From 420141758e4ce2e27df8269ab8e8018d187f9e82 Mon Sep 17 00:00:00 2001
From: Kaleb Phipps <kaleb.phipps@kit.edu>
Date: Mon, 3 Feb 2025 13:35:30 +0100
Subject: [PATCH] add results for sheet 5
---
5_dpnn/py/alex.py | 82 +
5_dpnn/py/alex_parallel.py | 103 +
5_dpnn/py/model.py | 97 +
5_dpnn/py/utils_data.py | 282 ++
5_dpnn/py/utils_eval.py | 167 +
5_dpnn/py/utils_train.py | 246 +
5_dpnn/results/gpu_16/loss_16_gpu.pt | Bin 0 -> 9840 bytes
5_dpnn/results/gpu_16/results_gpu_16.pdf | Bin 0 -> 64174 bytes
5_dpnn/results/gpu_16/slurm-25143894.out | 1270 ++++++
5_dpnn/results/gpu_16/train_acc_16_gpu.pt | Bin 0 -> 1796 bytes
5_dpnn/results/gpu_16/valid_acc_16_gpu.pt | Bin 0 -> 1796 bytes
5_dpnn/results/gpu_4/loss_4_gpu.pt | Bin 0 -> 39596 bytes
5_dpnn/results/gpu_4/results_gpu_4.pdf | Bin 0 -> 88265 bytes
5_dpnn/results/gpu_4/slurm-25137704.out | 4533 +++++++++++++++++++
5_dpnn/results/gpu_4/train_acc_4_gpu.pt | Bin 0 -> 1792 bytes
5_dpnn/results/gpu_4/valid_acc_4_gpu.pt | Bin 0 -> 1792 bytes
5_dpnn/results/serial/loss.pt | Bin 0 -> 158356 bytes
5_dpnn/results/serial/results_serial.pdf | Bin 0 -> 146143 bytes
5_dpnn/results/serial/slurm-25141451.out | 427 ++
5_dpnn/results/serial/solution_serial.ipynb | 1522 +++++++
5_dpnn/results/serial/train_acc.pt | Bin 0 -> 26362 bytes
5_dpnn/results/serial/valid_acc.pt | Bin 0 -> 26362 bytes
22 files changed, 8729 insertions(+)
create mode 100644 5_dpnn/py/alex.py
create mode 100644 5_dpnn/py/alex_parallel.py
create mode 100644 5_dpnn/py/model.py
create mode 100644 5_dpnn/py/utils_data.py
create mode 100644 5_dpnn/py/utils_eval.py
create mode 100644 5_dpnn/py/utils_train.py
create mode 100644 5_dpnn/results/gpu_16/loss_16_gpu.pt
create mode 100644 5_dpnn/results/gpu_16/results_gpu_16.pdf
create mode 100644 5_dpnn/results/gpu_16/slurm-25143894.out
create mode 100644 5_dpnn/results/gpu_16/train_acc_16_gpu.pt
create mode 100644 5_dpnn/results/gpu_16/valid_acc_16_gpu.pt
create mode 100644 5_dpnn/results/gpu_4/loss_4_gpu.pt
create mode 100644 5_dpnn/results/gpu_4/results_gpu_4.pdf
create mode 100644 5_dpnn/results/gpu_4/slurm-25137704.out
create mode 100644 5_dpnn/results/gpu_4/train_acc_4_gpu.pt
create mode 100644 5_dpnn/results/gpu_4/valid_acc_4_gpu.pt
create mode 100644 5_dpnn/results/serial/loss.pt
create mode 100644 5_dpnn/results/serial/results_serial.pdf
create mode 100644 5_dpnn/results/serial/slurm-25141451.out
create mode 100644 5_dpnn/results/serial/solution_serial.ipynb
create mode 100644 5_dpnn/results/serial/train_acc.pt
create mode 100644 5_dpnn/results/serial/valid_acc.pt
diff --git a/5_dpnn/py/alex.py b/5_dpnn/py/alex.py
new file mode 100644
index 0000000..7fb1a84
--- /dev/null
+++ b/5_dpnn/py/alex.py
@@ -0,0 +1,82 @@
+"""Serial training of AlexNet on CIFAR-10."""
+import torch
+
+from utils_data import get_dataloaders_cifar10, get_transforms_cifar10
+from utils_train import train_model
+from model import AlexNet
+
+if __name__ == "__main__":
+ # Set random seed.
+ seed = 123
+ torch.manual_seed(seed)
+ torch.cuda.manual_seed_all(seed)
+ b = 256 # Set batch size.
+ e = 100 # Set number of epochs to be trained.
+ data_root = "/pfs/work7/workspace/scratch/ku4408-VL-ScalableAI/data/cifar"
+
+ device = torch.device(
+ "cuda:0" if torch.cuda.is_available() else "cpu"
+ ) # Check device.
+ print(f"Using {device} device.")
+
+ # DATASET
+ # Using transforms on your data allows you to take it from its source state
+ # and transform it into data that’s ready for training.
+
+ # Get transforms applied to CIFAR-10 data for training and inference.
+ train_transforms, test_transforms = get_transforms_cifar10()
+
+ # Get PyTorch dataloaders for training, testing, and validation dataset.
+ train_loader, valid_loader, test_loader = get_dataloaders_cifar10(
+ batch_size=b, # The batch size.
+ data_root=data_root, # The path to the data dir.
+ validation_fraction=0.1, # The validation fraction.
+ train_transforms=train_transforms, # The transforms applied to the data at training time.
+ test_transforms=test_transforms, # The transforms applied to the data at inference time.
+ )
+
+ # Check loaded dataset.
+ for images, labels in train_loader:
+ print("Image batch dimensions:", images.shape)
+ print("Image label dimensions:", labels.shape)
+ print("Class labels of 10 examples:", labels[:10])
+ break
+
+ # MODEL
+ # Define neural network by subclassing PyTorch's nn.Module.
+
+ model = AlexNet(num_classes=10).to(
+ device
+ ) # Build instance of AlexNet with 10 classes and move it to device.
+
+ # torch.optim package implements various optimization algorithms.
+ # To use torch.optim you have to construct an optimizer object,
+ # that will hold the current state and will update the parameters
+ # based on computed gradients.
+ # Stochastic gradient descent (optionally with momentum):
+ optimizer = torch.optim.SGD(model.parameters(), momentum=0.9, lr=0.1)
+
+ # torch.optim.lr_scheduler provides several learning-rate adjustment methods based on number of epochs.
+ # torch.optim.lr_scheduler.ReduceLROnPlateau: dynamic learning rate reducing based on some validation measurements.
+ # Reduce learning rate when a metric has stopped improving:
+ scheduler = torch.optim.lr_scheduler.ReduceLROnPlateau(
+ optimizer, factor=0.1, mode="max",
+ )
+
+ # Train model.
+ loss_history, train_acc_history, valid_acc_history = train_model(
+ model=model, # The model to train.
+ num_epochs=e, # The number of epochs to train.
+ train_loader=train_loader, # The training dataloader.
+ valid_loader=valid_loader, # The validation dataloader.
+ test_loader=test_loader, # The test dataloader.
+ optimizer=optimizer, # The optimizer.
+ device=device, # The device to train on.
+ scheduler=scheduler, # The learning rate scheduler.
+ logging_interval=100, # The logging interval.
+ )
+
+ # Save loss and accuracy metrics to disk.
+ torch.save(loss_history, "loss.pt")
+ torch.save(train_acc_history, "train_acc.pt")
+ torch.save(valid_acc_history, "valid_acc.pt")
diff --git a/5_dpnn/py/alex_parallel.py b/5_dpnn/py/alex_parallel.py
new file mode 100644
index 0000000..e26b799
--- /dev/null
+++ b/5_dpnn/py/alex_parallel.py
@@ -0,0 +1,103 @@
+"""Distributed data-parallel training of AlexNet on CIFAR-10."""
+import os
+
+import torch
+import torch.distributed as dist
+from torch.nn.parallel import DistributedDataParallel as Ddp
+import torchvision
+
+from utils_data import get_dataloaders_cifar10_ddp, get_transforms_cifar10
+from utils_train import train_model_ddp
+from utils_eval import compute_accuracy_ddp
+from model import AlexNet
+
+
+def main():
+ """Distributed data-parallel training of AlexNet on CIFAR-10."""
+ world_size = int(os.getenv("SLURM_NPROCS")) # Get overall number of processes.
+ rank = int(os.getenv("SLURM_PROCID")) # Get individual process ID.
+ data_root = "/pfs/work7/workspace/scratch/ku4408-VL-ScalableAI/data"
+ print(
+ f"Rank, world size, device count: {rank}, {world_size}, {torch.cuda.device_count()}"
+ )
+
+ if rank == 0:
+ if dist.is_available():
+ print(
+ "Distributed package available...[OK]"
+ ) # Check if distributed package available.
+ if dist.is_nccl_available():
+ print("NCCL backend available...[OK]") # Check if NCCL backend available.
+
+ # On each host with N GPUs, spawn up N processes, while ensuring that
+ # each process individually works on a single GPU from 0 to N-1.
+
+ address = os.getenv("SLURM_LAUNCH_NODE_IPADDR")
+ port = "29500"
+ os.environ["MASTER_ADDR"] = address
+ os.environ["MASTER_PORT"] = port
+
+ # Initialize DDP.
+ dist.init_process_group(backend="nccl", rank=rank, world_size=world_size)
+ torch.cuda.set_device(rank % torch.cuda.device_count())
+ if dist.is_initialized():
+ print("Process group initialized successfully...[OK]") # Check initialization.
+
+ # Check used backend.
+ print(dist.get_backend(), "backend used...[OK]")
+
+ b = 256 # Set batch size.
+ e = 100 # Set number of epochs to be trained.
+
+ # DATASET
+ # Using transforms on your data allows you to take it from its source state
+ # and transform it into data that’s ready for training.
+
+ # Get transforms applied to CIFAR-10 data for training and inference.
+ train_transforms, test_transforms = get_transforms_cifar10()
+
+ # Get distributed dataloaders for training and validation data on all ranks.
+ train_loader, valid_loader = get_dataloaders_cifar10_ddp(
+ batch_size=b,
+ data_root=data_root,
+ train_transforms=train_transforms,
+ test_transforms=test_transforms,
+ )
+
+ model = AlexNet(
+ num_classes=10
+ ).cuda() # Create model and move it to GPU with id rank.
+ ddp_model = Ddp(model) # Wrap model with DDP.
+ optimizer = torch.optim.SGD(
+ ddp_model.parameters(), momentum=0.9, lr=0.1
+ ) # Set up optimizer and pass DDP parameters.
+
+ # Train model.
+ train_model_ddp(
+ model=ddp_model,
+ num_epochs=e,
+ train_loader=train_loader,
+ valid_loader=valid_loader,
+ optimizer=optimizer,
+ )
+
+ # Test final model on root.
+ if dist.get_rank() == 0:
+ test_dataset = torchvision.datasets.CIFAR10(
+ root=data_root,
+ train=False,
+ transform=test_transforms,
+ ) # Load test dataset.
+ test_loader = torch.utils.data.DataLoader(
+ dataset=test_dataset, batch_size=b, shuffle=False
+ ) # Set up test dataloader.
+ test_acc = compute_accuracy_ddp(
+ ddp_model, test_loader
+ ) # Compute accuracy on test data.
+ print(f"Test accuracy {test_acc :.2f}%")
+
+ dist.destroy_process_group() # Destroy process group.
+
+
+if __name__ == "__main__":
+ main()
diff --git a/5_dpnn/py/model.py b/5_dpnn/py/model.py
new file mode 100644
index 0000000..ffb02d1
--- /dev/null
+++ b/5_dpnn/py/model.py
@@ -0,0 +1,97 @@
+"""AlexNet model class."""
+import torch
+
+
+# MODEL
+# Define neural network by subclassing PyTorch's nn.Module.
+class AlexNet(torch.nn.Module):
+ """
+ AlexNet model for image classification.
+
+ Attributes
+ ----------
+ features : torch.nn.container.Sequential
+ The convolutional feature-extractor part of AlexNet.
+ avgpool : AdaptiveAvgPool2d
+ An adaptive pooling layer.
+ classifier : torch.nn.container.Sequential
+ The fully connected linear part of AlexNet.
+
+ Methods
+ -------
+ forward()
+ The forward pass.
+ """
+
+ def __init__(self, num_classes: int = 1000, dropout: float = 0.5) -> None:
+ """
+ Initialize AlexNet architecture.
+
+ Parameters
+ ----------
+ num_classes : int
+ The number of classes in the underlying classification problem.
+ dropout : float
+ The dropout probability.
+ """
+ super().__init__()
+ self.features = torch.nn.Sequential(
+ # AlexNet consists of 8 layers:
+ # 5 convolutional layers, some followed by max-pooling (see figure),
+ # and 3 fully connected layers.
+ # IMPLEMENT FEATURE-EXTRACTOR PART OF ALEXNET HERE!
+ # 1st convolutional layer (+ max-pooling)
+ torch.nn.Conv2d(3, 64, kernel_size=11, stride=4, padding=2),
+ torch.nn.ReLU(inplace=True),
+ torch.nn.MaxPool2d(kernel_size=3, stride=2),
+ # 2nd convolutional layer (+ max-pooling)
+ torch.nn.Conv2d(64, 192, kernel_size=5, padding=2),
+ torch.nn.ReLU(inplace=True),
+ torch.nn.MaxPool2d(kernel_size=3, stride=2),
+ # 3rd + 4th convolutional layer
+ torch.nn.Conv2d(192, 384, kernel_size=3, padding=1),
+ torch.nn.ReLU(inplace=True),
+ torch.nn.Conv2d(384, 256, kernel_size=3, padding=1),
+ torch.nn.ReLU(inplace=True),
+ # 5th convolutional layer
+ torch.nn.Conv2d(256, 256, kernel_size=3, padding=1),
+ torch.nn.ReLU(inplace=True),
+ torch.nn.MaxPool2d(kernel_size=3, stride=2),
+ )
+ # Average pooling to downscale possibly larger input images.
+ self.avgpool = torch.nn.AdaptiveAvgPool2d((6, 6))
+ self.classifier = torch.nn.Sequential(
+ # IMPLEMENT FULLY CONNECTED MULTI-LAYER PERCEPTRON PART HERE!
+ # 6th, 7th + 8th fully connected layer
+ torch.nn.Dropout(p=dropout),
+ torch.nn.Linear(256 * 6 * 6, 4096),
+ torch.nn.ReLU(inplace=True),
+ torch.nn.Dropout(p=dropout),
+ torch.nn.Linear(4096, 4096),
+ torch.nn.ReLU(inplace=True),
+ torch.nn.Linear(4096, num_classes),
+ ###################################
+ )
+
+ # Every nn.Module subclass implements the operations on the input data in the forward method.
+ # Forward pass: Apply AlexNet model to input x.
+ def forward(self, x: torch.Tensor) -> torch.Tensor:
+ """
+ The forward pass.
+
+ Parameters
+ ----------
+ x : torch.Tensor
+ The input data.
+
+ Returns
+ -------
+ torch.Tensor
+ The output.
+ """
+ # IMPLEMENT OPERATIONS ON INPUT DATA x HERE!
+ x = self.features(x)
+ x = self.avgpool(x)
+ x = torch.flatten(x, 1)
+ x = self.classifier(x)
+ return x
diff --git a/5_dpnn/py/utils_data.py b/5_dpnn/py/utils_data.py
new file mode 100644
index 0000000..0611599
--- /dev/null
+++ b/5_dpnn/py/utils_data.py
@@ -0,0 +1,282 @@
+"""Helper functions for loading CIFAR-10 data."""
+from typing import Any, Callable, Tuple
+
+import numpy as np
+import torch
+import torch.distributed as dist
+import torchvision
+
+
+def get_transforms_cifar10() -> (
+ Tuple[torchvision.transforms.Compose, torchvision.transforms.Compose]
+):
+ """
+ Get transforms applied to CIFAR-10 data for AlexNet training and inference.
+
+ Returns
+ -------
+ torchvision.transforms.Compose
+ The transforms applied to CIFAR-10 for training AlexNet.
+ torchvision.transforms.Compose
+ The transforms applied to CIFAR-10 to run inference with AlexNet.
+ """
+ # Transforms applied to training data (randomness to make network more robust against overfitting)
+ train_transforms = (
+ torchvision.transforms.Compose( # Compose several transforms together.
+ [
+ torchvision.transforms.Resize(
+ (70, 70)
+ ), # Upsample CIFAR-10 images to make them work with AlexNet.
+ torchvision.transforms.RandomCrop(
+ (64, 64)
+ ), # Randomly crop image to make NN more robust against overfitting.
+ torchvision.transforms.ToTensor(), # Convert image into torch tensor.
+ torchvision.transforms.Normalize(
+ (0.5, 0.5, 0.5), (0.5, 0.5, 0.5)
+ ), # Normalize to [-1,1] via (image-mean)/std.
+ ]
+ )
+ )
+
+ test_transforms = torchvision.transforms.Compose(
+ [
+ torchvision.transforms.Resize((70, 70)),
+ torchvision.transforms.CenterCrop((64, 64)),
+ torchvision.transforms.ToTensor(),
+ torchvision.transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5)),
+ ]
+ )
+ return train_transforms, test_transforms
+
+
+def make_train_validation_split(
+ train_dataset: torchvision.datasets.CIFAR10,
+ seed: int = 123,
+ validation_fraction: float = 0.1,
+) -> Tuple[np.ndarray, np.ndarray]:
+ """
+ Split original CIFAR-10 training data into train and validation sets.
+
+ Parameters
+ ----------
+ train_dataset : torchvision.datasets.CIFAR10
+ The original CIFAR-10 training dataset.
+ seed : int
+ The seed used to split the data.
+ validation_fraction : float
+ The fraction of samples used for validation.
+
+ Returns
+ -------
+ numpy.ndarray
+ The sample indices for the training dataset.
+ numpy.ndarray
+ The sample indices for the validation dataset.
+ """
+ num_samples = len(
+ train_dataset
+ ) # Get overall number of samples in original training data.
+ rng = np.random.default_rng(
+ seed=seed
+ ) # Set same seed over all ranks for consistent train-test split.
+ idx = np.arange(0, num_samples) # Construct array of all indices.
+ rng.shuffle(idx) # Shuffle them.
+ num_validate = int(
+ validation_fraction * num_samples
+ ) # Determine number of validation samples from validation split.
+ return (
+ idx[num_validate:],
+ idx[0:num_validate],
+ ) # Extract and return train and validation indices.
+
+
+def get_dataloaders_cifar10(
+ batch_size: int,
+ data_root: str = "data",
+ validation_fraction: float = 0.1,
+ train_transforms: Callable[[Any], Any] = None,
+ test_transforms: Callable[[Any], Any] = None,
+ seed: int = 123,
+) -> Tuple[
+ torch.utils.data.DataLoader,
+ torch.utils.data.DataLoader,
+ torch.utils.data.DataLoader,
+]:
+ """
+ Get dataloaders for training, validation, and testing on the CIFAR-10 dataset.
+
+ Parameters
+ ----------
+ batch_size : int
+ The mini-batch size.
+ data_root : str
+ The data folder.
+ validation_fraction : float
+ The fraction of the original training data used for validating.
+ train_transforms : Callable[[Any], Any]
+ The transform applied to the training data.
+ test_transforms : Callable[[Any], Any]
+ The transform applied to the testing data (inference).
+ seed : int
+ The seed for the validation-train split.
+
+ Returns
+ -------
+ torch.utils.data.DataLoader
+ The train dataloader.
+ torch.utils.data.DataLoader
+ The validation dataloader.
+ torch.utils.data.DataLoader
+ The test dataloader.
+ """
+ if train_transforms is None:
+ train_transforms = torchvision.transforms.ToTensor()
+
+ if test_transforms is None:
+ test_transforms = torchvision.transforms.ToTensor()
+
+ train_dataset = torchvision.datasets.CIFAR10(
+ root=data_root, train=True, transform=train_transforms, download=True
+ )
+
+ valid_dataset = torchvision.datasets.CIFAR10(
+ root=data_root, train=True, transform=test_transforms
+ )
+
+ test_dataset = torchvision.datasets.CIFAR10(
+ root=data_root, train=False, transform=test_transforms
+ )
+
+ # Perform index-based train-validation split of original training data.
+ train_indices, valid_indices = make_train_validation_split(
+ train_dataset, seed, validation_fraction
+ ) # Get train and validation indices.
+
+ train_sampler = torch.utils.data.SubsetRandomSampler(train_indices)
+ valid_sampler = torch.utils.data.SubsetRandomSampler(valid_indices)
+
+ valid_loader = torch.utils.data.DataLoader(
+ dataset=valid_dataset,
+ batch_size=batch_size,
+ sampler=valid_sampler,
+ )
+
+ train_loader = torch.utils.data.DataLoader(
+ dataset=train_dataset,
+ batch_size=batch_size,
+ drop_last=True,
+ sampler=train_sampler,
+ )
+
+ test_loader = torch.utils.data.DataLoader(
+ dataset=test_dataset,
+ batch_size=batch_size,
+ shuffle=False,
+ )
+
+ return train_loader, valid_loader, test_loader
+
+
+def get_dataloaders_cifar10_ddp(
+ batch_size: int,
+ data_root: str = "data",
+ validation_fraction: float = 0.1,
+ train_transforms: Callable[[Any], Any] = None,
+ test_transforms: Callable[[Any], Any] = None,
+ seed=123,
+) -> Tuple[torch.utils.data.DataLoader, torch.utils.data.DataLoader]:
+ """
+ Get distributed CIFAR-10 dataloaders for training and validation in a DDP setting.
+
+ Parameters
+ ----------
+ batch_size : int
+ The batch size.
+ data_root : str
+ The path to the data directory.
+ validation_fraction : float
+ The fraction of training samples used for validation.
+ train_transforms : Callable[[Any], Any]
+ The transform applied to the training data.
+ test_transforms : Callable[[Any], Any]
+ The transform applied to the testing data (inference).
+ seed : int
+ Seed for train-validation split.
+
+ Returns
+ -------
+ torch.utils.data.DataLoader
+ The training dataloader.
+ torch.utils.data.DataLoader
+ The validation dataloader.
+ """
+ if train_transforms is None:
+ train_transforms = torchvision.transforms.ToTensor()
+ if test_transforms is None:
+ test_transforms = torchvision.transforms.ToTensor()
+
+ if (
+ dist.get_rank() == 0
+ ): # Only root shall download dataset if data is not already there.
+ train_dataset = torchvision.datasets.CIFAR10(
+ root=data_root, train=True, transform=train_transforms, download=True
+ )
+
+ dist.barrier(device_ids=[torch.cuda.current_device()]) # Barrier
+
+ if (
+ dist.get_rank() != 0
+ ): # Other ranks must not download dataset at the same time in parallel.
+ train_dataset = torchvision.datasets.CIFAR10(
+ root=data_root, train=True, transform=train_transforms
+ )
+
+ valid_dataset = torchvision.datasets.CIFAR10(
+ root=data_root, train=True, transform=test_transforms
+ )
+
+ # Perform index-based train-validation split of original training data.
+ train_indices, valid_indices = make_train_validation_split(
+ train_dataset, seed, validation_fraction
+ ) # Get train and validation indices.
+
+ # Split into training and validation dataset according to specified validation fraction.
+ train_dataset = torch.utils.data.Subset(train_dataset, train_indices)
+ valid_dataset = torch.utils.data.Subset(valid_dataset, valid_indices)
+
+ # Sampler that restricts data loading to a subset of the dataset.
+ # Especially useful in conjunction with torch.nn.parallel.DistributedDataParallel.
+ # Each process can pass a DistributedSampler instance as a DataLoader sampler,
+ # and load a subset of the original dataset that is exclusive to it.
+
+ train_sampler = torch.utils.data.distributed.DistributedSampler(
+ train_dataset,
+ num_replicas=torch.distributed.get_world_size(),
+ rank=torch.distributed.get_rank(),
+ shuffle=True,
+ drop_last=True,
+ )
+
+ valid_sampler = torch.utils.data.distributed.DistributedSampler(
+ valid_dataset,
+ num_replicas=torch.distributed.get_world_size(),
+ rank=torch.distributed.get_rank(),
+ shuffle=True,
+ drop_last=True,
+ )
+
+ train_loader = torch.utils.data.DataLoader(
+ dataset=train_dataset,
+ batch_size=batch_size,
+ drop_last=True,
+ sampler=train_sampler,
+ )
+
+ valid_loader = torch.utils.data.DataLoader(
+ dataset=valid_dataset,
+ batch_size=batch_size,
+ drop_last=True,
+ sampler=valid_sampler,
+ )
+
+ return train_loader, valid_loader
diff --git a/5_dpnn/py/utils_eval.py b/5_dpnn/py/utils_eval.py
new file mode 100644
index 0000000..0f5ccb5
--- /dev/null
+++ b/5_dpnn/py/utils_eval.py
@@ -0,0 +1,167 @@
+"""Evaluation helper functions."""
+import os
+import pathlib
+import random
+from typing import Tuple, Union
+
+import matplotlib.pyplot as plt
+import numpy as np
+import torch
+
+
+def set_all_seeds(seed: int) -> None:
+ """
+ Set all seeds to the same fixed value.
+
+ Parameters
+ ----------
+ seed : int
+ The seeds' value.
+ """
+ os.environ["PL_GLOBAL_SEED"] = str(seed)
+ random.seed(seed)
+ np.random.seed(seed)
+ torch.manual_seed(seed)
+ torch.cuda.manual_seed_all(seed)
+
+
+def compute_accuracy(
+ model: torch.nn.Module,
+ data_loader: torch.utils.data.DataLoader,
+ device: torch.device,
+) -> float:
+ """
+ Compute the accuracy of the model's predictions on given labeled data.
+
+ Parameters
+ ----------
+ model : torch.nn.Module
+ The model.
+ data_loader : torch.utils.data.DataLoader
+ The dataloader.
+ device : torch.device
+ The device to use.
+
+ Returns
+ -------
+ float
+ The model's accuracy on the given dataset in percent.
+ """
+ with torch.no_grad(): # Disable gradient calculation to reduce memory consumption.
+ correct_pred, num_examples = (
+ 0,
+ 0,
+ ) # Initialize number of correctly predicted and overall samples, respectively.
+
+ for i, (features, targets) in enumerate(data_loader):
+ features = features.to(device)
+ targets = targets.float().to(device)
+
+ logits = model(features)
+ _, predicted_labels = torch.max(logits, 1) # Get class with highest score.
+
+ num_examples += targets.size(0)
+ correct_pred += (predicted_labels == targets).sum()
+ return correct_pred.float() / num_examples * 100
+
+
+def compute_accuracy_ddp(
+ model: torch.nn.Module, data_loader: torch.utils.data.DataLoader
+) -> float:
+ """
+ Compute the accuracy of the model's predictions on given labeled data.
+
+ Parameters
+ ----------
+ model : torch.nn.Module
+ The model.
+ data_loader : torch.utils.data.DataLoader
+ The dataloader.
+
+ Returns
+ -------
+ float
+ The model's accuracy on the given dataset in percent.
+ """
+ correct_pred, num_examples = get_right_ddp(model, data_loader)
+ return correct_pred.item() / num_examples.item() * 100
+
+
+def get_right_ddp(
+ model: torch.nn.Module, data_loader: torch.utils.data.DataLoader
+) -> Tuple[torch.Tensor, torch.Tensor]:
+ """
+ Compute the number of correctly predicted samples and the overall number of samples in a given dataset.
+
+ This function is needed to compute the accuracy over multiple processors in a distributed data-parallel setting.
+
+ Parameters
+ ----------
+ model : torch.nn.Module
+ The model.
+ data_loader : torch.utils.data.DataLoader
+ The dataloader.
+
+ Returns
+ -------
+ torch.Tensor
+ The number of correctly predicted samples.
+ torch.Tensor
+ The overall number of samples in the dataset.
+ """
+ with torch.no_grad():
+ correct_pred, num_examples = 0, 0
+
+ for i, (features, targets) in enumerate(data_loader):
+ features = features.cuda()
+ targets = targets.float().cuda()
+ logits = model(features)
+ _, predicted_labels = torch.max(logits, 1) # Get class with highest score.
+
+ num_examples += targets.size(0)
+ correct_pred += (predicted_labels == targets).sum()
+
+ correct_pred = torch.Tensor([correct_pred]).cuda()
+ num_examples = torch.Tensor([num_examples]).cuda()
+ return correct_pred, num_examples
+
+
+def plot_results(res_path: Union[pathlib.Path, str]) -> None:
+ """
+ Plot training loss and training and validation accuracy.
+
+ Parameters
+ ----------
+ res_path : pathlib.Path
+ The path to the results pickle files.
+ """
+ label_size = 16
+ res_path = pathlib.Path(res_path)
+ train_loss = np.array(torch.load(res_path / pathlib.Path("loss.pt")))
+ train_acc = np.array(torch.load(res_path / pathlib.Path("train_acc.pt")))
+ valid_add = np.array(torch.load(res_path / pathlib.Path("valid_acc.pt")))
+ n_epochs = len(train_acc)
+ f, (ax1, ax2) = plt.subplots(1, 2, figsize=(10, 4.5))
+ epochs_loss = np.linspace(1, n_epochs, train_loss.shape[0])
+ epochs_acc = np.linspace(1, n_epochs, n_epochs)
+ ax1.plot(epochs_loss, train_loss, lw=0.1, color="k")
+ ax1.grid()
+ ax1.set_xlabel("Epoch", weight="bold", fontsize=label_size)
+ ax1.set_ylabel("Train loss", weight="bold", fontsize=label_size)
+ ax2.plot(epochs_acc, train_acc, label="Training", lw=2, color="k")
+ ax2.plot(
+ epochs_acc, valid_add, label="Validation", lw=2, color=(0.1, 0.6294, 0.5588)
+ )
+ ax2.set_xlabel("Epoch", weight="bold", fontsize=label_size)
+ ax2.set_ylabel("Accuracy / %", weight="bold", fontsize=label_size)
+ ax2.legend(fontsize=label_size)
+ ax2.grid()
+ plt.tight_layout()
+ plt.savefig(res_path / pathlib.Path("results.pdf"))
+ plt.show()
+
+
+if __name__ == "__main__":
+ plot_results(res_path="../res/gpu_1")
+ plot_results(res_path="../res/gpu_4")
+ plot_results(res_path="../res/gpu_16")
diff --git a/5_dpnn/py/utils_train.py b/5_dpnn/py/utils_train.py
new file mode 100644
index 0000000..7747bd6
--- /dev/null
+++ b/5_dpnn/py/utils_train.py
@@ -0,0 +1,246 @@
+"""Helper functions for training (DDP) AlexNet on CIFAR-10."""
+import time
+from typing import List, Tuple, Union
+
+import torch
+from torch.optim.lr_scheduler import ReduceLROnPlateau
+
+from utils_eval import compute_accuracy, get_right_ddp
+
+
+def train_model(
+ model: torch.nn.Module,
+ num_epochs: int,
+ train_loader: torch.utils.data.DataLoader,
+ valid_loader: torch.utils.data.DataLoader,
+ test_loader: torch.utils.data.DataLoader,
+ optimizer: torch.optim.Optimizer,
+ device: torch.device,
+ logging_interval: int = 50,
+ scheduler: Union[torch.optim.lr_scheduler._LRScheduler, ReduceLROnPlateau] = None,
+) -> Tuple[List[float], List[float], List[float]]:
+ """
+ Train your model.
+
+ Parameters
+ ----------
+ model : torch.nn.Module
+ The model to train.
+ num_epochs : int
+ The number of epochs to train
+ train_loader : torch.utils.data.DataLoader
+ The training dataloader.
+ valid_loader : torch.utils.data.DataLoader
+ The validation dataloader.,
+ test_loader : torch.utils.data.DataLoader
+ The testing dataloader.
+ optimizer : torch.optim.Optimizer
+ The optimizer to use.
+ device : torch.device
+ The device to train on.
+ logging_interval : int
+ The logging interval.
+ scheduler : torch.optim.lr_scheduler._LRScheduler
+ An optional learning rate scheduler.
+
+ Returns
+ -------
+ List[float]
+ The loss history.
+ List[float]
+ The training accuracy history.
+ List[float]
+ The validation accuracy history.
+ """
+ start = time.perf_counter() # Measure training time.
+
+ # Initialize history lists for loss, training accuracy, and validation accuracy.
+ loss_history, train_acc_history, valid_acc_history = [], [], []
+
+ for epoch in range(num_epochs): # Loop over epochs.
+ model.train() # Set model to training mode.
+ # Thus, layers like dropout which behave differently on train and test procedures
+ # know what is going on and can behave accordingly. model.train() sets the mode to
+ # train. One might expect this to train model but it does not do that.
+ # Call either model.eval() or model.train(mode=False) to tell that you are testing.
+
+ for batch_idx, (features, targets) in enumerate(
+ train_loader
+ ): # Loop over mini batches.
+ features = features.to(device) # Move features to used device.
+ targets = targets.to(device) # Move targets to used device.
+
+ # Forward and backward pass.
+ logits = model(features) # Calculate logits as the model's output.
+ loss = torch.nn.functional.cross_entropy(
+ logits, targets
+ ) # Calculate cross-entropy loss.
+ optimizer.zero_grad() # Zero out gradients.
+ # Gradients are accumulated and not overwritten whenever .backward() is called.
+ loss.backward() # Calculate gradients of loss w.r.t. model parameters in backward pass.
+ optimizer.step() # Perform single optimization step to update model parameters.
+
+ # Logging.
+ loss_history.append(loss.item())
+ if not batch_idx % logging_interval:
+ print(
+ f"Epoch: {epoch+1:03d}/{num_epochs:03d} "
+ f"| Batch {batch_idx:04d}/{len(train_loader):04d} "
+ f"| Loss: {loss:.4f}"
+ )
+
+ model.eval() # Set model to evaluation mode.
+
+ with torch.no_grad(): # Disable gradient calculation to reduce memory consumption.
+ train_acc = compute_accuracy(
+ model, train_loader, device=device
+ ) # Compute accuracy on training data.
+ valid_acc = compute_accuracy(
+ model, valid_loader, device=device
+ ) # Compute accuracy on validation data.
+ print(
+ f"Epoch: {epoch+1:03d}/{num_epochs:03d} "
+ f"| Train: {train_acc :.2f}% "
+ f"| Validation: {valid_acc :.2f}%"
+ )
+ train_acc_history.append(train_acc)
+ valid_acc_history.append(valid_acc)
+
+ elapsed = (time.perf_counter() - start) / 60 # Measure training time per epoch.
+ print(f"Time elapsed: {elapsed:.2f} min")
+
+ if scheduler is not None:
+ original_lr = scheduler.get_last_lr()[0]
+ scheduler.step(valid_acc_history[-1])
+ new_lr = scheduler.get_last_lr()[0]
+ if original_lr != new_lr:
+ print(f"Epoch: {epoch+1:03d}/{num_epochs:03d}: LR updated to {new_lr}")
+
+ elapsed = (time.perf_counter() - start) / 60 # Measure total training time.
+ print(f"Total Training Time: {elapsed:.2f} min")
+
+ test_acc = compute_accuracy(
+ model, test_loader, device=device
+ ) # Compute accuracy on test data.
+ print(f"Test accuracy {test_acc :.2f}%")
+
+ return loss_history, train_acc_history, valid_acc_history
+
+
+def train_model_ddp(
+ model: torch.nn.Module,
+ num_epochs: int,
+ train_loader: torch.utils.data.DataLoader,
+ valid_loader: torch.utils.data.DataLoader,
+ optimizer: torch.optim.Optimizer,
+) -> Tuple[List[float], List[float], List[float]]:
+ """
+ Train the model in distributed data-parallel fashion.
+
+ Parameters
+ ----------
+ model : torch.nn.Module
+ The model to train.
+ num_epochs : int
+ The number of epochs to train.
+ train_loader : torch.utils.data.DataLoader
+ The training dataloader.
+ valid_loader : torch.utils.data.DataLoader
+ The validation dataloader.
+ optimizer : torch.optim.Optimizer
+ The optimizer to use.
+
+ Returns
+ -------
+ List[float]
+ The epoch-wise loss history.
+ List[float]
+ The epoch-wise training accuracy history.
+ List[float]
+ The epoch-wise validation accuracy history.
+ """
+ start = time.perf_counter() # Measure training time.
+ rank = torch.distributed.get_rank()
+ world_size = torch.distributed.get_world_size()
+
+ loss_history, train_acc_history, valid_acc_history = (
+ [],
+ [],
+ [],
+ ) # Initialize history lists.
+
+ for epoch in range(num_epochs): # Loop over epochs.
+ train_loader.sampler.set_epoch(epoch)
+ model.train() # Set model to training mode.
+
+ for batch_idx, (features, targets) in enumerate(
+ train_loader
+ ): # Loop over mini batches.
+ features = features.cuda()
+ targets = targets.cuda()
+
+ # Forward and backward pass.
+ logits = model(features)
+ loss = torch.nn.functional.cross_entropy(logits, targets)
+ optimizer.zero_grad()
+ loss.backward()
+ optimizer.step() # Perform single optimization step to update model parameters.
+
+ # Logging.
+ torch.distributed.all_reduce(
+ loss
+ ) # Allreduce rank-local mini-batch losses.
+ loss /= world_size # Average all-reduced rank-local mini-batch losses over all ranks.
+ loss_history.append(
+ loss.item()
+ ) # Append globally averaged loss of this epoch to history list.
+
+ if rank == 0:
+ print(
+ f"Epoch: {epoch+1:03d}/{num_epochs:03d} "
+ f"| Batch {batch_idx:04d}/{len(train_loader):04d} "
+ f"| Averaged Loss: {loss:.4f}"
+ )
+
+ model.eval() # Set model to evaluation mode.
+
+ with torch.no_grad(): # Disable gradient calculation.
+ # Get rank-local numbers of correctly classified and overall samples in training and validation set.
+ right_train, num_train = get_right_ddp(model, train_loader)
+ right_valid, num_valid = get_right_ddp(model, valid_loader)
+
+ torch.distributed.all_reduce(right_train)
+ torch.distributed.all_reduce(right_valid)
+ torch.distributed.all_reduce(num_train)
+ torch.distributed.all_reduce(num_valid)
+ train_acc = right_train.item() / num_train.item() * 100
+ valid_acc = right_valid.item() / num_valid.item() * 100
+ train_acc_history.append(train_acc)
+ valid_acc_history.append(valid_acc)
+
+ if rank == 0:
+ print(
+ f"Epoch: {epoch+1:03d}/{num_epochs:03d} "
+ f"| Train: {train_acc :.2f}% "
+ f"| Validation: {valid_acc :.2f}%"
+ )
+
+ elapsed = (time.perf_counter() - start) / 60 # Measure training time per epoch.
+ elapsed = torch.Tensor([elapsed]).cuda()
+ torch.distributed.all_reduce(elapsed)
+ elapsed /= world_size
+ if rank == 0:
+ print(f"Time elapsed: {elapsed.item()} min")
+
+ elapsed = (time.perf_counter() - start) / 60 # Measure total training time.
+ elapsed = torch.Tensor([elapsed]).cuda()
+ torch.distributed.all_reduce(elapsed)
+ elapsed /= world_size
+
+ if rank == 0:
+ print("Total Training Time:", elapsed.item(), "min")
+ torch.save(loss_history, f"loss_{world_size}_gpu.pt")
+ torch.save(train_acc_history, f"train_acc_{world_size}_gpu.pt")
+ torch.save(valid_acc_history, f"valid_acc_{world_size}_gpu.pt")
+
+ return loss_history, train_acc_history, valid_acc_history
diff --git a/5_dpnn/results/gpu_16/loss_16_gpu.pt b/5_dpnn/results/gpu_16/loss_16_gpu.pt
new file mode 100644
index 0000000000000000000000000000000000000000..fb7362109669f5c7f20755940618ce25af80629d
GIT binary patch
literal 9840
zcmb8#d0bXy-pBD90db4mQz!5!?xx}fqe)z%xsi&NI%?9v9rp#d9*e6&Ev?kFG#l->
zRoJ37qvqstA*<p#vI^)y5s*#xMYg%l?|a!~&&=!fJoK{PyU*_|_rgv7fuFCvgU!~c
zk?ntdwYE9imMmYr`mO#RZ!K7{rr+Fvpn$$B7B3m))5JF8w=dpc)=FEC(X7GZ*-TzS
zVo3w8Fj!71FPOmM*_uKSQ}|XW>^2o*bzspWg~cPeDmUe{F1XlHVbRs67|XF*wAfVG
zX()`5(>iOBGY0}M35EVo%Q00J{wO!r!3#r~!T=5|*ssch>#EGJR%QMmRp$NmY4Nbh
zFL|9gT?8=Oivs~qAN;J5I?4>8Rx?5jlxgEkU_t>0#%Gy;?{Yb<E00*A10EF|=;5i*
z?34`h7`y868jv=f17U-7z}T!xz$p!7M0Wvn{#l{%EspZsMFu=S3b3W7nN)0H8x4Zp
zb8pINJ+L<p3t-zffZfSeCd#<^2ElHe0|z#`8NjSD4Df9)r}g0VJOk+bG6Nc|GYPf}
z12o{^RtBsc$$|OR0tjxe@Sb|(Y;WcmSb21nQ{KYXc`ySyHWh&LLIsB|x>daEBd0Oo
zS5Ff#byX4ubCeHSD12(CR;~8Pjdl9LlLTh_qK7K&HN#plyCa#LU_WD>PH;&1#sDTy
z)d==q7RYJ6svCS1N++2pgAX!--JNX;XSZ{dZx8FhMOTI34>`)wS~;z|AAI^=gpAWr
zUMo`Yza)co%**x)KI-mvBMxz_I{!r{*s3PUDG#=l4lsbb6FG1#UX@!e3UzHYtnOrR
zg59TKa$2V!4OTd#KFhY4SsGTShclpLstK&<#({0?<+M)UdRpPlT?*rU1y-9JI55Rq
z09`jLj5x|rJk^KD)-Fbsv91K~`U_W0>lp9)W2@z~aMq6l9;Y-QaFHr64p;DYFtGYl
zOO(2@JVvGNwu=Ttcac+Gux{&K4g@ecty{Rca|_IFr;9+@l0jg08*B=nW5w(yO=Z?G
z%&z$la$0aK<^<d4y-{q7wlkCs>vX`ySx)Pv+mz1$Z&zW#yM7m*4zo><pu%jPyA;|b
zDSXmP9!m$$&@3>UeHKF*d5r@xPA0H7UExS8fim%Bg%^Suit8B#Z-!M}$sA6rx^!<j
ztrzNtmlb|(V4~!-5WvAtIq>~*g%jHa%GvI6TBnOQ73MfGlwEE*Fl>_mymbb=wO$ox
z7mhN@n*po4$!T41+>0FO`-=(8i&4eXQ$uO(r;2}=j<Vn!1MGKkU}dxkyj{nEF8&NS
z+}HrxhRbQ)^iBL9&AZO)9Zse8<Mu$^D{B>8<ojVAT$|>l0SWj~>0SGgdOEeS2RK%X
z*E8UQ^8$#8(16!MO<-As4)i?C0egRicIRY}_o{j9tOK8wbHING19CbWz!xh8(72hL
z)>&O^8KJh*Y>u*PhDoSSeP2%N7_X^<P}}{CDrqr5^}bh4TGeEp1L~Zw2r9L%@r+Q{
z@_%$fO-M0Hb@>k(mD)S+0yX>p#!)g;84#R_Qgc2QrM}&6g}vWkRhMx}vssU@&bt>%
z?I(UHwR?+DYL4wxc(0>rqo?Dyf)T3g<A9nbuAE9uA0dEgjs~H6|51gTM}V5M4dq7O
z^hx-)s@O%Sa-%V$Rdv@}9GEcIAUuh?Ab=+y%W0kN`p^XQof=>tC8u0f?Q70ZoD&5~
z`D&D^O+FgR;c0SO2QQh+36=M+GD6k-R&vT!Wt6*)@_a9b;{LWm=v*De;~GlE?#XJ^
zrm?_k@+)e!-Cm*Q6}2)J$V05VZ>-~l>J?#(P#fS3)Q3JtV78!8%~~<Lfa7w?v+5G=
za4NO~K>}zICZ}B0pMRGV>ZXln!08$}<*K$-GY!ap0jNp8!>H7p^fH0kO9FVetHO~o
zMx}Z}g`Dy_)dw4))Ofk5;^u5%wNSkUPev}~RG!2fVt}uUAXIJct5E%|hElvmmA`p%
zl$CciU}FP0<xQ)Gx-mlK&3u94dRLW_9wwE_(31j)P(MH_qjzvB6@_i&l-H^BxuP(0
ztwP8ehSh-y99Y@`rE24PlS)OYhe4=JI;vo&zC|mCy})Qykua14U+x#c$0OvFH!ZJe
zXaLoM0~_K@AnvLTjL;Zx^Ie5&ZX6|Gp$>ePZvusT1Tb`{4qSUEr**H-)camBp`A&k
z;$*G>{LeFBTdbVc!LwR%lt1M&lx@`l<sC<ZP&WND0VH43054xT<>_VPe^glgF+=g+
ztFU>#hT?$lRqxWUC{CsH?m_|VZfOEr<ghMSntW6MBj$6!%wWK4ughs2)4LKV$vMPO
zMtyGr`@iQv;V?O^gZ+~Q;Nxf#DmwLLgsQ#<@Z_6+05uaD>efpBy(Pr7;9WOUeW>dS
zgA7{P8u?*WzqpgIig#_}Je1muag0iJ*&#XQsyd-H2f8#dfhz}e!jq~?a$3jqw&R4V
zSoLjF6;i~gJaNevSZ#<zsp|DDr&77j)PWH*Rk`RSgS=p6#}Nv{ynza{k4CGC@A?a1
z?mKeIgDV^lFhcnUojA&r4g%%Wc{$}V<!dzq2<V_NqLDxe+NE&S1j?q@>sU4FY7i>Y
z0tKOb#V9$go2K?R31w|U8KJzUxgeBY&DT(-`^YH|E~{%Ifa<fHQ2s$@MksSqKh?|q
zm&;&1)CE5qgo^oL4CU)Sa$2kA9|%I}@huF+=xPwk7OPja?2~LxrPTL(0X#`Jfj;Ul
zLM0hKC?z?M4O*21sMonPs3Ak?e^db5N2~IFH&rgD$c=nFB~f<`;K(ROC_T236Uwek
z6(~)2qLg-gsLC2^TBjRH4CvBAA-I%NDL&w$0aMNjVAd^0D0*DYftdLy#ajv$X4|XP
zmeUG%ZUDuNx2dv5QwUErZPvpmTErAqeTY?ISR;njxsOyyv{Pm5LxmZCHf+}E#oMO|
zLg{)}2K1iF3FS%7s{A=vP^m2UlvAEnk-C8s%5BF5q4L9<C>38#l)*YCq$eYkpXsj=
z$_iXi%0j*XD#qBW($-gDh^uC^PA^TFgHm$+1gBEgG)|T60|b?l_dk`>dKO|J2gWpG
zKyep2tyMo~PAIzEO#s)|=)lEnRhGZbs1!F_ZV(Fdhs!BXFLLn~D5=*`3WLvTC~I;|
zVBKsEOnHw1KW{Vuk3n+E>lFUtB>?~32BF}_IyvR4U{xT><L<G7%Hswha$0A(I08j}
z>fI?S*rQg*BLy8(^h-V?6z0f5XXOplfxmvv2?ePaRCz;vs0&Vq2&{%Un?Td8a>|=N
z?w2VD1)Y)=YTuE;dZ^*Y8Q^(EBjkNKzyNkkROL{pppxJFx(N)3(g_6v|0GaO`pPLU
zm{*r%0<TQffjh$)A^&^>RYE=$R0_Ob0CGZ}a4NawA2C8+r&>9!SL_w_-CMZX3#FhU
z5_o*EyE=ydlS|1rcWY~);FoZ$3YKQ+w0b<#i4h8i9@GeVC&J{EXXV#-<$%+Eg-K%!
zDtTGERB80J!UcChtNf{l86hv=K9HNS$UsS3rvt|fLC7DezL9cYyN8n7P<^lFc%<lb
z%%iMCta4`9F)BHw9XJrE-uFirkEv4c0_3jVAn2Ivg9qi5Pv=o!tO>OC=Y;I|lLF<=
z9G#H0zkvyqsOOdSQ4dZf+y6NYNc~a}G9NVrvL+srLEa*3U#P<3g;-@fj%BpU9#_eM
z@G3#bU6!q(9KJ56JUusHE%4}aCqd;=_$>~Eu0Y9cd61!0%`pf$W9kGU+kKh|Xn*B|
zM_XEIC=HsRWZvG5lGXNYgI1ZQdhnT(zhNj79+-rOKWtECl=?Jf?N5*!>zRHSq%dKj
zNhRZ41(12KpFp|i1UwuYEQ37jVcT7dkn!G1laMw3>C1A*6jU+?IdiOjIwPlbuLso{
z@Wv#R%)CU7Qs%4!9o<lJ_PQ`Cxue#}DNir{ZapWI?45^F*6t*uQhZ9i;Kf7kqZIYq
zC^zz~qG88S3Wuv7Ade$`1g#2=&(jEnP46*6{(|X1fy3(>m7EvDbV6=bQ;ssSk3o1e
z;Aa86+?x@yiyv^5nR|ikuR540ovt!MR?6Syv|fz^Gf}epxCxYgBXq*UA0tgd)`ZXH
zlm|cD_e213%NQ_hkHYYe751IsSS|mHLCEOis4)1tpz^@WP9b=WNhPCifet($E`Wx1
z8sWjD-6jxMtcpV)P9=TUZs5TVPgQc&2jqT(07l2$f9Zw^Oy9@}>6^b3z|wudgJ#_s
zmHXqI<dn}e{YWQ`VAl3xKzc6&*qIEZM{j3T(hrZ4Q=VnU&lbR`?m8jes~ZC<C(3DE
zIk6ol+_%pKGAfq=561i|H}aTF|HcO4er{)!td|6%^3d-9kR7v<qXa(vfP82LdHTbo
zos96H$`K_a@M}ROV__TMLF#8%J(%24!)j$UBi#2IsL(ACNRO@&v`TjypaUh<2H}2X
z8xy5_x*(YA%jC3PjXpP3+3sUdNw=x*)q4XDV3poLJ$SRl42_Pt+v9*ixEKDsPDrae
zE>IHB$|<jWFV`C-ExnyaC2d<%Mo4)yL=aM&cHxAS)eb;f&SV+n>8Y<eGs4~M3gBMb
zIjZ#BtkEhhtT!jzYa0TjE^P^<Y_rpC)^+aWsTV5whI)6>g4Fj<${zKKrCil<)}1GN
z1f7-Q`;(m3!`QvdBqZIq$br<MKuUl$ty=`n(+No#aR#vIA_IcUIU#vprYc_rYAD;i
zbwa|C0gMpuen~^|PBRIKQyViv{C*cfNZ7GLBP32bjgol%JC2fH#(?{Y1|ebNaGemJ
zw??4&j$uH*9vUHW>LeXyM^8>j{<fJQq+GdU5Yi$`<&@7fX_&zXiNVhaLh_)0aG+-i
zkTQ0Mppy6(^%^Ec&qPUxYKUVJYbqHXb7w*VR!PA@CY6L%cjc7#ig&rm0b4hPQYVed
z?Jr&wg!sBsI&j~H5;vy>N<vY4lz9IO+-6<ycE{c*x4XP0gDkh!yo&PAp_74Io!e-%
ziZ9*F3CZVA$SDs_IoCxaB-W1x?m9PNRBnGgO-}2WwKsue=kc6M@-p>`C8zFWRMHMT
z1kz)21(my=!GJlmTBDK@7sv^ziEcpB>0UC(TO>GLV1(pI^;0WpUyM#AakYm=NLjI&
zp$wm5088RHAvquc<#uqCpc1#_3`%^njWWnv+}{6)5#rs|&%^kqe+9kWq9;}f;}3B<
zCaz0}oYq-We$fcG9<)Wd<!o=DWH155jnxP@owftFe&{2EJpIPu69(aCY%d+9;X#4&
z>H$u;Hfp|{^5E;)XBpw@Y4wx;>VkVHmw&RC8+pu?A@2*q&EU1b)x+wi@r{oAO<G<3
zOkfqw)JNv}KNmAvT`n8V30K;^BY^S~s?1P7h_1AFiP7rn-(C=e*om!CuCMRIsl?<*
z3&Q2I>ivr9FkA-t2%}Tma6;^lJs9Egg%dg<dP{RT<uOr*FEPTE(9bv_cHVS3<tie2
zn*c_KDkNrdDp7VhCLyM!ddg94`^#Wmu>E?S5Iub*aCz$pM&+`Px<yRHG(jbLaxy1G
zef5T%^7NP<t|&2%TTCk9_c}8|<Rm8n9FGOUk2nCajiZ3bo9crWcKd=&evS5yOxmph
zXFM4pqQ1MF))QPh7bVI~y@QudE$6g~YMvmcJSO_*5Jm{^eI2-@9hE_@qWY8wLU`{G
z;F9AuhT>ew36X#QSP&vsmY9U7H48Y(jthbi8Sx$vv2UwM<x<{GP6&7Vf)S#|)SyHi
zTZIxay2zka%-&G7%1)BOddiDF03w=<VJOS@n84C@Iw4}6dKQs6GZ<DqH_2(8K02Ng
zE?t<$fR8%JDOVBaDg@!&L0^Lq)+=95xe~p4pol2-*$H#C=d=oMa#0X2)^t-iBLCYu
zE8O<JK?pmgUX}~5Owg&E7gHHQv}(=?m-u6i5O#RBoYqa7&tky)>ZZa`lfk+&eL5w=
zro72znb+6P*QwFTU%iI8yV-2kAHV<l`!{WD&pq?|H?ubc&0D@|?z~l_d>a4z7jTvY
z1_dpdXI*uZ*Ms&A<vpxF|NW}1Y(t*8>e_j$RtGL$Hp*x4zpwe*FZcW(FtCYr?^_2J
zOlctRYyJ7}_wH=#`^>%jTE9}%Z}q%YfdNYb-wp_pkK?Vtxubk~{XZUxyT?F}L4!R#
z2YL<}<T1#-zx&|+1KsJX|KI`c?w*4@2YL+t^PoZ2qfBITKj|SKq4mel*Uoydzy0RZ
z@86Tw*5aA3|MYuMf4su)U;j4GoXzF?{QE2X_D=rs62E^LTRn4$gg;zDy={NI!0(@7
z=VvZ3xy!SU;Xl9je!dMGzNF4`weR!W6&#eF&bD?4mGY|AmpsqOn$F#<f7@)<+RxX~
P`c||#$bYQkaqPbU9Ki#1
literal 0
HcmV?d00001
diff --git a/5_dpnn/results/gpu_16/results_gpu_16.pdf b/5_dpnn/results/gpu_16/results_gpu_16.pdf
new file mode 100644
index 0000000000000000000000000000000000000000..3e27600917742f78a92d46653261a31b152f7565
GIT binary patch
literal 64174
zcmafbb9mfM)M#VdwrwYkowP|B+qP}nHrv=~Y}<AkH#Y93FTVG?_mBH*W@pZs<1@4S
z+b1)J?2CvP9TPn(G+Fr?;g6rtOoWVtHu~n!yu5@AN^Z7Bgbc!Zj(V0hCWH)M^h}H#
z2$=x{3WWUp&_>pVAB3F$VGy&ib|hr^L%^UauWxQ-;7G{v7bxoLD6Z(J=V(O8`WNDh
zo};6Yy)_}*2MEoeXsTyuW^F>q{`aiCje(+(BcTRBt%w-F6eCwhLIz1IfC)l>_ribo
zl7w3SF^B2@b>PG959{rX0M`9wK7)dhgN>8D0l=RR`vLl8jSS87glt>^8W{mECMH%^
zdPYt{R%RA@PA+x;#LB|L4A8_N4Db;k$bpdSFG~fjt!)4h(+Ago<MJ{6PatKCtW6wE
z37P&@E@oy4aGj7r%n}d}5hDW|L!%E5931V9^sJy=Gsm-}V;qO=vhQDM%nfvn^U~XN
zl5sW230)1$2pJ&hLW$V0{fX(O*|&Xu--itykMOSVIl{0$AYtyDG))YTC8RWq?{U2E
zFTVd8{O$94FYtbMm-D{8pYcm)^L?9a!@J<!yE$mp=4O+?=JL+r^}bhnO6UCv*?{Vv
z@#gnt$6ITmCqdm2gWvYcwT`y-+v9#lPMM#;VOCCO+ni?=S#hRK(YnETjU1bXMr(+I
zLJrQ$DoFI1w!)d0uK>=n-pT2vb&Sp|ZJofXbW4;*)RDF@d{Gkm3EjP|yg@yuFkO?+
z`|X*}nzzgIfl1i*0ETQ=yE(MCUBvy@+rj47gpM7RCzTAM3G^q*^lS%Uk5&Yyq?OAc
zoSz7f`vlr=&u5!+Mt4|Gf)1~T%GY35vv3o?BHKNmz2$#>x?=aWX;|^`e7tK-v3av-
z>vzu~D^!|OZ}GaHWZ+wmC^jBI{a#G7(STl1k0jmcb^jx!s%Nq3xA*l`)8?|uixK(T
zJ{L$%#zQH)flN_~LB8`to#MM6`Q)%qXWb|uNv}0#1EaWkec2mnZ#QJ)KIV_vgP~OT
z2@SnDS?>s}ZWHg%Sz6$imyhxdYd_wq%(Lx%On^pG>OZwKdRVkHxs5Hr`#1yfPwlmQ
z`t9P3)i;PinBq-96wIYj`IrS}sxt$ODmyTy7;>9hf%kR_?3p|40JQP@lsPnp;WO1G
zzWYhhue%K(Or4pR#rWS;3sc8bT9y|kzurM)v#;O1O;Q9sJ>M*5YKt({hqyWi{R~{S
zfN&lod_Bq=e%W1>!=2wZUYDPx1elz;PE$4_Ne`efw91SGYrDI?**aEtr`*k3=Hz1?
zo$3ZK@L6`^YMnk5#guJ0jO`}=aDJE>yc?Y!jLz{Q<y*9nxYmro<9}ZA;d)q%)_J1b
z#$Rcx9inhQ+J5Yv?grOcAvor~u|-;Td6?-1u=NAj_DVMr8edxSoF7gN-uHD`c)X8D
zq(Lc-<`Ux1FN$K$QVL@5I&Y?()X5BD9GL#2FERH#K%H~ZAY_vNP6s}^dN+kXzPUl<
z&9-t7Ugv4ao9k)mQkp;uv<Ch%ye<Z=<EG}-{-y>%W!!qg5I6fDlnFsRoa|-=Mv`;D
zA6f7<yBO-Xdx%sUCZ{Bks3y!<4Kc0Y${KT{yX@yHY%M`*8J8iQuQ{FXZ#o^M_^)&8
zWGhjqLY<hOw6_uL4+W;TH0Dg++p}HyJ9&I_rVw1~>y<5nBZ@qboo0;*)+u!Q4kI$3
zN*RW^O3xh5IbP}#e9W%!9=KmqIIf-p9>#zr<eQtm=a^6;Cwqb&r?dc*tb2&pb(VLb
zyx%})%65Qss`*_<y}S|fHvqS98iXRtjiTx>0o}szfs}o}Wp$VM1#S6YqOvu!$8+HX
zMsKrs4{@1>CQ-Cz0hc$YW}!}VVU+EM+mGpCLzGQZy`7`wC!O%SYZ0xvEpC&b&qSF}
zyylS0IPM&GUFJRr-;ej0uei9tUCo!`?FfuB!xO^0?$a-v$5%m5b0AXQlzjP=f$0Nv
zKX>!tNu(y%#Tyqho8yk}_b~2wzb`#!^;&!0dMuvX*BZRdw@lx|v{wbPeHwkdwi{lV
z&RYgFC+>0%8(w!GO`aDqvMtF>9n2_7>H^@rdzGU(vbEaW_Ls%6UBWc2=IUZ7UW#LO
zqbg!fXeRkdvnBcZ)@*Ms12=E%+G=j?y!MwvR`7^I!E?NLw>i9UZ7jVsb=>Zz5BOxg
zxu0EoY@RqbmY$kBu1<xI<;)eO=9dRiyk`e52NnizlPKyEs~3tq8v%>LyO%bGBirkb
z;*ke={g(fwLfra3dr;=+bQh}gXxG)?WcO*L4k4}|UmPoiKl>tsKboVFzp{$=IrcEQ
zP`Oc>K>k6NU=T-xU{NL8ONJBWy!^jj$it(K@I}+VJ{%eZY})4TMwt-UXuI9T?eekH
z7Kme19cJL){c3`zXW(5;L0mMCp&0hcw^`ZD*<9&tyRhtx_Mo-OCfy6`F!S;8ws^nl
zse8r$P1l6~Yq=ZXSuen|{@WxI0*5sXOJ2wScwq#?AmL*9`&RQ1&f7qdbHnux5WGn>
zM}N}t&M@Ti{6h!Jm0%|ctmtDSeUs>T+QG9{pN3ReNsi8DuiN!Sg^ZHE3G9Msm#U_!
zx7QOKUy<B6#q6dk@5kNW-Y@G2I-767*IZ--Qf*mG>m5%vo$tK_9m#tS>zl9pq)LJP
zDUQ!PDCoP$X!2Z+Fp|0?N!p5m@i%tzEfeT)=$uhdQl%Cx!D_EZX#r=g)1VM9;ax37
zkU`0;DDfRNnCa@{VoD}$()l@-PLh#)Zlf%ci^p8B0najQbtuLQ+pu|P0c1+fbrO))
z$kGuU@fmi3YUg_7lwi8$c6;Zx6)bY#C3fIKpL~A|5Y$WoS%9;(!1q$J;DtX+dtH>U
z$+Cd2d|_Y<HqXiLuwaN1w0K#@odb_a6hHDOG59{eG9@Qv1|L-=Q6B9!9&+FqY6lH>
zzx>1QOJ6;qo%(D>9X@_Va=mQ>yE}VZV)`zl=2KL1rC9F|R{4U|E%%>fka&va2JsmD
z7TsOR4$T;Fw4Bizv|Hjece|$~^qM&3t3rp6H|M6)poYWxO>2EH^<UV-wuBvMRG*c(
zZ_HcGQx<BJi6Mqb&9+xuxRH(()rHH4_1Z^5Eb1heP{6FX=r!|=81>8L$i~rs&!OXa
zO{QbfRmmqIW3|TV%i2=wjTbhSY#3c`Jt{)9(Q5wmpyLP8yF+hJb6d2Ve`Yt3l;_>b
zf6ZqYL-r`DnSdyFi1<xNF<xL?S<9wBXJ{d=5=D--t@rc<ML>cEjEfv-iSDC`EsG_*
z27<X+#H4YiCu>v=T1eWXc$I0j-R?L+kti`p#T~g>r<(=wdj-6Owk>LE8O+YM@Wr*C
zvRK0%b6)A9Oxr=-RaihUV=*7?l8vC496H668TWfb@UXHa6ejX?rl!a%QoYJAM&9e+
z5cEl3S?D$A<XMfPgDl(}Z(uQ?WO_qp=T%P0)pVN{(ZgU;h7gybu$jg<Ja|A#zGBIE
zt2~ng*W122q&n~&N=ps#Tc!UXwlDTBdXjvJ>}XF|5CDfd)7QABe{H(yu%)YiLMEFE
zk>C8HPM8%t5g$?h-RdkfsP2W<Ez4y%GtZ5Nd!$|Y(0vxd9H*$(mMKs#vCh;^noSf5
z!5VQ_pCJ*5|3$Uk-I*(p7-*T2`hxc9LeH=?nloM&el<=$1o4QJ%yhFy)cWbdfHF%!
zZtw@WA_cJqPm#s`Y>zl%i@IT@1zfw%na<beA+A2PM6EZl-)=kK_LubO)pkXED=K*O
z=Ijqt+MJ?uvKJk9a_Ue@W+^(x#K16Cp&aPjf_el&7h0|o;_GeI=Fsy&c2f_q!?{<8
z3_VktG1)Bo<-={KO)dQ(BX^?Mp04B2f~+p9ur=Je3Ir9(Z}Ua(V$k5wzhQmxctwGT
zn`M1v5|S!+)ALf$IWyFhY=+3A|MmmXvS!W{`sv-QPoh@r)M#6l7%U=Vi-ziXA~57o
zrncELS%i>6u}8Eaf}Pq>%CnKD9E3Wph>V1%;1UR?3Y*gc27em|CMW-Mc}1fN1{=q^
zj!jtKJXfB1zRX1nAWPD-HfN#kMTfJw0ZZnFa8Akjps+b$cnDTirWIWr&mhU483!1N
zDX9n_x@Vl@Y-80~%v01>YiuPkqZ+>Pn764sujGSJcN24c`;ouTkG;LawX!QF27Wre
zI!!pk8ttDUVz>c_f(SXfmZL^p`ir5NsPJsC2`?kFq=wseI^|KFm~&q~{+UVO5Nkq1
z0+OWFol|B#O9O|<Id@&!7}uxDZlY{|lOyJhNC^v}NmR|atYCsQ6Hf4vZ&S7lzmB`i
zDm+<5GN0fv*4VZ<Ws;`NCOX(eEm*7Bvs5EW0_P(qG_vw|+G9&mGn-KggfVkYzO;xg
z9`((TemPPszFaM~Yizklcl3XOlsv4o>W-(Ph=74xT!x8uc=?&mKJZ<yqJmZvwxBnl
zzBi*(`qtVBnapQ(N2qs~o2n<o5ga|GLQ*H%WZal|%Xhq3Ss@e#V^j(qL+PRwwWNc1
zI{q0u;s;3{l-gb;@i-ie;oL@~(Xf`kD2oTyLPg(amzTo=HVgfe_4GyMi2}N|<E4=P
zJki=L2k9%gZ#aaj8*$lRM*P{f0(q<GevpmHi3b*B?U{{}C_E5~sye}|I;eWXS(J%<
zE|@T^8wlw;KB(yJF#p}*<j+4WO1mxm6U09EqR_@Zno{lSv!ws-%9i?g{RMqmN(CX6
z<)Qd5Tam3}MKKP|NXtuyZRF3`v~ogu!!6(^p73FA{aeczBq-$!RHvYZZG}F|a~wyQ
z<lx|rw0MOzU#88D0ypl$i(*U@0xeuqR@kOnGA^jhaq6<V4}nQJg?_3E-t;bH45IkT
z^Ua@GN?EE2PsfQi6e@E1Q`%}Z&q2-3Ag(pU(l<-+SJU;Sgqv35yKO5NGd-9gHIx&f
z{dO|#VUA3kW)0qLo+y?uJ6(uQ3_SEtya<j|uh@*_H{W#=Q^@ctzBSKwx4kUw_CI+f
zkG4oy$O~qBG==sT)vi|B2fsGFNq67$>dC8YTw6&TnqwaX?UDXfw;zmWL6Z_B&2iEX
zCqryG^A&C}E-63W^Sog^KoNbr!0?@%>1$I^gkKN!fV%CNLq|rSixn-ZXd;6ZQHg`r
zN@S|cT?;|JrB^qyo4*w(U0!3}ECSWa6P*Q#X{!Os7ZVtaXU6rkUmoA-ky23>iD8eF
zXZ_%RH7iYgXQE&olr#3-RxlyGGNtjw(CjqZR>uW3KEWtsqxZa9YfI}&zGB<flKd>}
z8f0&7g+c$AIP@+3h^oK`ihz5BFV9ZGzZHe})bMNb9;a-BEgsx(i`-=Jghy$7c6k<I
z{Bio;7LKR}XhiNYP>MXb`K_;N7+yH`qLZ;*Gz8Vgcy^59&oa85DFQxyhS);4!Iz_*
z3|l*NSa37gC7r8ftk_dn3S=jwlym70>{2+O0-!TiuV$MpA|gqgFXjHueft9-<cwcD
zMD$tesVmi{P#eS81T%$^H6$neA%tRX;thLK-j=kUaLW<PV;9Io8OEJ;cQK3nC?pK+
z@yRs{afLy~Ew~b9>0cxizAH2w6xHz?3+B-KaT`8=Cv`qDG?WRYPKJtujtgRau_)>f
zz0t)nLO9+&c1Lp7$?7<)W564d4N!pH`eIzal`B2(%*=eYrdz+&dSiUuUPWdEQH`Za
z=#avoIf8<dJ+4Og;JK~%i{YXj<;O2f=f>>r9NsDx%@+yz+FbXh^{LqLwNg}yEAehi
zKPViBkQ}b<6^jThvlKZDC`ZJbj%7OC_#u@ySK$drW8TGsY4)o@C;K9fr5&j^*%`d!
zo($*Z`fdxChT9tMTx4t*4~XZ)!>nEIWNy{6r9BxL2f!zr9xQz0BI1M_KiC*2-YD=H
ztn5!Z^yWTmS{T~Op-!aPK7T0?6F7+cFV5Uxl?DZ6)h*Ll*9}P*xsnp9xU4|P@kI+U
zLDH(*F9?Cdhra~6#p<<Gh4?D3byh$ySo(Hixz)|8_+7Jv`#{)8divtV9JbS?VBu#%
z83z)#$92=7FmVf2r9pqC1uceGQ&d1bsU--|D+KT<8iFdC4VJXqDltZNdort$_GIX-
zD{$Py-#9-#f)d2%_kSwX>O#*<DDvHH)K`@_;Kks;gdtd}o~r@{Jxa1Z>{3f+yf?kV
zQRY(2!u^dY!}=IMc2TjEPROfgD{<Eh*?dITWo66VpP8H1D<DPutz7IY>BJ@zRudS9
zqu98oQPdB`7t$osTt_qbvCT57Lzt>kioOY3x&BwHGSX8BR%Ezh-LP3I0+>Y?lvn@I
zN+xVQi0shp-t@SM26hP%I{Z`|Q!(1_cRw%u%6HHD2%$EM!DV60h*lYIt9S={LNG1D
z*g4eW*Ipgid~l5yEx&T;Gvm4pv$FUtYe-+<g}zE_AZ8%_7(2T3c7j#wHN)oo`mK((
z?W*0Hdp0y6ZKV)=eH!QsGgDBSpU3t}>vACB#Anq_8cmZ`SQsEG25#wTcMI=uCVq0>
zbVnd=F~(NlG$8ert%?&!41Kw%+K9`a>7V3EXwN@uG^X-i$iNG;d0Rx^{1nTnIS$Ll
zDT-dY!NVx({{E{^y5|kkXd9WRhGVHkcKANpShF0)0fbfDV>|`~I@Q>glLwUxf8GvH
zez>0w>nmc2d53aY+DUK&n;jY3t_OEodR$jp+*_Dr6*nESHtWuE)ZyVn{S<m7+TdE(
z_00J~i&W~aV`Bt?#`sIR2OXvHrr}`bzFM8U$v`Exf(8c9rvY5r<Fyx70~h-dnl@=$
z%K0YaVe3?B%-F(8ic)D?GkS(LNwMk2tsC)j;RUN?enp|*-nmRY&%dB;n&~*zOGV>Z
zE!~+n<G1IX^U1o4A?U$F+kyOvYP!a0pq^DsmvFlJtqM?(X;so4*aH$>hBb=#vllb`
z#&bz3x(zp#IAI;PkWs<q+btBj3dF}yl1D;H&zlqR%y>?N!;?wrUj6F&m*LEV>wf)+
zZ!N}xsU0s~!E)->QWE`1U4Sa8#wvio>0!W<8YAJj?I&$Mz^@179nfyc$t42!45~w>
zu(_xrn-I!e$2etD!ZNjJ2zQ7P*d@ZVMXk*kygd|UoF=_ziv$neE}TDyxzfYcJVOsx
zwccy|08)XkoXm6_sdoobhMR7fz;Gl5`=jcY(~Kk%NnoCOX4~&eVj#MLtkvsglq_p#
zibgV(UL<s{m2kPAVuR$}D<$JC17#!?QR`NzP?P(r)i*Y$=#Ck-rLHW8Gf=HTYz*I~
zi%?BpahK>+Lm`Xqj|89Sm+0cvLW~`(N!s%BE?csWB#XDNLMFK4v+z?sPrnYC1xe8m
zRW2jcbnGZPs`N!xQL7SorgBf$>9t547$A@OY!w(i(AHH&uhegw4w{1humWG)^1==j
zmWx}XgO__*45pI%u^ztPBUeL;E@0oo6K9tWKE~$+WwhdoWFuW;JTi8E3t}R<1e-6r
zqej9-?<8J0J@pA83Z{Y-%U@nq=lVMGM1h-K1IVCi$vzSW<F=<*??s;$f$KqJ^iDS`
zt~kTyv|cJzEUEvTI~f56H(2BvsN^d<0XnOwKY|!sy4lK1hiEPvH-%~K^bCc+Pl9tR
zKc*<Uim0b!xC*EK3}Z5F5UcwMh$y(c6SGRv0T+E{d;SqS5>cRe4=*Z)*g*L1W$3sW
zXQ1kyPOZSWbR!(PT5+DaK29MvP#yxtieFK<{?F<Fj(}(E?xG05Fj`f0)hAz-7;XJ#
za^OdAP_2m+tJrkvDJW;OXtykkx+n$?nXs`PJj>U^>x)mfLA`-YUriH-)=wV8bNXda
zF|$SS+|`?hK-i(1ePX7gG;G<RepnZL-jrWALT%QP1xSU3`Q5P<A60MzJ1am&XAU|+
z7@H=}h65(lL3+!^b2?+n9=^@9E?IREj$<x)T|L%fdqvxx7ACjVo93z>3-gFPXSWI6
z(?B&Q2E;QMZ@ezGx7cc2pbNS#1D2aFY@FA{ObIo%zl4wFr$&iQim&q2(>n2)ogTfY
z<pK9bL?&I0Qa5->Z?vaBeO$<|LU|U?laad?R2&_pw?Q`Hw=^iaj6uZfmQNj5>`K5;
zVkVS$B=8199OF7IQmk=iv(fgqT019x7TFXHz|$CusG-R2apO1R#@s!!f<n#!1-0fv
z2}tY>3_AH+wQ5RO7SXPlTfd3BKm^{y_2ppWaj@BFdEPXto~3xhw>DeSjxA7jy&sO6
z>0jfnax!g1P-lZIa)@U>>riORZRvTV%=$5tXX|GZ)#)QXu3OFkV`<&h2!H>YxTR2`
zLML>AM}|8vs=ejrS;>xlrl7fb3TCJxdEL=`dZ&xdWpY5w;IO@eW}GW=u&X%|EeCfs
zZhJ=}xS0iAymZ1_hgC}kEVpT)P()Ud?nj(Ky#542&?nBjz{uqh%a|3XT>_`*(U%sZ
zo8@+w!qmR;J)$Q6bE@%_dzdHYw1H6L+Lu)YWLzzJJnjQxc1xr{3b-Y(20^)X*05ek
zC+z?cJJTVEYLqDAr*7<C2~a79afy7bb-9US!wAAy1F%w<luX@Vc+WKT{<p58Ttsys
z*=!VV1pXquFs!`$@Y<8a)-F<2)RJmZUBtSUriq(EYanDsNM2>{&owv)*5HV<bZnqD
z;_eDV#-G9reOHZLaI*@q7iYDpjL!ruW>3?BKr6IyJ=g-C+0uYO(`gruBG_sQ<m-Gz
zWX`OiL9Fu+O50~5TOHqm91aSwK3`1fie<qpripxE^x|!O2)Ly|9I+Q3F11L}L$F>i
zC#WH+C#8T=tth0cL0~m#sC7W+?`Ny_z9>8HC)<-8Q(-v~@7u!<kBCGekYhf907tNL
z$M^;nID@zZi9VQLbqu2-Z~WS&9rMu3Iqk^iX8cB&SyqusVSD777Uhd3RjOuC%@-b|
z`>9)%{8-IOvWsk152zbmye}k9+X2MXCXdu0B!5Ce)75I3me4qTu?eGBj~)mHvRDxB
zO&=QCHEqao>1KqR?_k9g3k0&DdyTctNs>TaTNbTCB6`e=-cyG6m0hZ@<3LA)uC@{h
zxH1b-_!tKLSKk{DZ4cMDd^nTeP`3nmnu_=k;O9gjWK*bd7VNoi-(jAtNHf?Z!Vcq3
z&#Fj8tDc<Iq3_!ZAu|FG^lNxgizHkdQUa0{Gn;-BDq=25vD@evDsd8ZRFJg!=h=qc
zF!{l3AM&_fNN8FXx;8{0rPCV|dqw=oN%LX#P;biT#sefCN4QGdcvSQHp|uob&F|N6
z`ZmG4xb4=?sCI6o<?87CB9_a<U<0XZJHEd(KnG4KO?l%fJZVH=Fql0g#jcKx_;eue
zSUjNeTOd2ss+rIvRT%>$nS^4GRhuUo)ns%(&j=O!D3kC`(UPNwPOs5JYm>R^F?A;v
zj8A#J;ML`+$&7WO<Da?6C>b+#OC^ul?l&9C)unzevb9~U3>R4gtxIM6#>chFj*^<_
zNlOkS4~mb~8;5t%OaXZ0tWh@7&n34fQFGLGU_lWi$j88ps}`27zcJ;ug{AGE>>G+_
zuw+4eF5XSA#cLs4I1llkqRt9KJaa>pK%u1bEX>DDE7P$LfTHSJf>?$F>h&p~%@_`Z
z%op0Z>N?4o?fLQGvnNy9uRUW90pmIt+jDJRGFfjMrw>bYusw~D48bpnk{KKU(<(4P
z4Bk-qgl;>`cMcAo(qU*f<eFK(nQ*@vD%2hCA5k*DZRd<%pjcCwDlBc%4~3KW4FrC0
zSKr{wPbwy)apo!22#JwtUnJ12X>2AM#E~`=@~7N@S%A>}UIeG?X3BZwB4fYNK&@|~
zX(f7Q3>|G=@F((eJsEiRhvS;Ct{o3dg|Qftym2v-H#A}-5HUGX_C;mF7pc~w2k|h7
zyC8FMmsP^7>jNuNebfT0n60>cuWg0O@S?F&m&i=^-;??%&1<0`EMidfLJ$~+#^io;
z><J<!x6#ICC#2v%2ehk!d%WE6_cxkol2Q6fGPshyh7J1T82PY*%mAC?<L?SKe`m$9
zq>-s+FZ$gPd2iZt;;PB5a?|bD12s)EjzL@7S}X=pkV=2fZv@vxVUH+Bsc#(+=ecD`
zrQygaEsrV~SIdw8K#RnNft#T3=!q;BY6P~nlYiU;F+doMA~Sr;7s(`TIp8ebWpUgC
z+3bi<OA{H=d+vclg9<NCE;y((*M0M53HAw_95rHo@$)bLhV|-P#JMj8@U0v<Vc56}
zp@kcw=Ru8Ifjuu0-CnrppXSR34FZNG&h56@P?Ve^tn*~OQAx#4v}h?oJx#?OBMPp4
zhke0<1oM+K0>Sv9M4PU|Pxm}lWz<eT1HPZXo{xSafJ2*1R;C@8l?mOPv-YJ-)8QK+
zLs2DS7;#kPA+>p0RBwltp_CwoY;@{Yjk~^C=N1iaOOiY#;%)+Co7<LB8PF;atljAa
zMQyV6mnIYPF=Q6)M2{h5?^vEDz40&I>CkmrvFxTZElp2^;=dKpu$^$Xgb^pTOn#_L
zIx8fN43eKN+`*B(a(CQO88?U;?q9(5u~%zWbD^azlCuf$i}Q9QDJ&*$gM|5=e!pU=
z#ddW{T5qop4W4os9Zc2=mEwwRNuZlDL(u3e;xZoH9i;7qKAT><C}br<Cn=HB5n%@t
z8Y+27co+7`)<h}LWjwe$2$L?11;G>1EwsBIxHouzhg;4w50y0H2)XJMK{Fk;9t}iA
zRZlXILSClzxRL#uIRjW?%qJV&dg7Y6rT-}N!Y?Uwa-a0b!_s|da$cw!h0)ht2COuC
z;s#ZA{HrPgg%UU{P2W<(l;?5HD}}zIP2ll2&DUrY@c5b)rU4@36UDJlp;P3esBTR^
zIkf#rwnLY`<)_^i3|<E$@GzC|$v5v_{G!<*xl1<1`RQuO-AsuI5qW6}pS&mRyBN8o
zK{+k;vUD?P-bbZx6GqM?c3%*Xkg>IHd?u2E1Fzk+)-yZf$2SG-89G>lR1J+}zFlE)
zi$a9PJ`yNwL8~c*dw?!h7Q|qYm_R8yMu&buZZ`p?Qb!8}N`U?yxfxDoZILmBCB}qO
zRgDw9Zpj)yHMQ8ndS#B=bV<W<SEA67ybEczOg}(LD5@yj?HL*QhuhN9Ju!AK81Xh5
znC=SyL?WmuOwOTE@0}eEf<j?HMMY?P&{6lozD)3!KvQcFTj}B}XK2Of_8`7eyMbHY
zg&`)u#~T$ZFEgelG<GtYzm4L>X+$KZQUMZDCMIY7L64oUS}_ZIOL`_M(ykGeSTqH3
z1_pg?AZ9#MNR;ULWwjVB!5t?Q{D}obUO1+*Wkf{QWAd5tNKduChFfY*@1>NK#vL7H
z_>uoo-3FT7c0BbOYn(V?Hf%3+4jEf}^A(||y=5?;0A$3~FFR-vS{hJyMU-J$vaN#V
zH8bauq;K(I0$~zn96O6-X!k+Kal_iWiRm%|xgcWeb!TA2+`sf+AY-P}DEJ$;@5W^U
z7kyusNEtocx}#28EC)N0=gb|1y8}!;IoyvGMhv#|<3bTGo^CL?%D4w;BSV{qC+xDo
zPwCL>wNmtOJ(2xR?fs`AG^p%$i6_jc_8u<dAqad?tn^3fp#=KfaK0+3j~UrPp;DZ}
z5U3-tw8!PJeE$vVdB>F6A7raYJoT%`DS2%NIm~3pSf9j8G$*tl3=Cq>@0w{_OP!Yo
z^cfwte@nu$0810RV_REGjKneep~TvLi11~V+$Ws|HUK6bjNoL@3R{q%I2CaSgZF|3
z1t?Fk2mDC&*9c&93>LO@X^7NiQA3hff<(YvjyuWKfrfu0YD(`f+D2B-?Ak%ez|tfn
zo;mTXAs4(}hNV}15k9Np%!Z_;2X!OqTG^AQZJnp3M+>VDX2T^8idWN&#o=D-x8?Ur
z?JrWN{!$Z)bs^W#sDBn&p9n|Q{N&EkytQBcblzqg292qCdQ;%`c)bJT=SfuHq%U2$
zD-*#dOlDwVP6L{e>A#a9sb-gdZ5!g7V39IHjBecyoE#b55>yEHJu}rW7x7*kwxOZ6
zcwSHPd<t$93aX-f5G5`uuos8k-DVIm6VbyNa$F^)uF2f3EA+aIPBR=?kH&B&7Tf4o
zp6UZ55yP}TaUt;_7z$7qxMXu`yvZc$ger;Wjj`M39h3?N%^3(~ObdLso>r?q9{nI7
zhU(#ti;=RcZHXy+B*iDh%V<?S34xIv9533oP{jg06PVaLO$b|k^yzB5i*WVjnj~g_
z7#tc{Nki8X4Xx$}K4r{-s>jjU=YAN?+`{1*p7x(WjCl>F0uL@l%FpGZSy##9E5y`7
zl?k`u^Em0T6)IP2ktco}96bRh(wjhXfr8FC2AQnJCbVKz-<Or*TPD_slxVjZ(q+kz
zw>j5`Kcn$y8NOI=36f5yWIoiWjTIw@eNubM`F3V8ZAr~pmu5{MCu+-<0sVTg?*U4{
zk|Eo_)2vGue8SqK%Z_#$Y1#}}%h|??l1q@hvtMhOPGI+Y24yc~wHd!L<;yc+wRbqI
z$y%_ZIaF{o#rZ~Hrpt=bG9!)ew96^hPo*<e#nL}_HB62b3n|Cn<v>jaQL--DNu(se
zo7Cqog|;p$4ZZe;9l4V7&qPwA%gwGa0?`lHh@o8>eJ88rj{ZjR$VJLI%z<t8jVeQr
z45tRKBNSbkaWvjkB(uF5cM%nlCV*d-TayCIZkwd#CCox{u7p#Yce0(69?s#(l^Q7i
z3lfh?Jk+i~S!@PS5M~KJx|1a2Mn5d&0aZG^v2_r0_b~9-Tbrc3u}}?Pk?=BlC7dof
z>dZXkz_nQ~df4ag9idiF=`@!gX;5Gk#(DaCJz*MMf$_AsTu<Fnq{0X+c_RMAV-|Z4
zz8?0Z)gUU9L)3=AVv1;7)as^_X0&}7UqdBxq<gAVeG~2?O$o4nOkDRBs6Y25g$)y*
z>hg!It~i>H^h>e2%M6*(69;X6JI+Caiv)PK2*W~;me#N@BL(lYlTkLvIHv4B-wHPQ
zL`kY~k}(x08Eh@x{3(Ueq!GDxFSQg*K4O=IK=U{;2jV4DI~_<Bc)@`2p~+uOYKVPu
zmd-C|in;;>+P)rkvZfgM{R_J^RA1VVl#e$ehgmT)WLm0F70mBcpyKVl>B`~mPw8XI
z$aI2q`yIB5#~HU8mt#2AH45{VlnWE3a%U2bF$V!AyoP)$1NA*y0S2M-THGe5FqK8A
zQ2oHAmnk^WaQvxILAK}9=<h_uk&x6?8=RNRyxBSZd)CH^JMMP;%B=Mg{(jro6|Mc<
zAz=oLsv`*8>Y?eZBvM_~?TIuOf_9(_flJeedLm#^;mnZr0UkaB8|cwRS#bX1U%~@(
z<o0eD$3lf3?rF0vVB|jI{&eWl&EXYlie1;t$NK7TCz*26MK-{=JtQ%{hSC)nnHz|c
znVbXDOl&~TP$v2kzMG&sVjLBEMXKm3*sltywr!^0FsS)J4H-ibr)z?v-vBBDOB341
zom|`>5rD!J_8RCJz|IGUulh4FdFhHU*n&A;EHt9;i~cawvUDFRF-x4^d_7i3cni8O
z4pKH2C(G!DBM~Js&2XRr+%Nj1&vB7afpei7=<@@fv$k3rZK&e*3lh!jZ>P>IB{1FR
z;1=Rwo`zjhLoHvj*vCjB)sNL(d3mXLg+lfy9`C9`<EP6>AXAgfk%D(W(HeshiD~n2
z(y68siXkc_k*0G$Ry-6l70ejfA#2o-?(jAV;jXAmn9vqmq`#7=#<qs61)#8CWwjZO
z7%Xu&N}!V7A#?=emUigkxv}_E_wuQg>Ry_n?LuIK%T?CseigMOa7}`t9`+}5Pd_42
z4K=CUa3%*vo`qOtgO9NV%1$(`Sx!crRmrJtiC*BQBXG9b5&MaI#bMhAvm2#kP+89N
zh+V|JKTuDXKm80R&oZ66!TxO!c_B6|J#p<5csZXhlcD-r4TKq*{_vOYtjKG?%T}{p
z<Eb!gV^<@k^qBjipI8M8K_x3>+`3&O$@Gx|nDM2=dgVcjfaDBo3c*yQCiZX!K20kL
z7>dNnUhl!uDC~4Y6=lp~KZfMu)!?U;hN1`5v5}O;2P^L$mL2-{OOTe`%k{Yvi-BrD
z{3w%jc$=d`?Tf^tH*Q7pp!p#itsG{d<K)@eZ;4l3hriaT28m9hKO<Qp*9P;%DMu-h
zU>R$e7YLzUluX7$K4Y`yO%g9TR-nd)=t5m)1%vTQLKQC9GwO+gL=7%nMIK{1vUWq9
zOdPE;{@qVZDOLi5e#iG4NKZ`B!X$r<etfJb7F42X4Aua*MzvXNV-cH6cF-^B@`m!T
zoNPAwuq`=Ax$53A)^sA==7eASyglJn%ZN~r3Hq8=K9Vlj6AIvQOIUvP>aiKm^3Z}2
zv<hssxef4df`I{Fmx`;q3w`10uq0^*8G@GmF$-lv&-}9#zFvoB#&&$!$#3XXJ7P`g
ztE`}Ug8#-W7aI@}+#8S`!sg#Sw+!W{8>p*sCi!)e@yQ>mWuYbz*ubRT$-oE2yz9dL
zE>^#5A{d8N&^+;e@#}o`H(f}Sy4y3ZvHajNeN|uEYjRFcV`3JDTMHaeFF05<kQ_;9
zenVnM%a>l_BlHa72MdvUVp@BJ+)QN+DeMM!S8U}|mZ&E5+0dwfLas%yH0P03Q^Gwp
z8F>p-yjo6KFCyw$OJfl5ECrN88>fhhThtR^t+@<!(y`nm&?qnjuYg7&4lvUPBqF?c
z-2hJ#!g$K-#w|I$3<;>|u>L7M7J@Kf)u{tqFM1_$;FG?t4z+bNozRDPe^hL~V6n4!
zFb>d-o(F6rUC(@MPf?K<<Ddca3zA3LJOm<a11!ibIpSoy8Av8Ji#N(@#v17@=!xEd
zZ@v_6%Iwzp_A0B@C4rQoc$x`k3P5ei%p74Qw_Kx!1tfBja`o)gcRNAm@5F9S6qm&;
zN_(t@_j0r2@ill!_JLXXIOFW}I+Cs>9Z~ZxQ3wP)j6}8FDY{aUwyAWVcO*G*j6Fn|
zOsg@N)l%i?NVAZZL(lSA^c<9YRurFA9oxnvI$$a>SBR}#TKUGN!+Mx((Bp7kF_lOS
zDC@oSLh_Qp6uE>VqQ%AJIegN?jgWQ&xonmBQ5exRvw8-2;Wv$w@(Jha^>F*nAKJYS
zk@$4e8Oc!=S=ES<f}zL@D}_?K+|gSD#}(BT;abI|4|<&pSFy9ch4oTSkINq~H=~`w
zIYWOha}?k&1;vv=<}{hAD~E$F{i0T+HcK}@4{j~lHpX7Uy;3{?l+fRuZURFBxzVT{
zbStKtRn?qb90+W<XZ&VL%mX_S$W&Vl8W};BaRlbN7fF{7G0LhgpFsht2&xbg<VV}g
z0aJZiK5apg^hGrM2~m0zGrSwlEv$^<fJi9@tV)?S0vmauNF?D@Ua{^wsgAZ{m!}-~
zcIbF9sVL3qPANR~066dBKn@1)v|<KbVPEF?1t}(yqHxm_MN1M`)hZ)oCPybI<8;nA
zG#V?6$<-;^7Lk@BPI{c_uV)41LN=NM2#Z2@Ur#n1{lEu>b^jrfjL(q{W{92B)!tGB
z88Fs!9g^&12+^7e^ANb~M2=Y)OH(aYxbfP-O}=vRHR6O1VH9buaJ0yyr-;$O(lum+
zg2CT8mb0_tUv~UQ>wh}A>WUt!#3;dKc6V0On^?*^*FX+H7WT38c7eOfXfig2K;2%K
zgh9qje6<;+TD|gz2zFRq0YMrN(1bvqHfKKdFIp2oUiQOYVihV2S;APW-!D1SL;bqT
z%4Q&M7E6<0T@e}3yGQ$N(<EJPlUqcah<8al>B^A!6J-LHC1JkA1iM8euMsAAh$PBy
z4N#Vr=IB5I(C0QxqKm%cI6u0Z@eV$Nis&UQ+FIcfDtgyy8#-2fRGV&UBx0N$R}+qT
zFbk<|3@b?)e$$?EY;y3m`zU{VnDcr6K2uUj=_^4+x0o*S)%bh?r#wWMINli>`)hZ6
zHpN#_<cm+_6m9)rO{NRDD_u4F@)}4&kuc--T6|?t%?cM{1lh8Ysr+Th4}D?X*-Y0x
zeJ?e%x%Eg8+8D62CM1Y*v1<v+5lIAwS#;-=&Aq^`Urxz2Na6<<W|au)X<&y)d(gPi
zdB>n;BbOdq=+KovHwOoWpiL5i1$)5r?27<DV7U_TkFF`U+^Frm$#3Y^ZQftMW!0PC
zt&dtw(%4kmcpk*BT|e#$<q53!o+Pya{&&<({{(lEznPLOGeG?Nq#2V=o%Uw>o~Dtb
zv-ADHUHjnS7D!I^HAc(#q6e=1y&DD*dsyBuAd|!I?82l9>g997FE+owPCZvaS@vJu
z=e5+exyT>b0rl&60R8-yP**y(E42#;I9NR_u)epSzURWz3cG`XLFVBm6y|{RdQ(z)
z{Ay9<i%L<?Uz>Y(w6Z-}*#(i#tFrYvK_Bq&01Yb=A{8nzF!rSAa{JbxqPx>dJb%=d
zZXF`f3QS;nrRjJ7A#j(5(au!=Iu=o=e>tbQ{JyFk*pMmiI|oDLW0!!J!Gkz<`x@4p
zx6h)V=gz_fVg)1}cq`UsE331o8jRp+Fls6HljzRUg*-zafReWF`Be+zDlDkle`79v
zTA>f{-I}rLfph<Ycx!8&gU4@^u;#-B8iprelVCaPsZ=P2=VkfNX=j1w!!b<!q6D1N
zi{-($(e4<Rml+7Ir$js+71t03FTe?GFTe?G1j{x2aod}k?iBt;ooIgNms5jkyP;B^
ze-1=Xv;dAemr8x^>Zc+&`$u%=jo@J!&O{79`~RHV9y{C@V8EF9{g<*`XMnOY_&|#+
z510QaYZEGb%si>-{YRO-g=ICS2+}$X)Y>J7i=TkcPrz^JSKBj)q9F67%bCewo#$o1
zvD;;z<~tBsl5`TCr{xO^fmL}EfoFt8?_3KshOMF;xX!z}SI4`$_lrvnFLAb$d?V?(
zyV<6vw+6R|;u{T5`MQhNl6hSafzq$<Vc<CP7n@rVKAlGqK&rOCS3}-2*=1E#ig@LZ
z&fls6K?kB=pGOrB?#g;XltDy68(Vz*b8?>ByS>h1*gL(ds;thd5PUq(!zw8h>UW7L
zb#n2W#?gk~$uz!PI$UvF%qIA-Hp{*(A2j{Gz0G{#0MK5QfzOpiu8(Ox?U&iEYR37?
zz^hrW?%8aWP`sPkX1vsm+iaeka6Q`Kzdh`&?qm1yjS00(Gss)<oVICB%u1fF!uHGm
zzk=V>hO6eB3FR_ZV@P*E!|$Tv&WywE*&2SCYTS9OG`E){ad`#B)SSzUOEY0DiZ?Y%
ziq2k5)E%kZn7Vn~O<z22$t_*_MdO}jFIVit2|zu+y%gM@^m;t^f3HYUr#}<8tb2WZ
zT)bmI_<fdem_cc6Z}uLRbM||(v(d-R)h5R_Hru{!>?QbLRLNN)MOsjNkoYr>_srev
zqxkb;$o-ugG5dFeoc|>aWtBvIez)kA!n?6k;C<z7f4kiaSiha~b@>fNV0AOc$Ljs+
zLT-afTDT_IuCa5Q!VQ3{_wfQK#oU#Oz5d^vJB`h=$F=VqFP1(&Zw;@b01swI9ktbc
za!pnzO>{K)30{V*Qs(^;$_6;>L=|syzt~#DH#x33Ki2*3eBZRGYI=El?{H|^XLy?x
z$WqaG18X++R>Rj<Rds*29vzS2pcUy5aruoz#AxfqQH84O*IBV_<d5F}wfrs*OqbO^
zJFJW?#QQ)@n!MhLxtu#m{gfRPL2n(_dwr&hSGYCK?(7!W<WAuj@kH#^dt^g?HqyB9
zN@FXvhqf)7Fq)B8t7N$-1tg|}OvqI$dHjU~aPs#TypwsN8&{HEQ#fcu?K!Ad9IJRH
zQ&Vv{a|xgN6Y*`C<U(GgK69!r__zc!eXqTm<F+vZ>BLf{as7!P)sk$qd1LO)Tj|pB
z_~9{_9c3_vN}D-FSjbP4=*6GKHLLB>4$ktUHCoKDKDRK;_{><cr6<Z%8#_uvA_z~J
z+nl3%f~mGOeP<5r+2JlfzBq<!uFTbKYAoXzwNehb?y7U!!_gt?*U5_gh?KAdZ>D2*
zd&`E-RE~R;BV{qe>pkvEY>pS>M(U=~@l?%E$D5GDv?Al|PtW_35+rPSAk|71^U!I{
z$swaU5Xe_GZkcx*Ad)DT__ZGO{m>07V9P>!lxU)<q7@3<?m*crbk4@QuAZOcnyiT=
zmQ)W&HJ}xaF27IOxxq<qjF327Dc5iwABwrr)tJ|B$)$8qtX|ftA2d-=1U~Iswg6VC
zAYj3A<I+jHm;cxYM}sH@5g$)CM8hfBg$x1p)v)W<d#v6ft+h3fsoCjc1KX@EsB9(p
z!ACGk-CdieP*%8b(~UmE)+`hTjiXJiw9~0VL4Y1Tv|gm#nY3Vaf?&=S`%@ChHC*o2
ziH$z71SQQdX_0bY56d{iEs6V5GebX3C?G;@=<$Suf!fN9c54oWXuhkm5b+dfC*c|c
zn2bGv?rB6U2Q(2}q4T8!-kgE?J!3fODR&LdK=4~aCcgst%K%PA#%2riFf3{BT)JR*
zLl^a0WcH!iEwjPZL$f3`Nk2=wKjZ#J+hPez0&t-Fbe$^chDTtWPR-)XcWurHvZ;h{
zi(AAxExk|CEc89!bf42c8)bme0JEMd>$UP%jBlm9bYS~X6_pH2p|i2A{uIx1y><3V
z)|r+@T`=`7QzW#UHv3JQ8^?-}B1;*vGl|kGEhN+Mj4p1iE$7Yxx$4wrXteSW<{8sx
zx^jln03lzRwN~GI`(<rX0y%zP(BFV?eEr;~L7lYhIjg%ooco*uS`V!D;njMrBq;Ss
zHV8Xp-6vO0g%~%<o*%X9+M__|rl(i)3Zx})4dETAz$|Z#QPd?|>)1pfc6hCvUAS>d
zmS*F6wf9>(-B?lt<0(YX8)cF0=zy%^-RK-}w>_#=tI-cU^tI1I%fJhP5w4JA2*W`<
z+e$Gpd-PvxgQn&V;rQSdyrXVMFs2X;)9esmpcmC%SKumvny}L~)hyV;2Y-@K)qnGp
z7l?Id+<iK>{&q!IuBQq#`_o$08@q<ewVRFhMBQMbU>)AAE*>=oyF!VTLL@7H)%k1r
zCDyy2=<yfY?+S+wuiovSPB8BG8wGw~ci-y0_5RSyzyPmbH-^6Z!LY4N``tWN;6zsK
ziE`(2U8_a^ms%akAl}ahq#SVgDJ$_FGD+!M0wOgH&1xub??d*eS*q+Yo^O};KFVXe
z>Wm#{iC05dmc2$lWJgAo6Gs~l>G$~xq+Ez%(t(DnM%IgXE7QlkDP5>yMxQ<3=!LXP
ze?+b*4%JyMAoKD#C0tQHp|^$4)9;IRwvcHT@?L2xL%j-*L^!)e3Uo-jL004kpAjFZ
zS$T0x=@hmRk66*Z$6i5d{LT(>mvGx_qa*f_a7)at%r<P1c5{w!rlIGvQpjH4RDaP9
z5FRP$y8oWgEbS&TWN)O^CgldY8_|;TBT-xsX_)kudjD%#`2${ONTT?1?M`<+29<ol
z{b);-g7)>*aBJcqd10_{i<H~4wQtj&uhs~OET>_ohHJ_dq>P*#Otu34K0||rR{2UH
zuQaC9otK20ZGf3DZy>L=Nf~_;z7YPt8K3t{>}9o`>?F>LQJX1PYQd;dYovJcuGH^k
zBVPL|a2F9x!!}d2u;lWFQP<(w>Os+gh=FRp$J^LTiP~HHoRCoQO6nxH&2U%4oyQ9f
zq$>4SQ~6(fB5m!fvV8?N!bR{KqJ1>V;iEM6_{8{W8_r~xqJ1R4TYpM)M7S0YDR-{8
zV_YufJJ3H+Z3~u3a8Q>7iX#aqAdFINZ*%4A%eP3l7M7Szk>^+=XRjm9$I}pG1d12k
zNE9WwNx3RDgoG~Vt&;-wlup8r*w#RZ_C4l!S1DJM&Wyl8=R#^XZZh5<wUZU*7*Ggd
z$vkOK-+Jm$659)R>9^I2!p?8YT%ab;6SC+*&XDo@qYf^3IM|_L9@7fHX_E2MsIc<s
z9LHYxYZW(Fj;xXK?iBn)x`_zHkRVFejy{aN_)>TMQ_l<IVgiF5Xwo_z-%T<_9%vFq
z6t<JX8Ra7PMG2xE+I8-_*w(BC+7*?QAjU$>1!$7`Ta**SGyPVaj3t++3Q!D`X^b?r
z3lIW5W(me`UQ%8q5i6@sb@!ADdK+FcUg|*cQz{iGgaGjfBgtxj>tF?$3!-DKp%9bV
z+I}~&7k6wcU~Rf;RMy`)L3vvX-5&gSK-&y<Pl551iWIJC6JtyaUcxMc1d;-CgEz^U
zPx*gbFU#)<X=d~9LTuW+X5=J^qX!o+dg=on8D7`ky^){_SOtRvkq~Ram=r=xdQ+v3
zz0On%zr_7)kgBDakuU>(eB1Za)7nu`?%cRV!$mc#|DO883)KI`*Ao-x=9PaBE(9B+
z_4&0gZ~fM<<IX9!s`u&5s5d9)Vim;ub)lm3S7IAa&SpaT)!FEZWio;TL5w9P-^-I2
zpK|Uh^qc{hf5qo@Wx~qG2@FGA8bq(!=Ne#mb+g?;Xw~Ow^BP{|uHYKM&ELWsw#8KK
zm0c(~&NPEUy$)d6hYqUsui*j$+f5K_*)$g`n+{SNS(6T6Z*68?3zyurOBI^pRp^i^
zo83|Kg)Tl*K=ldtoG$8TAui1tkH%EXYe2e4^Qz9Kz2<gqaL_AMYitc|$eUhP#3f4&
zS_(z{9qyb|lyWk9bN}ACl8*t+c`~bT$N*9j4$m8R0M5R6eC2sNxb}G~quN52O~+O;
z(_y)Y(b+8bm)r*zMHcLQ+PCp}nP+%kjOlz|KL~Vt^l`VIMEkhE{r8RIzi)E?xt{!S
zWtoYMm7V3E``v#oHh*9>2(|vjecbW>3&%+KaW9&Uk?~JsVrTygW@95{W@h>`{+R!p
zit~ey9U$?8{sYg<@o_yI@PhvH{@3{L_m6odF2+AJA9O5itb}X;xgU*_lar7gpzFUn
zSO7RSP8R4tw13O~82`ib|LZ?!|Ih=h1jzX#>w|}hk%{$>vJZtE0H;3M$N0k$4gk{!
z{D%?X)Cc~LfdB*k&;sTGIOu=q{^k2y1{W*Kzp_8%{bR&O&&m#P=<h)OYdj+(^grW2
zI{yyHM<71hM_@mCPJsS@=>F;dNz}i509hY4{*%1FlEOmB%*F*s&R;?JD=mK{12p|f
z3FAku|5p1i;={*(f&b?61No=1{l&%f!TvAhKkMKl3?Ez{;bHtE=pPOL^dCljj6Ya8
zSOBYm{V(_<Q~%;Vba4Ujf8zl7|B=N4SS^5{e$f7D?0>B$PJpBTw2uV*Nzg}f|96A_
zWB0$A|Ig%y(EqkSSqDV*PsRar|Ls4P=RYgqUpfHozsf$g#^0g+zx$6R|0h%bGyi8L
z{8#tK`2XhnUoaprf41NM$oPx?pH&6O&HrrjkID@HR519f6vRh;2N5F&1A8-DN1KlV
z1^`%2&k9goLC(xd-^oEy&sv5~!N|nvPi+Ge2SQfpKQuyr6;_~Q;$S7DW9IzWQ;bZ6
zTx={K1r;O#<t5Av1g%XhjQ|Q61RV@M>NI>zGJNRz06uu20fi`RC5+5WOdUVi0hFqL
zLmyH3qXtk{!HAIQ?+Owhwu$LknOV9KQvPKfA;3HbLK!MRu?a^bD`h}5|64cWgY6%0
z|Ekx(1gHW5*n$jyLCk<!5L^J^e}f<E_3t4eD+>oBz`p-qSqI0T28rZ?jcl6ADQE5@
zjaI9GKjRcYw`_!?rJVA3H`ecYKk=&kHWXSXkAJ#^)M8&jN~ZLs7>%`tgBh)^$sy0E
z&|nLANNokXg_{OvWq67@Exi@^4O<OPPeSq=t(K+CJKP%#Bu^UlPt}#xFwCDX{oiOc
zfog;0K9PfeA_2hwfxw{I-DCo}_TgILu>(TQq_sM0{{&r3Uo6L-?Ce8?<YAzqp~!<_
zqyhnfBdv%+rfA}!LSdk2Lhrl0Ir7Fw^Lc5bdv<tctq1IL-uFT|Nm<!z1{e|w3W~0w
z>rO=9(6YLs7Chh6-(BN#0~^}}X>Y-Z91Sc^Pa5^Z0-k7Nu%fcGT5r?zzCAtt{eBVI
z8{K15H&sy+%aIckb3xfxHhKBZ<UMzd?~q%t*9a3Lv{^Dy@9g`M@11USjdd<f*KZtk
zX+>GX0vgPAPPr-#b&if~NvMzVzdFfNeU`^$wLX=@%D`sA%6w83_bm&l0eclt`0mP`
z`i5fo60$rs(YLtTy#b_aY`SZCY++{nbPv9JW_f4~ZEI)Gc6<Bw4*c!_-a7=zSi|ZJ
zgqfE9t`+Pf`fkC__8ikvo)?i&1o1h;FWS!!sp}%<q*MH-55`>I#>Bw2&bfVXW3jJ~
z#Z#vq5z%vJjVac$+NXLM+;ShXD-Ki&n(KG?+fC9J450UVYE1Pv6F=B@Xeu)kV}z}j
z82&fphn)6LhwCy~neX3`KmQ*9J3z$0@lOF$0{%yWza`qcSQ?uGDE_4y9R^m0zk>OD
zWBfnW^gr<bQ&azzM9kI3=AVpE{0Ad|zmzxwWB~sl#m3O--~A13Ep0sik0<`kQ{D7G
zwEzDXNz%p8#?n~O&fMm|$O6z?I*VC)n3^b9x)@vhhg|*>P_;Ap=SDPDuy?ln$5sZ=
zGO{xMZ&1a;(%9P0)Y<v(qWlvxwKMsjY3W7njO|S<?aTqnE`N1t=w$NWhJV1uu1-#W
zS^DQn{eAYo=4O_EYcw_WFg1p;Tn7>qVVU4Xc1AS$RJIXDfI~-|VRjWIbDJ@+(ry$Q
zfQBsT`RVpbMOdSaDN%;&G}v%|7C*ZAPG{q*@u;$mALjib4Ot<;T1xR}uU6s{yB2pn
zx1z@imP+S6(A^j18%r<6K2%Lo8=e{Ot+`##Z{CljJbGL53g0SY|6I98Z4+v~gK6}|
z5kVd#f{Q~qs|yO`ttD)=Zr;araNK+=H0Af>dVD627CxIb2XSJ$U{dz5y#C2TJ}3W9
z4<71G>)ZZY4qD5IH#Z*%j)Ro)HbN}^E_nU2%TKKtFEg*WaN$P31nWcOxD4SYM1>Vr
zpfH<_DoaoUt-dd3zm(4M$%WAw<$!NZ!|86dbSUClxPceH5j@WZrSm&Om=9kVCsz=?
zVimxYiP5)^{bK&>6RQC-W;8b>O%~VaUM>>8**$2xG{u|C9=+m^KbhI}8>vxB;T$xs
zTqH*wjRg%IaPNH2$OoLMMFdFNdA4`H4ZDjL7uZ_E?&92%)T?)@5;l2w_7s@HP*`;m
zZ}34BFX7Nbe4<3#W;4-?5jp98IzPQIful|8WJ<qr2aR1-bLj#x%?IErJe7F}T5%vz
zme8>t`7*c-+?A8ld#moIMp2XkBEnf4aOcSR_QAaw>%P`lU;BMEjb*&k=SI@F09EQ>
zCwl7&p%5SIc&Pc1+oeS`T}e=R#NlbF;{wbplD%Hnq#wKNWSr-3(ejR16*AbQ>b-#J
z3}-0wKX>%`>jDNK*o>`J?q$|oK_i2B=&G+Ppu|(Z*x}!#{J(@Ruj%iB5eU<l&nwUh
z2&IhTmD2^<$=&SsocKG!k}n3bl@g=-1=;9)!S3RzBg+`l<(d2P7|rg2rq0G*jb#x2
z3~TDc^%^17dVkT*Z#e^9(9zb&pKKz3;+fiMrxBkuYo5m61+31YCvTs(1uHu>Qwa2r
zAJX*&^eA@J7k2x#phTxl^z}Cw;L0jM@WnTGuTPIt=&QuS^~GONh~ZGH*%v3Ome&oS
z3rlmB!=1CjP~Bhg!8E)SZ3$=GfR$i>Q-=7RgYw3)V`&#SQg}DLzI)EfjG3w>oG1tV
zj%>yi9P_3n;eV*DP+NW^jG8XYtJ+w=rfG*m2jP}_{>9Kq4z(_A{bHbFHy3T@*o7f7
z%BqKLj&u~Z!?@sXYle(~lJaAE<?D5jcD4p75Y;ySqnrGe&fM-Xi1`2|@N%_90xjBz
zPRDdK;6<6Uf>RLW$HngM{PgXqY&_xw`Mg+|k>sQo0XPGqapk(1LbXPuob+P3rWa_c
zDXA+_K;Fbi{=FJb8B%CLVz=4OCZ^YvCcL$-IUmV!!%d%OPk)jn>}>EQtO<s#)_^FE
z>g-Xu!4=EFhp6};)Si2m)kV8oE)ryQ#kdTJ&zbkt-_4^E#VCRvZtguL3M@GEcq>Pz
z3&h|DC<puQ8yO3<RD1h+#u$ULGhpXIwO@+ts{AyZa`6RKPKVfx>nT;WhJmhp<G(fK
z{UQ(}wOmE#{M6Fx^~(#pAN*_ZrNSEH^AYDBtvDorM78(Wt|M)~4GG#}zdq}4kD}61
zpx~MmBD<zaNaa0B<pdJ0?%hl@4Tn^_m!oA3UnV>qot~^TP#X+wqCagp45qHOG#=C(
z2Jd5r*$)nvXlAFRQMGDcCr@Eb#m_DJmz0zBeOmND-?u2|kil$<9*m<=@IlX`WW#QL
z=N`B`ycL~-as7_*y>g0V9L>se@YzWdp=N{hT*~*wi>fSxa1nW5{()#5**`jv=<fss
z?^<@NQKgdCTl+|{DYALaKq!Kj33Mx4X1cz8NFNn(4O!-^3}PU{yR%($liqM$SMqd)
z%m;g=cmIl#h)Cq14BnI6VeaEwkKaTAzNp?hB@pSclvKV3L`gcL^A4!IQe`ZbuONfw
z!fpKHydRY_-q`NNDd_957MZpE2fQ?>c|${Ww}yn45v@&=HWXM1-0wu&ILHt!;ALNk
z+v03z3_f}o#p15t!RfJkXlz8+ABpU)$Lw<;H!z{OSM3@YH#yXipJI`hN&C+04s__3
zo_nXyJZ;?ULX40(r#+XWYPqw27MyQAV@{2M{}<5^*0sqKta=P<&9CzHr8UEN{DXv?
z(-Fy2-RV0sle3S@`NS6m`bw>kO1TVeV%@2byi5*beD2IG!FXM33n$$W+MhOu`WMP0
zLftS&%l4Txmy-o_5<kn&j7`!G9{J>5C=Y(VNSM|)B+nx{uV+^-C4E@G@mQS>(B-65
z>i>S=0=TI)88lBVmTWQu-12q~l01hKLXO&>bDgWysK-c1jsj*NF2zXyB!uKPC-$^m
zwe2_AsJJW7#XGn4fMgsj6VNYI$sh9=kA*5dT<munSLAW0o7fF&E2dv21JR{gMw3_E
zt(k4Z+!TQ;0_Qe={W3Pjq&0U<0U9bZB9-cVD7-I^5q9;@uRCQCHx^R;(7^|KUNB$b
zYxxweL-ABbV{<Ka0B#ov*ugLt9u#QQvH+v1mb_vqu^i9mJO(R<aR{InY?jjiDyuv_
zWcPMi%#p;aH1B<=33QkaEH7)E?Jpyhr-0pvjGS=AxzC2pBL-tq)TAW<a%pd^B$_~9
zH5WCUvMekv$Kt4emT3=d==|Mps^ya|ZQYdc&~!=l*^z*#v%n4buB=OP70T;_@YX>t
zK_$TjiOrfCS@HYLuN}#HVFY3rHyS<k;WevR^5_$Z_T-dNQ|(cS-oQH#n{Snv%*|G}
zbsQlg%yW;<<uC}+V-&J+lS}_P*dZ=IT*lE)x2ofCCK%Z+HKL-+viM%rPe4<K37$J5
zjdR4-Pzl&aB7TNUnq$<iEc_s*D5|tkuy$=v@j)dpmdi8tU$94A$${`#e!0dD7T73=
zDy=Q*>I`i9miCtdG9U!s3k4nV(TiPK9Fu<H@WgG*9<9)(K$<gbaujgy{a}odJB&72
zvePuJ2(($YPQ=KHO4>(!mZbz~U(0FReym3pI-K-kY}&#&PC(pCrEYUbI-{}qc;FK_
zA#At96y1%FB+p+2rh5yO@@8i@hz3HRzDmNdm0MPZ^e0ioDBjs`g<01MCb2d?PX2QZ
zwwd*6*lZu=cZs%6gnj~|w%=#brOLS<NQw~n$;PwKQ=S1g7sLOG1|a+3kRhld|5TEY
zKJjOuBE8eH5x|$+6B^J?7F0mv$$0-}Rbv*|Z<T#q!wV~c)IE{GkLvL#fhBFUo$$*@
z7l1~Ib?J(00j?A#cClV(Qc!PbsoP7Al-drk({fAs%7Q`dTY%&;mR9bVaxv^cB9Zz9
z$I;;$O?Sxf_ERpw(Woh<hCtst2rtUoHD|py_~VCe47_P<N5CjRQ{XU5x2ux%u)Frz
zOa@Q)mKo_)#y;dSxvZ0`Vj#!>{j34wD7W%xV!TkeG!phy38?+y65K_FcR3jnpQK>G
z(0C4}aeY7hbzg(@(Q_k)Z<&Pb9!MRNrE5#1=DSjQY;ei_=@jmK?_TAXr{(2?+Gk3d
zs&11HG_{M`h9yLq;0^9EaoXgGR@<i^Y7NVM21?h%<!9h3_@RMJN`N{E7TbKjAKsHX
zwHRmK6&b5i)WiK{L{pwOFlSIY2Q@wU?~S771)3A!WRxeXz>HK-^?D2?hu_)HCWgwy
z3=kO6nbyO+Mdf{1XD##-dD4Z*Hp&e*QY_l^e)8#{!cN3+)NConl*TX|>{|EH>yk?N
znVh#1d#lyKIzl2%f(-!WGo!;HTcHd$RgVkF1+ighP@aMwv)uMl8l>d&Wf-`h`Ps%q
z3RQzw)tF@V1HX<2N<V#{xW)3NE1WL+8V#~Vtt2h&k9*Iq3u=e{Y-iw9clGgaxXNvW
zb>(H+RKu0xV|XPbUQGehdw23o;i43ClI1VxG*L;%&>Q86XOOtMN_eb*AMXT5K+(A>
z1G_918|8Kr0Ig7vs&KxaNu3OTJzHny&!1gKbLMX%v<}&*Ldi{CJARM!oi%vOOWcTJ
zOMJxC*5X}aFPD<28l1}?$#K6hxPi8iI8ZlGn#yA0ZEQM>dd#dOL^N&%84}BOj8!mi
zDN|9BTC)u0>$+#ZdO<8H()>N3Q7yc4S#d^|DS=`_aY-uqA{`atcqg2K{)Cs4kXKP$
zu6}{K4DezDMV;8LB?~xI)rtNtg%ASekvkE2rQ8Dt@?+8|uM>(W&VE+O26H7oCq0U+
za{&gg<`u*4jB#fVw_98uJoOwfUfLMKw2hG|(2K*pfg%%L0S}=)um*TYYrpYBvjvMz
zAR==#8qTPf7Q$`oE=Gi1e1??4rO`(cXvXoZv};%R_a-T%`fvkyj7UB_&hQh#7gnFC
zpaD`>^rCooP)w-UM1(W2w5~(mWLz((e;DvKlcOpFwn5bo^)SIPBtXcBLGRJqxWcE`
zO$0Hh%5m_8)fJ9MO`kwM>uEX%;vXgI*dFi+^B`ojP1a&}YPB|y-ure3T)bmGk8zRX
z#Id)Bqbrr%tL-5(CP3$7j0dUiVefD0F&x;PqtK|}^e`A?GggaF{aT^N`d{piIyF(@
zYL;eB{Pv+j^wM-cW4*BBX9FVma)^rTcwS)$2tmsiKlPQT!;wJYNP?f$<iV<%a=?=g
z$$t}au2jr)qsD7(XJV0(6_2pl^6IJCIn4HU<^Vhn(maVjT<Y!XF52;r*sPdlxc%;p
z$=#~wF3oZLgJ3z#fh(L7yYZOV&X+)6Ar-bats7`lkc@<2t<c``yF5tZ{RXP6a4k=v
z=h*t^b}85pht|!DQK1%{mp`32^ty@TF$-1FX``tms}3R@n=|WZii%LAzP~z@HA1%|
zJ9X*0_na#7!Z8P~^x0)GYj~*^#v6Vj@uRyvaM~obLQT@2Ws}BP6kK(LxX+$rfb-kW
zHwsZ+?_AdxrkeRtq(aAzsa+PQkgvxUIrpKJS6SV?z;jU0I~~C5W#ZzFhk8IuE9-PM
zc38x@U1^nd-jI@uRsebg?^otA9Q4ejFf!@>_-XK`gUePgzCw(&$$6Zz&BH~)KOLik
zw&H%yC=zbba;=2-F^LW^>F+AQSN~knBW|?L*PcP2Bin;r;U4@o0+d8`qY2c+bfeI!
zGr|5SCy|(wgL~O>io-K*RO(uzEx=bzkSK5v2un6pv??dx?llhY@#@DQp0^_Tp6C^E
z!%m#Je<JS^e*B<ib4g&xthRcgFLqr5;<&j)WNXPVKYSRA12mViAC6jbg#f!!00P=-
zKv3lT!T780;o^HCf|FJPH~^Dd5rW08e|}DuC;kALL;;SMpA_ep1va-mgnsX{jx*Ho
z8ZA_Aal&e~alnSfOL@$hsWBnJ)w-xk_6?=Mr3xVBYhue<Rt#m70PgsU?N>WX50(NB
zK^qUatM4!o6}8-hYDAyjBd!jAV8}3oIEbW7P^`aw{1SVV22PsPzOCXo+Cb`eB*>5j
zf%FG@Eak0vsg^o^WIKBEUdp{mry2?iD0cY*4dq>fp)yMPoYR|;n*KG2LH*Vk%UJh-
zGTrK81=>7G+QsWm5`sc>f)UkRx3M+Dp$ZW~cfjNfJCq8zS@gQ+ul@KjiFMGx%Ztz7
zjo%4Hp=XZVBRZ$D70m0>jhww`B*}}*-r$u$ytR2r%OiS`gm<=jMVc5U+tK^Kf5>13
zT1jzLmAjs3e2%&c4uak-olqm}(HmP@NsKc+;Ofv>kQW&(1qy6))4_a%eF&7_4@=yo
ze#U^~(ABzx!hy^7<gO!&+T(3RSp@<rM(7i=(J*fBR$wBL@ZZU;p+t4Lo0SCp@xr^m
zVB#_I54R&{+cm?X;wbo1R=E9N6h3kbcqCt`wM?Vk5c5e(FC`IFK*HD5(jF$`sChnk
zHg^sFRL>HlH>V>zbA!l=l>hFc=mUO%g3{2bzR4D{+=?gGLBoui6&CuQ)BiClw)k7T
zNEtNVpwr8@kx8k_iWZY`-9sm+%PRgS9x(ObU{|@H?%r;{5*_N9=9aO18IAS`wz4e~
z^~MnJ!QPy!9nT}vFp!)<Deuf&of%$NX}SK;^`A+3fV~y<^3ZqwAZ)-w#6oH5)1sM-
zE~x5Db2Y7l{gM+f_uwTeRc$;f_PU?us@kju0M1R*T5M%q=(otqy+-UE;f>Qk^TpTx
zqu|vdyDL@|g)tq^$`D`hG83Fc17;Dmr!aD6lzoFv3tWRbQ_=-}&By7-_%QG{LJ3h*
zA49t@4#~VB<HJRzZ*6{zHhLx>g|qtAT$MI6tIVn?gA!4thdH~Upf%-$RewZS0r6$y
zVIVuKH9-BnIE;N9b2F(g_B6Tnr+ZLur5w|yLm41Bx3K#|r3=fDh{L<gQH0=YYGv<T
z<z^wdXdZf=m{)6<R$8_+{)o^eq$^F;V}uLpOq@c|2&{pu6Um7g6_PnCC!D`K<XoHE
z%G(!ac8d>;J{XE&R!!UM?`O}l`~vov6_eJoG~YY$nz<U2_|hHFNR#n97h!ZldgWu$
z?7N+`p!<4j-vQ5s@Ns5@f(xZ+ID=$5nk<}4mG6LAEb@GF0i=6dM9@C-+$XD6BNO8&
z6t`-_nZUCu2lkv9?nY^Q@g?(k)?Q&<U#p5m6?&a-fAI5_tdXTs!0E4JU0a{?3Ci@X
zIm$>Z;WSz0_8T^?r>fGmZa6eW(|HovAtgkeFNkKDtT|$9NZIa5h(6TI_`r_tO4n4;
zhe0pM>f`42^}~>Hqw0H*`9X~8$%$~BKAWi<!C;phkn>mN)G05y5Le%<9CS}|4M`GX
zUWc7WGr^GGGaf>9)-yz`5||-6YSbLSbu+(KX?roz`f!@&;`S>mgLka>$f5X*stDO0
z;-|GZ(_C!v>)Svh3`3t8oDN)cIXo6;f<Khl+=+GKyY|L+i?w##-z>b~>DCN?PO@GI
zTPNHn=}B-uF7i^X3$fUXg5(aj2tpUMELs{OhJR~SfOAsfL~K)4HGO#Mjn<C1y!Bg^
zvU*iyX%^-w9#!MAu|K!G7{!0=$&r-{Jv6XToSwRy0IP=(1IVj|ft*mn)!X>;mQ^Sv
zyXIHEGsM7hWME|2)$GczL(Wf9%vg{+gNu~sV|ERt=Xm2tcy>0!`lH4SfCSD}2P7dw
zrDHGgr$QGkPf&S~=CD7g50lxVHGbA?JKnO#PCq^#Me{_kPS2ua{OrISSUSdM5ybq!
z5NKG!6IKlkwq`O^CK4}KhYN#6(*s>DzhT~)#!3Yj+Vz&U@~+ye93*5^??c)pY=B<f
zirtLSK_sm#J)c@wj+i3eyr^3&wV$Q$5@mjm!QN5fVRN6S4ED2g)gJgvtU(>{^<HOa
z6_cm;-0;n|rNPcaIP`(w%&;0os$dC*2Z)!Lg8vebKgiBgDpMUsTNPNZ)r_?Q;eEV<
z$GZ}4O1eQaT+7~^1;2IXdbRqZCTW7zb<!pDb>WKw;d@30_5V)c58Z0vNQ4F&qLh(M
zNsrE+W1kek$_t8D5`TA89iM`xCQspii53}m*lcQBCNCPq*XF_S$}Y<spqyA(3PYRP
z0`PRPEjrB@=zq*MPA<N`BW%L6xP@lwV~sqIn&l>NQQ(Onw;S{%W#OMyR`vrY<b5PO
zYwWSfp3lJE_w}WxaQ^v7-^S9eh!<zFHilZlVLzu{6Tzgj%>4M28~SKEkTJgn@UsSg
zF}CpjZ0me0J^1;H*%RFF*rh-_*R5$de~=Hw^A5go_G{#JO!jsD@J=g<Jur2R^GtAw
zYL>aD|DXhQ_V8=RvjB2nYATXTY2Q9oLEXMDIuZ8y&Sn>tsRxNVOhvQMJ>!HQ#denQ
zgxSPU=%}%5Y-P-;VHv{P35f^eGfwmr=$xKDDU{skae0(MYGm)-$7E7t;%lzk?Bwa(
zS@YiEm~1IcT|3-|W1yM>3=F20ym5(6KVQa@Tv(E!=cm853vRM%b3pP!qI56%Wi$?v
zx{>UL`UwA@avR96qSta#od%~?c*K)DMXiTq+ih?y_77eydkWLA+f|S3#bXVi^rDZ8
zx=xgC<eo91g`7f}Ita}Xc_Vc~-7MNY?E>1w2H{#yuyqHazHX3Zd#9O@RJy*wO#&xr
zPpDREzwg@OZ;!s0iCQh5qzCsnI>vB*wElTvZGTOprvq$)8rkHj7~MI{k5kv~3RQT(
zS~hfq^jb9~4y>20(d%60kAXYhHpEd@Y+IECl$EuJm7v^uXz?j)6+;Ra`cEdzR?*F&
z_X;QQ@2F7nyVIG&I)yC5W|aCteV^Rnn-Zj)zNCk=vjPZ_TBVX^C*>pJo{QEJDL~iy
zO(`!OgM(TK_x!90ZZObD^4q)#M5EplKx{UF!1U-WH4(0((8YPGlW3kP;78i01ky=6
z25S|=iO#1@xN@ccO7w4k3RW&JMS1URfz(jydQ3!|IQX&l2&c1J<0jJh-(47DQVqrT
zQI6}QkAw+s7VliT=YY7o^cm<AzleP@ADh$b1ck=Fg5}XaL2}C_UJ_1Kg5jdiqC6O~
zic;C4ej$B+BIBjr+l<H_5#Ft7)8^CH6L+2^WPT2=jM4SUs!jbqlF4GN4rQ>oB_}-;
zs9c}7@4jri%w~M~F9Wo|#`rHggu6{Ko<A+|`-j+!$gDuCq!|5eXh|PzG9J__%HB&5
z&40Z)$O`)zot7iZgCInYBpVh{pYHz2i*7!nQpugjDutw`vNpRt2SR(b>4x&E=)DuP
z7c){Y@n<ix_^nMnwJTr5APbn-QcHrnlIv3)WPftY(uCmRtH)?4HJ(lm9LtK7IF_W1
z8&5hM=7G8$DEDGmt!M@Pqa;`~ovVbd8w+-oLQdXT3Tzm<aLLr%9nqphL+hW@SOo#A
zOo+q}W-$=H9Ep~e6s~3srYq~b=tujtY0<x7<|+Zf@2<zB(FVScCJ0U6QfwC#qU!{^
zUXrGeNOasY@6ZfH=j3IEA-f`wb?OAHV91Ga#Q&hyGRznm#2uWwtIXfFGzR^MPq$o4
zLWMv4fW`S88@Wz<{pSY~RQ@8~w(WWq>+C}n!r12%<Bn|~8?eKK7F1Z^EAgdMM7@_{
zxd1dVI3+ttBA6Ufa*B(j_RA^6rARD{7wxlrzh4+(dX=K}kOXXw19U*AxJGGCjw+>g
zZTw+-@f5B0g;GqlL}XZ0L}s)Q`4weH`5fi?X*`L-r~`S84i4dDQB!#OIhww*Tf@(m
zqM>C;K`GyS|4eu5NU8UN{)|Q6Nf`|2EX%oCLQ~gdVuyysko#7NcsEp8eB3f$5;BR6
zs$<d@A0z{sXXIC{KyR4z^<$Z#_hxLkKa2QSi80feL&(Pdn6vWT4VHnzcW*tLHn0;Y
zZDhH=pRqq?2ZYS@DR8YWLDFGryRG8f2Lig(m3wifD-}bC+u(a_nboQk9s|Uo5|srG
zqb<N;nLOP<k3bd`+o}mh*R_7qfH;<eU;8ibjSLLJx1Gqwtlp8`3>%Bbp-4ASJR<~k
zSlklTPSmtc!C!n|n}Ic4?X4;5SnW;J3&hYCRUnTL5K=&S+^703uuZDYirh}TR-lcq
z*RHn~&GR7oV~##BI~m878RKu)dNh%MXb^m_Kde5onh+v7ct&^nD~@GqBb8T$sfJW%
zx=Va?KuU>E^RPpXrugSICrE`=S*-=$G3N21qd}_J|5*E&OltPIK=$3pr7O1>6nLY-
zpyUc}`t{Vl?x;J&U@vH_kN%4|9PaK*x97WT>9M${VfQIqW?ODO)qgCb!HkSmcdaFG
z-gg0H%nc+={icyCZDk}1e@I13QdsfH#|xHihE!BUr=G$216Uf5HZ93+SqD7ffoXeX
z3TcOT3mMsnn)UA+X{#c7v-7DerXd9qqnG0lb&hkh!`v+_B*b6TDy+L`;)ky+J5r;l
z@@Wz&frEq#GzwdH5L-F97>0;w!bO65L#S*eyh9%*O*7C4VjVi27DOZU5AxAGRB~Ln
ziJ&K0!nPLmFGuM%`vYQkbG}gmH-zvJ98?p6i9jZhqH@FhL%YIo$k3)o;wQg5$%4#9
zrL>^LPwvs;em!@LXZK&3$ZI*rMy;q^4B#5OqyXdoSH)ZloT{~i6Sm%JoIX)s-LYSH
zBoPuvkz(Vc))jw53$M+7lMcKI0$<GPj$Dt7nd*HG+P$FO?yaR!lj@B!$NBhaX?|=m
z;@wQhz7owwQfG<sM?0|@@Z3-r4JLl&a3ucXjaf$ZFns;E+{A~Kt*;Ctvm<$AW1AB@
z=7reUA<}@tHz_m)apXFc^}9=quhT2oj=JI;Ar5O84wJMp)*)gku6YT(Y(`k`-ZLxe
zaQB~~#rYOG7o-kn5lR9Kpk+cNjPEtlY6p=HLA|(P0+qv3LB2WFHsQoIMs*lIJK^HI
zLpqyKm(jT+Ql_<ysMp8A-I`x>xc+<<AF}04EHp{W)gow%1@EXGz^dNnS756c_3t<$
zow(^rjPVES%EkSqor2;UbzN*dnz=XB0>S=a_tF-)&QaySZ$vDIvUfr|s8MZY3N36&
zVeTSboGyhy{Fvw1i|Dku;f<d-KD5i8JC5x`spSHBoLB47O5~Nz(sw!eYJ%b;+&Cc$
z5e8o2@^#W~0<ysV;8SQ=`DjElN%79Odzr<x8e}m2hBACZh&}#d>~(Fe>RER>q26ja
zYE$Q|MX1`Kw{&+;q;Yzfjj(kZ+>M>X0L8|(8Q#}syAfzjEN|e*6--`x**OHUJ421k
zI*EX`ZYK8?#$c|WGt;UBIn!?Lvv}%*NL%xJ+e$rUWmPlGUeQX|UZ~j^i`8Ppb}Z$S
zxMV)|&vW%y;grl-T2$le*0LL2zhIW(in5vnp7dyxf_exdp5jc@mx=r553_ZME6#|y
z+r1?Kn%{)mVyoj5NThO326u*>IhxkSM;u1sC0(-ni+T)PA~g!!1kdQHMuO|mR<Osa
zE7c!SUVIHm039_nA-2m7i>Bk7j2TDk&%_q5BT${R%0TrdF$|6rF0v+n0w#SDF4gI@
z82Pv7ChI>X;u7WBEoSj_oOtvy)hd{wZ7%KnxcVlDWo0<&k8;`BV}*QB+;;;gIZ6h&
zKo+LfyUuYLQJMW=Nn_8waE;(IJn9nj*TUfag8ruwfxWx_n(*|;*N3Zu<%>UOe2zl9
zu2Bn!-VIdL5;pFA8_?Q<g*fjBQDtyt53o3pJD7&d5Te1+ApGILRGt8suRy5<orc&a
zx9jigMEev%0z7R);BMsOaMlP$$Y}+YONfVm-?&F!ce10&okf!v;4oQI&7HJS-u3zY
zKnOu*%p2aP9EpVkO{v%On$cEH9y+Qpr&uvRe%lqBA1|w-O7-rIw<P>JN;NKtS^3I_
zZUEV$yoS*htw77F9U`ByVGl|zE>E7$Q?eeB(z84JpgqS5;}fEZak8Wx@VL%{(p9E_
zVv13-8IGYiZJ6wd7b>u%%&-&Pjyaj7-jUsB<)f_V(%x&9kvyne?0i-1X~fi$VH^XU
zw*p)IJZ==e^P7#JX+WzToiB-yFT>iYKMAvTOs<Sp=$i4E%{)lBl>a4>;(4Jmf41BK
z2wwhHy^qp8@(R>Lbd{??BAqs=MIx~tG2=b#J3-1SnJI<Cc7hFf1C0Pde7_hAzn@6;
zXJm_8WAQ#2^JyJ%ZfSi4!i~#nr?BXtSwKCH`(?JJ!EYG!GN`)l_EV`K4yms(`BEb;
zl_S>~ikd-Ge_teUQK#sI0(I`@Jmnn7k(z?eDljmRVhu>;z7ekn@?%RWReDoT0mVW;
z_WHB1yG4m0!E)XUy5kX-rt}MHZ#*h6yqETl>}T!|fqa6RzU<3kGh0UuNHBT)bCtmH
zKQpW92+m>}2%vHg#V**{&(})S85qHQzc*EmY@e^A@Fq7`62zX+Mdy!CiMXp&e|M-<
zG3TCsG$4LDW<_U73rHVM1GAy|mYU3IFz9L_zzKnx`~ti68v*4{sT!bj-T69r8baoN
z9?P=NRyHS2Lrw_<R&Nt$T`3AeOpLG@c0gZLc%e<!k+5Dl>O#lCW6u^Ibk<dq2d2fo
z;H&ElM?VHB%w0>M;81<Q#eTcG{^C_>j(nklROt}gmHty6^5hXdfqF4@im6)h^<Z^6
zTf{LLYgwnCbhXT$dE7^f?83w@pRzeQ*GE~1WScp&ih%U(Lki5`oVn6iY}Rs%q&r(v
z&pN?w6j(l^dm;idU0Czp?Utk@^vq&{>^4TW_dBg~<`kYdA7J#&^n48bb*M~yGm}Ht
ztJLgU%~F=jmJ1*IRK_7%<#u9_)hn6%ESgo9CM?p_QQ9Bh0ytrJ3ycvbV+F22FfM~5
zgD@~uPjs7vxKggdB=~6pgbstgy~bhgqHk{@9H8u9?U-3#IG|fYC<D6q8kV+OHmSW4
zj3k%j^MS(9h@>-jfD5mTPkU#nj+S|~;7hTpa@m_u?s-Ql!r%d4U0UKOAJ3Vsk5qgO
z9~wN*hw@mcwY`;fydX22%OU`>TDqjEH#MfG*KzdrdXln9TG*z{uh|0IW_ki&Od2_m
z!(l`pD`1}RQ2K~g@f8yRVH*xs_5-6#0f;J3#Hd;WdmHy=b)HJH0_M!?(^2}Qa;7WE
z<5=@^d{RaH_g8~Xv5}IlZ3|-U2>LGG`0<)qxf&X$s+6YI9@6~274nA0=Ulss(k}pH
zY__3EOU;7y%;f@g&qWLzh~P7`c_6H<y+4G=F%*3~2$aa2`|oRL20jfQ)$dd`@xbYC
zkia_f%Jqo+A4O|0+uh-vrMf~i0oyXSsJ~uM%LNoN)9S^iwhW=qltcHSLTHzfUA>+$
z=rOam(;yGLkdM0T<>sPD=e3Rml+H&Qoh08LmHU$Ef>J(VyxZUMe5Vxl=%+pXPMXwx
z%NKvqHpcvp9>jW0G%SjZT%=D9#%eiaY^x~Sh#K(dB@Mh{Kjk3(d?1(;G}MIe`_;eM
z;73}#u$l)?gSWGt@d;;+kEz7^1J1U)LGaZb1Oqjf&J~UXh5zzwH~-H|X%?~K2(xRM
zHhk8sQ`*ZJ3d+)J0%4-$33KtarwY=Ku}fL6v6kay2dI^Ap8#~h1B>iq8i-4eP&cO+
zXbss4xgmq@gbys^B6S3+W!mfL&JdXFUM-9re;Td?4(VRIYujVxS{$RGu}2Jbgi=_K
zzdPsZyj+VtmgWY4W%UX&9zR_Un`>HfMzxqV^T<&A>QhXT8zCiWpo)znc+ev<8VNj2
zlwT997dU5((EfzJa##0aP1p>JNMY5e_8q?LwYWt2_QRQK*988_Fkp5BgYvUaB`|%D
z?(kljI_an7Z&QfcY?w62<r-UG0C*Sjh4<!Ml?Vx7I5&cjhfU4!urg2$iS6FSfy$QS
zD@%9nMg>(PM%uU{5%m?SPO$!%Z`XYOpsGw(z8F|x-!Fk_Hc(L6&AN(L>z)%<6ss%(
zHzodh-JEz;-9BKUg2V|8%a~eTxtALIFfb`ltHz;~mV(8-N@J<g4{}z~^n2f`(LKi8
zdtemCdS=HoPy6G=BBQv)lo&q1=PD_tDdA9I^FXbLx@9O^pJ$+HU$lrBOiRSkH;?j(
zOGWDMY1<2^S(9FW)F9s)#0*SATM~EwiPS|9-yELN9JxQ;G+cj{U|(Ubz`xc2LJfZ&
z1jS1XL6THRN(<R}K8BjskIM(DxWJ7u@>-&N=q8-*2EsO?8H{mgvS<|B1=zRE)trP(
zsZ&zoil!4)JK@T%l4t@y<J2^0Qp%Ft)QSmU+X=tPB~1qQAGfia+n4YBvKwCKr3n%P
zq@wOfX!a7Iy^tih!SeW=@S3dwm!bUdFyeJpy?Yrhix|KvltXkx6iuV}+L>M)?kWvf
zuakrK28PPHg|qdvF~Qa;ljxxz#)T}ykN4Vm+lkZC3-Lol?E6+)qiW2SuU?>dlQ^&G
z0Ql%=S^~eIy+BHo0Cqt+4*|drFW-Js6M7aQ>-8R<Y-xL*1Tu;Isx`;CF3>zuN)fS7
z4n-vMXN1tg1Vu09L)7z`QX|$ZAVpoN;PWyvrq?PoR_)ICEug+D`MG2k?ICs>yf-Wy
z<8RYHPh^(dJn}i3ugL0NM0EI~JqqYRe>GioJR7}JAb11m+a^6fkk!SH=YPe*IZNCe
zr9qq?bCDE%luQ!}eC;dp@4jtU7M$wM=B2^kk@f)b-QZdcKC`D=M})@WjY4ja&WLqf
zZdp{MxSr;$WWA|xpp<6s!pWRKWO&CEsl_vU-p*R-aMpwp2eTJte~{92Dbc8j8CP>x
zUyktu@!T-|NiuWiyH<|?t(^o>k=J>6-kr=y4jnr6`AxIpEw;_>(;@QkoS#v2{A`^U
zFe9#<*<Eb{)@s8JR;vvBnE%G-`dAY-O&?;GT%YG&HPsfrsc-mh9hxXY7}<ava&KTH
z-`ufAt6=knW4VD$S3XUW-0@p>-DoFaG|IdNui?9y$ztJ$-QKh#ENN}+LYQU~f4-=o
zxOrIwo?-#l;_IKYbcScV_N_3a&o?zKDt{9JrxU<DWDsz@5bcM$C0o_0OT-<>Eve93
zyEap#Gxh961P6p6BZb#obpZ~ZKrF8D4|*wem@!eDVmK}Z+zbzaCwwBDE)KvIAtn_O
zd(Qr%=yEaXF15DOUiGHm$qm|um1`zc`A#><udV$K*XgXN+J2rtaf%?HLLB?rsmf*6
zlasuOmVTuYUB;*zJsL+!yOGG#<GF}6D@cqDZfNJLkP&9wOr-c{C+i;F<afquuBC2q
zW7X2L0W6-$*}9%ie#(7MhDT4M(qC|BXK%~_+PmVvqS3#$aHxqR4?L3GhfDy_L8?+i
z%iKI9ekAGj*HPah&#hUNc9my1R?NrZ3(1(AbOpkH40DaYN;2%rpIL!JwTjfv<UMw(
zK`NnI5CQiNU7s0&7w955=r8L(%xurtl<8~V7Cc$j{j4Y}&4l{FD}dHopjz%_fF;)f
zO){1v<~R0QG^jh7kRRoMYARvNUmJ|_UC{R_g+h^}i7;^t5TX@M^BQ(&FI~_RoNW2k
zm=&PRf@tP2Cx*vSWaqe*>Q=3M&|{qEaAWL;4~9Z4Kb5tYvth8O+akWgaAzLLiG7gW
zygK*P*j}hsO<%)E4P~3(RbfNq2ZTo75M4|L%Wt9lam(DfwUi5VUk(fQTV>`GaKHxn
za*dJl9#Rl~bKW0#IEo1|%;#o?7P8k}LREg6ywQ`mQz%c|w{E>Kss$YsHw7AS$Ua`|
z(?9f;(BUCsxY5DK%pQ)x>F}6}rHy%%J>{#KdPF;Sds0zglun8wQ>hz`96#mLA~6<N
z7boM{#p5!;C8dmm-CwMm#8e2BU8>QOW@k!d9@_oY1!2)imcK5`VkDpV0$Jh@(wCZ(
z(aVva)_UgqbWj<%L@YgyC+}XZx`s7YR+!@xN{UA&1fLT-1)R1|yXYM4FU;pBcBq%@
z4&J`BPMAysv`p+eUDv>nprL*~d$kh)p*_kirk=P8hHX%z5TTR>EC{N%FwNsL`$XM7
z%WC{XMzWbr;J|{Lpx)3NY7~_nuNa)%fZ5qn9%ZJzxu+f(_9JhqPP&C?D~5y<{cJYu
z9qVJMKV!tVzJRe@^c{Mt8jA^QP#<nIq#AXdsl3Y%#36b1gOnlytBc*#QDE872JC@=
zM`of(J%jiA3^qO^oFXOgiFWK1lFHDNvA8=@0>?)!x;u2y&UGO)zj|=i-Xs)+x6%_o
z18vjM3A~Vce!%iP5#}qnOu}rq6<F{o6v%dMg%Qyk-EHn;fbEL!gF-Aterc5+HIxYL
zmAZ8qWDej%>B6Rb9WmRC*?6Qx=SD02OTO2nZjcF#Ds`p+Lo*IQm^>VQNa$4t-6Pg%
zP7l&Kp~y;&iYJ(SCqP?x>IcT%L6Cb+_Jvk}%TtZXT-?P|1zK2y*HQ1QI}e6G*wJ>3
zDdH&)4whq+SHJU!5fn&mX6zFx-d?H!VTx!SWV)N^C5t0DXWg{g;fCSKj&qkc(5HUw
z@+%76wdvPk;>+c}Q7~f}NK1N;B>`weTh#m<Tm{z>^dXCgIxPqy<;T@S(QgC;0P=0o
zsttZ{n0Nh_#BgxMvAjSIT5>goEBc19azaoNwKz8s-BAs!8_Pl3d5;$e5?;EYcLC$K
zaWK*R@{-%e-%sS0Eazbj^LiHo1GZUui{&&(r1RCv5M|uoIEn2&wB6EcLYPfsuAd<|
z>wF)zfl;e%)A{e_B5yzA<R>*P3N#g+F{)9$K|ugtDM$&)E%%D3#dHeZkf{yUKI(@O
ztoQ9~@N6A@U>x{`K?qE88B}>ce(WAUiX|hv?xRPH6GVpg)|4VO{`!E^TM;4!>&QY^
zb}bfuP|cDeseW<KQ&Q{8{i=e_kFhXR@yG?M)9q&1b1~lGCe~=(y!Wf1Fv<o9D4&mA
zk~yNyxlyyFT3UISQ6_*qAlmRnSP0JJr)ZwMADY!if+dD>(}8%q2!9dsgCm++s}VWk
z*&N9AO3nJfjk>Hj(~?zpr277JbHf8U7_(6=8m7<F1@pOH&<qjpaHuL3Lzpzm?)Ikc
zxtYZon({M+SeL5-s$hZ+wIoq~EuI_{25w<Ww2cgfsW43R6MRz}DQy7Ztv5&DIChAK
zoGgTbfhy$Fj=$@jmT*!Kmk?36w7rJslbMk1)QAd{|D(8kNGB;tRTj{Tb~M8rpoF8S
zVT1+fYVTqm)E4K+&gNe7qjRZE9PtY#F!t0nyuccQgmkJrt9DSdhh7_pN08Wx4H`*!
znr|fSEDS~m<p@=NLG1Nbx`!>#w_T<%ssqzv*<JkUTcI3>z0Y#>qieqp2(1x*y4BPh
z41~+2Bsj^>=xLgKcfw|M4CTn&qM6@<cW^xEnDinxE&rf(<|;;$fVL%DVlJ{_B5rY_
zjp+QBPgOvC3$sg7_DFpa^X^E(Z)oYC+HT2+w#J(LWEx)<;ZUk$Gj~U;0a$lkwi!db
zIO6Pd)&g7%8t!rjcHTx1U3<62<o3Y5SqfWZBg@L@65+FTBtF>pWMU51#5O<J>hcn^
zih<RT9w%eGVPWOn+^+-9x*T`Ko&8&kQ(}XMc_n9lJYTI6g~XO+1q?u@Di`=2nx@^f
z%0C;&7MOC%%(gJuJLi<|WXRZ}D&?5))4xwnMc(}J8wjoYRl!c78R4>&UamUTMZOx*
z03F6^@!Z~rb~rc#BWug{WOFT$;BZ_vPSDB27ZS@z4wN?r=T(AGRh@W!#|pW9KZSv<
z3~#VVH&0X6+MD@|I%`Qn3BG?nS~zYs#K5zb;eF9WXr9_$alTnIyuDhdy|!JQ^@&Ig
z|8Tz<)r?!`{^8&yDAla=>UjK`i*L42x}}#61u4z72Mp7XPq!7nQ&+_ImH>i7Q8f)G
zsG{|Kuo=_lhcbu!vgg?rIF}$=7gr=!m+&O(#HeZ*Q9&19)-XcJ+D4B-;(=3<Qk#U3
zriRq1F+2R2OQ^Sylt=zMIURKZvVx|d;Aup$AMFcOJjh@9PXt-myR#YTE&B%tg7NTB
zdn}#pZngHaBv$u~@mb}jDhh;G#=FQRcCK2KWIJyzQa-w*Q$L0CJeHy&MEu<4?_omd
zP-6i~@P+}BT=R67#-K@{8aSg=q9)Fw1MMCv6VQ_$HAxW}uvtfs?e$|#pQoyw)B(j=
zffKwKd}>D2NJ3(rSR`e%QSh;cBgeW!!t=sxK<_zKT{C_l7y*&%l6o#i=j)@JGoK66
zDw^(#C{wwT2enpe11pOzME7--di!<Ev?AU{N<#$0^lu@cuxWmsF)$-Dg7aC{uM~7Y
zmQtW*Y}@#AqZ70#oY<XpxHm{`eYrc8fV>AHv67^_V)c?|+%wQ|9n2qs=(daLL%a<n
z$OG#--cojw9xD|4gC5N#z@$<noUe6%3}FLacxLYD9*fLyq0cK$g-KvD$cBTSK~RxT
zE@lpshWMD5Gy<f%td=H%WJ!udQ?wGN)6}ml9E&O>xD2A$VdCYj5By>l>jJ#T7E&{8
z9xXn)S`CDKQ(Zib{Z5;}=(S<4bQ<<NaAno$pQ*wb93|grmG6VPLpkgZ=rrU_EgIUc
zA`AuGr3#Lk4sonBT8^^uhxUso5*pfXO78U{!z2*C(2h;9GqNa<MEphUF8&ndXvGjv
zD^@RBO;Cc;ND_P9_uE)awv-}r(=JSiZ_Xp8cjo#xIzw!mJNr&hjgILdz%)8NokV0v
zP66YGA(4#zN^?D_^O)RK)|w|ty~6<ApmhZ<)1E3@64kf|i(^^q{)EzFas*22bZyqE
z%&j?x>2VN_EZ{fcj%G9I%&K~JGk6+`sM1d5%Zy2va=)e>?V88)2}$Y~l^N3zj5E#&
zSZ(DtZ(qnhy>;~BHO{Sz&0r~hjzRpRU6yDxH$K263+jd`q(uJ1^#LtE|8tgoi7GO3
z5T-pAimp+-sASTwo{xBmJQ0}Htlh>f^9%e<7J6M3#P{*25~gNM$L`JI0o3eJWWfmP
zM=V0qj>i?9Jx>IZ>G){8So)#5qI<pfB?!y)gxvY|RR^q3OnTB}f`=_w+xH4$xSkGM
zK$qB!*ss$$@2?YYQ<H>ouHvT+3U+-S>KK<1JiU#PrZ$;A6llqV+;{yN%6M5=3<@$F
z<V{`2on;X+p+;=iu_*a?8xlBUcDKCXxFOQsKp1(jc|0@y)IpJV5*KS|9)5wuHbwi`
zyQ(~%!qMu~H*=FB0>a(_8CnX$AfC5hn9U>E0+h6b<F~t*$_fV^>10hrUno~|hikv?
z(M@KugeQmfcBjb}E+GPD>lgfmM@{4jH9YS&)S1b&Ge>RV4MF)Fz<)U5zvn=5rKCeD
z1KUa8L`A(Ld*e5D`m@BPm+9V}&B0s@(B4d~o*tyT(q23CR#4f?xF46=A0X(2g}1OB
z6>g%<tWg?d!`!{#u}V6P-E7p|Z`kv@Cs0TqdxxSA9PxHpi<o5!=F$Y4hEEb6pS$5c
zRBk3J#3Yjf<&WhW_nO52<Hx_5SF3Fx_6vz^_B50)&y(m2^~Q8P*sc3*o{XP`@c!16
z`qFr25kD&8uQp5D{0@RFN_w+OrM1-2S0W47q&S#chMe%Qpl%$^q;_u?27nhG+4228
zkvM>=V)$QtijAftC^9|Hg(-J7mzd8hz`4_vtA<(c(rkCZ{4GMxR%hYsgmT(<-u;;%
z35^xe;j&~c+h7Ctyv4aRKZGplcwjDB#iyAOwZ6z%2=YOKMS!HIMz4tAIm61!Q!xbE
zr2e$re;2|NlnZ@VBVwh65QS(3V?WF<@(VqF7G>p(_=|kV5@9yL0>RXZ9zx!-av35D
zU4$_?b<7D)&d7@kjF3PDNxU-sY|31mSRdCS+CL7?H7gWgC)sA0u)6x;;%vh4eLa^*
z^Aqs0^0gFK4$iLi2O)U?i@Hyv#l*)20ehT7%sTQsfFQI<Rv)f5w8b|YZN8!apUrI7
zKDB9C5Y~T1xpoGpA0&k~W*Ch6MNy1MeZ-W;87Lgu_pZDVU#8rb_7T|HpMl)ycVd24
zj_6vEd@~FQjw~<2qFP;{c_6Wt>E^Rx!WguL6s}gYmiO;+*6F;I=!zko02&r+XtQo`
z8%#v6XdLyNvz*YvVk4A+@llJ1clcu5?O4HDX%FO~iCOeA+Egy{m?tcs?RIBh3d624
z@y9QkyI@prHIunO?E)cbPXU+Fw>OzhJM1J<=kG;xdEMllz~1sanC&Rz;jE?i2BvzL
zKSS5=EQ8KBValu=F<&DbCaAb;mL2W!un)tQHwYcun)tR@1wcH#de-`0JhlVNXZ*7F
z*C%pB)JxFP#*8U!ymx;vLKltTb=B&JTtsGaZ~e(XFjR3tn8t4}+lVQp7^1Vr_rjYn
zv_x-_LQv-Ixw{UME*P7wU59;8o(wAzn@RkXYCZ-(5l)=R`_H;0EX!_03nVVe{a1Vr
zjMu!mM_t8_Z0!W-ORI?V!-$}5lBAAjaT9SXvweK`JNBW{#lT_U+!MGhHts9J_DZgL
zCPtoRpDT04IQ^9(I=&vGFe8?7#UvbQCWbZ_<~}Y_>I`y+JS>(a^)J`sR^ZW?2iEZQ
zWvtFG8Pl<fF`J=He$bAt6fL{;hAK>~AhYu}T12ShDlv<Hgiy|bdjvR7t1USM;3V|d
zBj>GJxDVEZw;-gm@J~SuI?GnawX4ZFB_R@kpjj=hCO_K6k#WCiQe=DKI4_b~cWtnU
zAWlwAkfbt0lGbVLfB&dL6dvrh%)X<S8wHLjX_W&z7ir<28kG==P85%C$ReZf2Brpk
z&QgaIlmsGJ@V7O9TyYlQ!yCyqA&uYBg+&)80e0zQ4GC}2?Y)tx%0iIUOtGzq@6m8#
zj?S%1{($#&f&gQ$qjE<2<4@@Jex42GXLa2;@8K`z113Ny)lsP|jW>V4UwYf#|EbPd
ze<MP4sQCt~`g9#YdN1|aS^H;Zu`^-4M5EiA`9ZF6_8^2ID`c~*k(CZi5!y4XsUHRF
z=>#2iP%0P-2`zZZx$YQ>rXAa6xc3lEx3L!oN)wXo6t<1UYcmP+^4w1-Uj0E_%ObG%
z?sz~@xSi>gMWk8A1-MTR_V6v)zRiR>Cg(aZl`We~1YUeKL1UV<;O7`L^pD}wEkXY{
zY&4aItODBD$C*8@$@NrI1IoR^Mi*k<3)PUY$vypg;wE)vmYFnr3;k&cMhQ5^nYQtx
z*wJ9P3z6Hq#W-Kyn2si&W1}hL(d7eRY~N+WY9#xE$IAnEg_C~J+^$M*SzDuu?KLJB
zH}w^jbX~kTfDEsFXqB1VpjjCok*#+^<3n8)L_U)SNzUy>!6LN!)}iQm*nph}MSaZ~
z9t*9+9yT=Fie{r5aIKypxIa8+@{LBcY!;&3FDreUZo%GBH$_J@aW*irW;+HR#@yav
zTXRnavw@@T@kaYNs6e(cp0^A#<FX$0X_l2E!;N>~Y`Ydl%|19v{h5sW!a}t#Bf<oX
zNURogP@c6ppjg)ZKj&U<8n;blG)Wr8p4z&c(73x?;fSk7Gni%FFH)L{g@YXAdeMls
z6R8ja&n^3nqWpB5SN)W&;>Zf~(6Z@Im(14=lP%NyqlE&#dDiDXNAZbNCydU}JNC-;
z3VZTUJC2Q+U;hsSK>WYDx0sCY9Ud#Q_SuJiE4Bo6;XCOgxDfXD&4_m6mwQ5|O34lr
z-cGz)Cs+a-X;&3>_IfV}_LQ^%xVyUlj0_!DyChOnmwb|}H=32Tm`PdbS)*&yDAP+m
z-jdrHTQll+u9xK?C&)o8?)3mg;ZS^NgIYq?90hRnNGyc#6Q)Qr<kJ8N&IIfJvz+~g
zBPc9Ub<`6^JD0?kf?Kqnx@?$K#yl1<4PdA`w~T5)Sr?Vdx`61vyUSLaJhc7MC#H0+
z3aejq98UBazz;jGj&YNW4Qt4=7*Kyx)ZnwTz=<A!f%eeXOj#O8Vxfiw>cFHOL>AW3
zt|I1WzlE3<w7Ztsiz!-cGYiy7uPfH~#jkVLX#z1odkunSQ7L#Jf7MPl$?vYaPl$SG
z^FGbFra>!e%B#~A>H`g3Do>j9o9T7YvUN9@YT_4U?}$JAalf&ucEX&^X6Gl4zjft|
zA9i3$pH0qTYc-$(7SMMEpn2!I9@enMd2{S1qCFLej_uR{K|sF0;Es&2AmdEv+C?P*
z1u)n4aK3T$S4y~In!wPkS9P)}S+BrRFscmnD>4&!<r^RXRE@6JG#Rt^hMLW%p@Ci!
zsVp5Nog8wvon+^eDa9^vsY6YCAf6|C+i8HPpJ_|rB$m!iY-=v)2w1P6P9d}g9cYGV
zJ7KU;<rqS>$Fq7Eccu~ZW-YuEYqp0zzC(woHvzM;-x!zlaP3dJ2tzX-^(@mGUM9q!
zQK*DL+#I8FvLK_kwPpQ$JE^e*wri0>;fFPYOjR5RXB(z&%1Gy~l&tGQ(3apAs_4;S
z!nP(nTt1n96pG_${ScfrQ3ix<z-b_hpq}y|16FeZx$;qK0zLwFCponc<wXMutl~zV
z^C<GD&=A1Z{)6|G6X5}II1{Hsd3l1W|1a%IX>dqTi!-|^d!LC2fA<fl)^>x#Z$^SG
zC*XH`-!J>`LnyB_ufNG14W-D({_Re68VqWbz%hOb`^vwWtf9J07aO_wT^{NxwT4V(
zu6w-H<l&}|Z>$N<h_u1WojYl6v9YzACH+(&IIJVV&fRz?xG+$uyzKznp}b$(?D=<6
zD3Xd{fdat88W>V!{|YsmH=bySIHJTRh>F;>eC1XTW=7{$a`S&}LVe`SXDG!6fClsP
zl^r{LY%lGID9R?tTg4_&F4H+&h*g~vO0(g49B3^NR&G<zskPhxp){P>R#n-SEw+#E
z)V*e&9np%(w%96!nV?MA!e=>p<jiw3-H;Lj*OcKo2-YDMm&IP^R<Qk*pyBnmI*Wmd
zEk!&GkI9Kv|97#0DOE8R)qXc!F7351>zKto86*0RyfseoyA8B@TBHKv1}HZpW}Yzy
zMfAu=B_Qjx&ZvXNOqwEd@}tA2BPyw(o=_m4j7RF6<!<{8{Y>i!l`aCF(=Hq`(92R&
z;RDbc92XBcZeE)fT?CT@)aJsx{`|=(dAN~g3eac)U4r{@N0{mVd4jE)JefsL9qvfr
z%Wp-KOle;4I)(BMaPtW`jBKb9h*0ckp$@tSbD7XQdOLva3}`%AjU8NIT;?tr^2qZ$
z2P2ee|HJ-vKuF|w#iMC`P)(o8Es3H{Yve|BU?48O)}Tf#d!{O~M{TMQ2#Q#-9gYVU
zMMJZgc{vF|fJ(k<n|L$G4Stmk71R&rHxf7bc>P?kB0KBnMh4rE%zIM%0uy_@Tgs*+
zCz3)}BIAqoU1D+N)#ME$4PWtB(U<Wj7=^=-H40vlS*Rtcf+lp-`i|W*t=0A468$uI
zjVn*#({Uh*@%UVnI4JpqZ0`pZqyX6cpR5tu8Osx{Y13gLU{|E9Lu|#qb9KtA9ElCr
zAueB#Y{Og>!uC09>CL5pXcOpSD(d0hIx1hR1gNR^|8--&<r!2^RvP0+gdkw&7q?xn
z0!`)mQWoL5WlP1t?D#)t$*EdwdrXq-VIT=k7*agwtK#_`#)mm$IC!<wj3gb&DfZs<
zPOA}g4VY2zCGG>2qL8A>y%LaUZDW{F*D6o+v#I0YUj~zMb%C`3a8|gaUUP4{Iuq!X
zPu;c7%V)NnAxjQS31b^o^1mc6-!vU7oKB=464hPeZXLNgCo$H~I=WY0WY5R(6oGD6
zBWi5Bcd-$bE#M@TS6V9f%0&lD`1vy1diYgIq3OZg_L>t$I!{>k$S|(??=lOYogxl#
zE=W_iS1@$fRfT*xueK=<RtySiu1f1Z0diZpYkW(jyav&Gt2@wC7d|$#uP>!SlnAZ?
z&&p9Nf{UY$?k;kZ?HlYyTcQwi@kEs?Sb1TiQ9W$Ik0LF3e{anHh7ZQx*{&m5fCcNs
zM(A>kjw^h*;fKH+5k2^6k<nXT5kaOeVu=MaC1`Iggj7cVT$cO|-jxonl}!?2=5fX+
z>=B06xMshDW1x_6P%>)|`u@ewtH+L*`Ff%bAOhlhP>1Slwtcln$(nj)kxv?kWr^!l
zA+MjgTgS+;_hS9edrf$&?K*^GI>a`(PBH4Cwv?_N=ZjP50|%t4w$whEmo`=zldeoF
z0FK$SkYxI&aU?@`DgD)RC)36t-;{umtEVy@#CFi!q2K~^YE90%g*v;(qCBj;0rmn>
za8rC_yZtq!s~Qxt%mS7RDtUu5kPeDmbwrWqfNp<oJIdg=U7Do#Zg@Z1nmNgyFld-J
z@#*Y9@C2mh5QZS~eZlFgI6%%Y&p_uU9ZS^AK!M|<&5Vp;=mgzjkp;#ozp>{q4|tvU
zmA$k_*t@kT<6E9wVMn?MV8c3c<?I)4Zvj#$c2x97JbUsBp>SemCph!!ov*6w4{nZ9
zgxHKA(ho^&7EbkL>Tq&qwX6FaR{KRABQ}oOkc&U-S6ZhdHYPkgav5x%z@Hp%70_nQ
ztj7`k35sjH{M}R}bqui6>G3f}aq{b<T`)a=Vuyb%5=5Ei^3<SH*&VaMleHsRuYrN$
za@l9!$kxU=@j|?DO5C)w@(t?Owp?)7WUNsamPvzO7?+#k-b+?k=qr)#?%jOci@J-<
zSor+XoRd&E_EF=#0-$u#`s5(PM(V1T4ITSM@^k2b66Siq&tI*z-I7`Q2C+ANJe)C;
zsjApx`X;Wx2qC-!M-e2#Npt)2q#vV^Q+txpI*GskzUXycabNC-tMU$m_T%b8I1<HS
zh+4Z<umrwTvxm~&9n2npG9^0NK=r&+9kMa^kuR*iA!1HoCXg&lAPnaM8v-EM8D|3i
zjB-$)e)a1IC;7`!4tJ=ApJtmVM%i1G|MMZoioz}nUR`NNLe`{Yhuu+Lfh+uEFI%Uy
z92D-{#+d~!RncT+Kv*jzl7PfVHQ3^BIxsV21a0jpP=sTPEs_EX)c?2bcLOdUx1OtP
z3KB%9+f<15lQNOtgP#ZXJ%RDwsfqH2<6twZ6$?GWxU5H^(G$)7oDy1t)nFr>oH>{=
zg$X8t=?{jrq~*i>5=80hE!5xob@DKWhSuulSUEfcSk*pO_pMT;d%i2|lI!9IH^7hH
zY2Eq;F^UEl7kG2po;^htez_pJ6VLAZ<9wNi&L2G{$F|AwLbO8qz`Dh##&Fm2@Hq#B
z3T&PJSb7dKuV!Fza{clVohdw-ye{7rN+#8{wkYdl`z7Jlv=pPNYR2Z04R0PBwg3Nr
z;cOc<(iYp_Y+AeV#1BydF=XfjhMlsZapuUCTK%Pp);$HxcbAQkX%qYhfc01}jx0C>
z<!c(H>*Xls!s<ddTY3s%rMI+)f4F%#p;2yuEo4s~U`k5w1}5`ED*y~rO_5ESt{{;M
znNZ~yPP~C(O55{%Lvv$>UgBWWA6e}OP#xoscct1nLgK`RleQnQF}SQ6n)0pvJ2y@_
z(LfS^an!V1r6N?MU1EpE58TsxK}Rj*&mAgRFo;U81|HQ5s>?t@4AaP@K738x92df+
zq;Fp)9X-$$7T}$(w2GVCT*W(G>(C;}WiQ$^a&~nCM_dx|_AJzBQt6erb3_@R;69Lg
zdLJh(08pVW{XkBVfr_P$=o+(W6u7y;W^srTbAG_TISvMZ1B6)n_zP=2SCXACqsq5+
zKniPd?Uq~P*Lys&J)$YPVB`=%Rate(`V^|`dnRliO1u_-?UNfMU2pk|N6>2zkg0Ij
ztl88d5#gd=rM8tuVeX!qHTQneiR-WZ0O;Ed!D}AU4AwqcT3wwV$Bz?&`UVEzes@5<
zrM#!v9JX=9v~_5;cG@kZ8R?TMcX6pEWIi3{(TSLxvhQ*`9uzW32^5~GyqxsMhN_j%
z6yDo}$|VKiYg^~c>lQsmQG?RA<lgfaN(mKqkDw)6LX<DFmkj>un%Pd^-z5#f!)mq$
zb@#{vV{-yl<d<qC_95-@Yn2657#5b_R9-cl1M$*Dwcza^;}j<#IA8?OCNO~Q+<mvf
zM_bCf&1C?|(JS6NNARgI%jA|AFV}PKCGM?G{KcQaz4ZE?wnV0_%-Y3rRc1PtxYykG
zwm#2B+ra9_yc|e2w#)8XEViQC1CU}mb53a+(QItRL9esk12$zve+L|cu;h1zARw>D
zcB!=ACkh|W5ei12E=R8y{wtn`FqmLDM*oO>qW`Ia0f_5cwe2hTIo$FsbBsxbbd{!i
zFm%;Rffo;gECi~9wD?)oRV$;s=?1|@MqhmK-rd{9Nv+Kt{zd5Y!cxhc2r}WBk2JTn
zRH+o*&0{(!&g3E$Hxtudigu`*ZZe+y6$bL;;KX-eJuXHi^CxgBd!}}M%9$8q=w8q@
zRTDVIc&U!Mb`B$nJ^i$O(7avYBi$HNWM>er|E9r)QKMiWFdenC!uVwVo*gc)k6Tp)
zS1*5kz|<CMk&7F_o!}aq`^<HU26ekYarYQXfY`a>ZJQ;o?kN4YDO&wtW~Wwv9I%jp
zTuaD%?_;(_*aHmBrYBWOC0yw8JE?7}sUL?PGjqM7$oWiLn1*<lFz6%1PLZR^p&z_U
zJLR-uEr%T0o?BV#N2xMV^%vgiPe(VHZb_$!+S<(d6J`e=7=%-9gp~p4RYVWpMs1+O
zGJ?qCmuvOWDIL=9B`;mG{iX&Y)}u2Bh0=aU{^S7s@_3se!1EzfTuC4zmpvxv%cw#8
zJED>~AM-o$M8?;Awq&Q{Sy*VY2lHcK%W(-pqLoRYLY+4Yj~il@b5rGy$e<cx^{A*{
z^({>0;~&Pp?dH^|!27R?h@6q;x8|z)OEzhE%;6ZPy3_O#SKB|}yfO8xLL`W216?=X
z<)n!H;HXMLZJCEgZ)#_efvOT?TPD{#{Wh^J(fMeRf%DvyN-5q*`;N8x#m)5s`nKm`
zn}usnsjw-(mO#ud3FjGL(4D(V3jbf^QWfYA-|4h07h2Q(xDAd9nw(V)v~hd~R+}pY
z6UM;ZIT);rYc#rj{vO;z^3M0RRg<UQw@4~0$Dt}hXIO?<dRfG~XHQ-qLv)p%Xk1+1
zekE}JA}g_J<l@le(dLzGeH2EGJB84M^69BIXUJUy1QVXu<;C0_v=&pAN8Kd6zA7Dz
zY7tX9`!8>xW}X7jE42UJOnAi{ai^M9^)Hqs8(S#Fz#RzOz)^|wsNyQerIv#I^NF2i
zvw)sZhl}pr<CXUL&c2}k>&src>BIUh+@g#1BFrar=?FC9-Iy!J=)jzCEbS6k56+eY
zg!_LmBY-AVvm0;MIPEQ#djM(3X0mj~vFxPwpzC5mSDsXhoY&LROIDSk>u#&95>AOg
z%!uz?tcE?uFR7ACoxEGG%HPe{LfNKkCF3ftz8Za))GABi5j4-lfJWA`vuaMJ3sNz}
zg?Gv&LN!X_W;;QVQ{|lUlcLqk>-`Jua*eOjW3QuJigJi1BV9zEg2(Ozt>Z}R66_F<
zV?~6Wm-F65`<Jj~rVT+rWnLZl!Sm_LG|CO?zL@!d{uJLN=8&z39v=^oj<ixEiy|fI
zWEB9_PWicKu_+ScHlu{pA6Lk){o{K#vZ&p`PtA%7ToQrkh!EDLRwoA#$^L9hgw;FU
zmfUeWU!&5OMZNvT|2b}U#3n`**y?U)0;j~<dL*TuizB`KUm!TsQ19135C~nAi}zwr
zFtQH$f2Ew>f1cBblrAU?Pbr^g7EFc%h|p(FBl(m71TYIV8bl15Zw}nY8y<A`A66P#
zS#>imhwEEZ5yb)>6)@=f+l-qk=f+LE2d7;^PL7)HC*_s6Bn|o%&|x<}kk#y!!%GKJ
z%&eMJG_U~Tt)kVMK`pv#%!_ri2e}2aBQ`OtE(gMcr*leE43b*<qW{%lB86Gt=j<)t
z+%~*<0;kfJQ`kB|Tz{`aJ4Ngks@GGy*%m?a%K1Vxu2k5TmfxkM1Z9aksk$CXy%<Wq
z8=_)~tBH}wr<6kLZFuCO*)+1^Pt#{5lldZ0r`-nFrd>Uyh)A}DIX*)2>0Q{@Vou%q
z+f-Ai><0;8dVjU`&0u#zJxm<<W4(5=+z?Ksn3Pcz#2|T`TIPEB-u;+UF+v(&rUJk?
zoLhP;!s)iu&G*&7#U+F)RwV>Ql4F<<ShQSxa<CsZV`A^%zFb4!SoU%=U=rC}{cu}|
zs$4l}FMnY;>u1}ou3pT)DbX-O$rn&DWo7t$a!5jzer=!;PW8QQi1X1>gtZzw9abqc
z&yMt-Osy+kS=Y5xi-0{OATov}e!0C)W`pQXYm~2l;LP?z!5#nJBPh>fb$=DzZqyrh
zuoIn1euP#R7Bcyx0Ye*I`)bf&)a98D=*i+|I#MjYy|Be4)HKO7Uzk8>PEy$?Dyj=?
zgy=pC!O4y;%keOWH1S}&CDzH2J9brwBPCno_sN<YGY%SNV#fA5l~=`e?S*iI2&d0l
zYe#uzHWE#jrdlknZ-+O6{>ep~>OYIJuQ_T*5iYkH0Al8N(Z+E|us7K7eBiu70hBYP
zLz5;N{eIw@a;yqlE-J4{%WEo~+gz1&i#!-h|8m3v?z!)WHa@K+!<t{Oh)h@O@aRd+
z+^Jmj26667&(#&>02VJLzh#0%q+eF0lq|T4aTCxbrXy;2vcZIoVF&hS+^iHo(Wvwu
z8SD-_L74hfOmRhTX!+|4TB}oP2cdRMFLi1U;_|E&jLu7EPa-O+FNZ3+UP+X~#3@6x
z^-;~<5P)XbGEKiOy|*gvKr3LP#T<rJeMcO<V8o~_wP+)p?QtY4u~1yFNkxEi#~m}n
z_;ELJ6z^-t-wf8NsG~s8(U4iNXn3_d?NZ@&gGherBQPxf*FF7bh1&pFB;quk4MSe*
zdwnaESjGRdp0k3uiN&q0*+Vhj{O>{pK`5(I-CqKE1rqC`z>_-TR~d)FY#@PMx@gI{
zZogrbBl44C=aPTnbd@D{nVdeAGS_X5E&|f&2=pH8hJ{6+o!-kYZA>9oL-e<5x1P*P
zSYFn}^*Bi><VT5Vh?%aHtR`MpqI>fs?muuF*ltFM*vM`dA2ofVs`x;f+kA}_>sTiU
zw5@aAxY35Q&!no#>B|Mo#JscOtF3EUA?{{INY$VlwUeRsJr^4_j6%cK%dQVNBurg9
z=?pg&4?0|-jy{e7!e7pHa8oQgo=`5?zrHPFNmSR$(+TFL-vQfbz!9_Jyk3w@S6bk8
zVC`KolWZU>veI=tbkzpxJJ9Z#kn8~9dLW`q|D}@wsZuxuc>Dbr%YQwO6^*H;P}`Sy
zymy&-g@oRqe&I$1OtT(xzl;=^-`%&z)>yqOG(YUemtGo;fO}V)jew&oCcTL0^{u3>
zD0~IH&y<_hFG(eZqJ@aoL#$EC+rH0JM_HU{S2D{r7;crr(_BuVJ>GJ$v(VZDt{;k{
z<3~6q?qcNaT;dM+6VD=|CuXus&BPamCyu;CONyQ8gMB$xz><b8Kbe{(tSOuWKcyk6
zJ<0EUB#wWnss&&?|JPiT{=W`)zmqbgTJuypO!n<__xnrfYRcb(Xv`aA>|Q6D)%@HH
zNX88ZqtqD&f|f+z0rbJ-Y6OLIe3GMyW+Nfch)}Gz))NMqnXcL<FrS53Ita9HD@Wk!
zs59uwxMqQ|L#2T)A4$Ve`{&K5g9$|ENg*PEOyLY{sCbSXHL6Nn$I_S;F7g_ov2zY;
z*K9L~_OcUQnwfTObP-dc{2W^N;DqZB8CZ09kkB(8&P-Wpfa8uWNpMg8A@m7K88Ek^
zrD5Q>L<>G>-`~S<JtWGGlfQlJ0UD|Ld4&A=I3?9<31$-y0#~fHuU$*}B3adPqPxaY
z-u5fVW;CR<J!F3}CnelmI7uOeYu&XQqFq+41EI|yLSi&?7D-&Z6?U$%0bmH#y9Ckc
z$tbqhq2PfFZYh7qXr^kD0Nfvuvf{}&PIEW`{IB};mIW(D0T)OzDe|UY)>4ZD`fY3D
z8=m!#Fsnbmql$5fq?9=>#-WBqc}#W2jqtH6wMDfY88}Wa@m=*@p-K~q`56x0iXlO|
zV4sT>0Z_1`p)HF+{gg&qU1F4x;yR}tS^gXhHGX1rCua^nV)9{?ObA%W{N+_D7aLz>
zPHge+0B!z|RvcWyNS%BTg|^`$vDmGwGD5Sc4CoqV{8d7JR(+`mcskDbaBGW2R^h%^
z$W+mW$*Pk~=J;Uwa2W(PXjwe6juI2R8$m&vHh4$l$&@z@x$hxE&{ivn(;_yPC8U3Y
zJ~le3L6DX5#|t+b1WPlp*?Yq<=Lc*IwN-dE#QS?Fq1~Vvt3ATM9e!+-E5q1yzOh+=
zbo!3f0!r%UGNW!h(D9xVlA`ypWmW4CKM!s^z3+sXk$x$C6ZELA8i&<*N4lab4QXVJ
zHtfxVy{gm2=EYx!Ehem`B){Nv<U;kQi7sY9xYN!(6PrYg;9!2~SgWbo7B#2cKDm$(
zLf)HlmEF@o3Kf0h**E9Mz+8fC$@%U11e!Rx$xS6div9`bC)<n`%5$+g_=h!QTu9o;
zL=qg?JMn&XbX3XwPARI84>OhAcp+=X5W|4p4dH`ra`=Sg-K;^6v6}NA&T?pG%C@F6
zgDk|D9!t`eXVdpqSJ<HbY<Wd%Sr86Rxwn6qAuF$*Cm<U$^G_cA^)PckD{BQz^setS
zo%fFXJ(y~r=~4t;TeDM4BIUqq#U~QYUFq1?LX&NQn94qnl6}G^M)Sb~b&yOIINU@I
zRPv^Z?L>lcs7BVQb5$eIuYQ6Y8;vs%r*1#QjkQEK4tUL$xCeHu&D`b-kX~Nk8LjYu
zW6Tt$e@S%L<r#{Ad3aRPT7utr@OeG9s47*hZAG@wthX{Y-ce+Eh*ncMt(;NW>VH91
zn{Lz2{KD6?gK_PUGlQ40`GH%%*{j#;x)E>GNnI>K9qVU(mn!lH7Db-AL~v&B==7|N
zG|!4aGr_b`(KRY5*0c~lyCQz4jznOIOUtRQ#LmZ*LipdCqu^#A^a7Ny=~#HFQH2}9
zk}i>4YD!w^ddgZDyA9fDYoU%RFL)ttdnenD<R&l$nowAVxAszl5Id0^W`GpLYeSGI
z)QOyKzcDf3782){3E4Crx;WkCwsaa*AHV)Civ0mqo-Gcg<qg7$Fk~D}a{%eL92Gf}
zaaKLSPD@QvurxjY^g4wAm4_B9*}I!1eobb=$%@-*j9YvL>im8Ax#WKV)YjsWg16{_
zjT%4GvqQXX7tNy4zN4{Rw*KL&*Dkbx0elwS1FdzN9b%ApQQr$?S-Kptal3=MTTUD%
zg<GZrl~aPxvEgAf9rR|zSU~`<oV52aVpq)(4cRw;3C??My(re{xcpx6E?AGS7TpA{
ztXap-Yam40g)be;$@Y~Isy=Ji|6~tjlmYl<kMClPUXxn2+FOo;NPQeS@=7*Mtn(>L
zE+}F)EG2;tVxGSo)2JMd*mp9hyF?het626TPrSm*Dwm}L3VM|9TWQguGh(0h&)#YH
zyx=p9SjItAo_Vj)Q?@O1D{%!Vc|pK<8+#KE4XbnaYt;SbV4FhV-Jy<zPc$KM_7QA>
z{m6IpD1cv%ZA6EJ)N;LPu=OytM>IRt`5WN?nc+F%h$-cRBIB{{Ee|yhw`Pe~5}z%&
zchjc_Bdqnyil}kdfh(NGxk;D~XUl{2uPPOeV^qt-La>=TG}d|rA=m184gE=U7u}KT
zLPbBN=3+vJe^u1y3c>%6Ov|JQLd;tyr!57i_Njg7W4gN@LLQ4<e!*LnP)ki%XwuR=
znfVYG4_v{US+M|xg8$T5lo-nUz`p?QShgY<{gkFC+&WzXJwO?APnK1p(45Txwwm=+
zzhI~y>W~HP|8hT%;r&xI56}k82%yc@*y2H!-yw!+4$#gdggi)2QP<W|unG<IB-s1H
zX1n$7gJ&@|ALoAI^vJ7akTO8{^jF~72sO14D1GQ7INAR|0wAN{`KAn4AG<}E=&v<X
z3|g-fA#ez0U}F8C*`o=Obww&XcabzxBXU`*%c-<_vB@T%oxbBq_0aVh5{t&B#Pu={
z8H$?#YIsY6`=Jiv7ylO{D#qe~)Y(b9M1Gx!LX{Scpw__Tmzci`_%#(Bt=W#CoFy9b
zdvK+PF~9tDl9~1<f~wN9DGbL14YW{-5l}cHusD$?ztWY`f+-9|&kw21sArV{)g7+7
z<|sm|NX-OhxaLaCDtff&1N6X&vXCK38-F%kgx7_zi(;#g23~!c*llxG9VvpIHt`4<
zMBJE>IjEtVDN(mS9!&WD?)ya3Bqr8_Xo0w;`&xoc?)<<SH{uQ`>MipyM7Xu3QohH>
z_1Fh@_qi|ltP+{ZQ)I~NS6gOz+Q+?xGIywgJZ;q3e@Fs>kC>XI&dD=)&{+7g-5m)^
zYf@oPyi#3Z9R%@#2gJwl=r5e_65Rf1gu5N~Ir%Q!=TDYTFZNpD2fJ^rT`1Ih)S_Fb
z{K4M6U=1|l_5-ZZ^<BI2nq6(Q%Iox5U%D~K(vP^$?`bVpTiDz>g;2gHIQ+!TEs;bp
zd6<D~aX$Tn9!D|``hYj}XE?nh0fnW)O1Zzyqu~Dp>Y0T!x|cH)&&1hFLbTaGTN+V}
zTU_Jd?34ToeiZDwELrQPNgM>GN;4=UcBo<6IWh~08W5(H)ch_o&HqdwTcPg4COlFi
z>kkEeD(Geazi>)v;%3SG1s1r4Mt+1J_ir8NXiL@<_&~a_cP%(n@Q-0z;VL;uVQ>}p
zfC`QD(>Ljmkpmrfu$Qx=15TAr7YjomUQn8ppOf%V$QSTyQwZDEIK;Q(WuNepT0$g~
z!@J+GPx^T5_pdx4O(e*NiMN-k=Y+p)!kp%BRpF48kB?WMB^9_5tFIz{<^3vwKJTPK
z=OoD(ykuZoADek1ej=ve0o?L1wy|caDW9O+0e@x-!M#p5#h!k&rM^p%qNVYbtlj_0
zM)eo<&<uGp)8GOEG?4g~OhAv$70kUY6vS|GK`fwtEi~+hR2H_cQ~Gym!j2k2kH>Ip
zh5z#D1BWdtOin*_nT?fG@RMnY@)g77zo@svKu3*|=P9SKQ=M*DAIqWP+X!_9eDX&r
zIp~F2*|Ds&7MK&Ack4;<YgR!U+z%jM$Y;K+iUggNx<KRSk)k2MmQ=RJ2dQG6{45nS
z(2MO1+d=~q?KUGJ?b-J*ux%@L;{PSa_`t914Q<T5Io<FxmPz3-Io}_M*J>~O;A~(b
z1USRPb<t+#HSye#LBVrHZ_VLFHnmI6{}l8J-!6$0<gya37{<M~M<#VY7#YSv3N9A&
zXwb#qpLi~S%Xc?KMji(JSApNWHo!$>zxf?Ko1ug{xk{7{u_@gH4DQ-(;48cittX`K
zWc3;^mg$BFAv5#or(DSK3X#>fXjD5^cVH8HWghppD`lyNxvi2g<I+UJa;)>BYlCVP
zyokgETd!=bYx=z3D=|t}jp1Z&+^hAG0+02NPed)qn$rkK&GKrO-Kpd8(`kvOkC?Z!
zfhbj7YgM0PG%d&CxX2gm_~l$Y%_L`HR(Hyyk!E%rBRO>d9{l7c_c7Kc81@gdUCKs9
zo{vt@<rXUz${#>*EPU614Mqmr?XC~meXBH)h^e!5<Hj`Ls51H;KCuZV@Vv)W%T>yT
z$mem99nR*Psm3Eh+WP3l2gI!0gxi9-zZ{=`U9Qs7-xJ~#o)<y(Is8uZ>S}SZQceCe
zt=$HH7iT&5X~fPTb$?b=TSk<tV#jJZz|G9Gc^h;~HIWzCJ`}ylrqLA_X*w@V+yHa%
zGCqRU|H_DBYOOGueP#6V{S-!j)LbyjTyf$2cg?C^s!V%N&x`hM@qY&=1}>R)jsRsB
zj}Or?-~e*{`HfOKo3|NMCGZoqaJ^)OYPV?|LMuL|Cx%1Fst$>yUS=uZ^fhNXOf7zp
z#3bMthR}Q^G_i}*6<plFlnlW+yGI_k3YENQO9vf<?Hv9<zGE-AW}!|sud^;PduG_g
z3oCO;=b_sERpmj@oKyU=y{``>2%G%1Gj&x`tH2gS6~>xrObp2U>DD4@1jl&S>SE(D
zCGcigEoFsnLGNKQ&Cbk6mVyM8DCW;G8x^2KnOh;Cl@l6Yj!DmzSb96&p0q+4&;Nga
z`||3>Lq~bYq$2iifWzC6<anFl<&+IO@X>Sr$&`I|5ff1WR%1o&*b({6l(~LgES8pj
zu{M>wDA;3=U6hcX4=fxhnr~VLJeu+9Z;~y?2ZVA42<eyNenNjVlFZ7LaJU=f5^Neo
zMMhcx`Muh)K2Ac%u<CbH)7>yai))Wt&gyOkNxGbD{pf&kloX?{$MVUQ{)wNn!+lIy
z(HEXU4s9+$cU?{`qEk%hbV8{MQkA30nq2e|-HoZ;TY<<W<$)GymVIp1gydnNaC6bQ
za2<SIAMFWy10@2uBz*bO+oYCB?>pD#zf!u}f4mlZy7#d&cI*S_?gZF@x_xx-!E0uk
zuq>e+@{taR2O1^g^ynRt6}-&q{2Z+$D&8v(*y1zNfKWF}KV8t*<P;z)$g<j>)yObu
zYgY57SV;bEcCYB_I(BYI3w-5GNT#~S-pq5XtEnVBW&{X|SnSUbU5^R8`uqEX)BSK&
zQ#cwsBoq}EInO?`-!(tv;K1~@8)`!s{cMtANQRW_6f3dK77WK`$=>{tBCn65ku_LT
zUUb4)-2NwCqcD(7eFs+Ap0tNg(M5Mr+$gMvwHBH_p(r8~HkFFUu(EBVsKCa1c78%M
z!MSo$XI*<?ch>F;g0g_mc0s)dll!0}P<j1jX5+e6UrqJis+d$poAD0MgZ<vhUe{<s
zIH%iRwx=%`2fkF^F9^xM`e29_t7MI-#2*Iq)-^{x9obu%yvRCRYUtt`k~5x#@(<{<
z%mm<=Uh+@+iax7V^;XR<yJ7RD_QnzYDc2oG{HK5fTdAQS@9M}=d6{4Yn`*pL;ZKN<
zs}ZRO!H{kcwa%OrPh(Z=x!I4e6}KxvX*0^$KY4EJt9>5ponB#YN#lx!$<j|#rqoPl
zNdoB4o;W*)g@x_x+*NjS&juJYWJM_zU$6{d>g0(egrr!c0ftGxziZa(_aVfVZT(YZ
zVtA&daT4;qXYoR9&d6#^Fc8;h{N6WCtD=T#mdIV{+;fPvoIDX}XTqN1QZR1Hg*wVS
zzQ>k^ZalOfjAEHP^_@9+Jr4o^eT50aUwVwrBZ=ev=!zugA|F*aPa2_pNXM%i;OsY|
zM@U<W-lY2EgdiQe3x0?rZXJ#dwaQs6KMO7qmHXSL@cgz(P38N|EZV}`;<GX8u#x-~
za}#*Pnv1UMA?q0wSHApwQ3%r=*v0fT4n{zuxMqfmpkevPkJqqxF_L(@V<(HniQquI
z<fmmIo3>W6T2YAe^914epVjFi%$00cT)(H0&smica3-GehS>qVL&mXNCPs1M+hur5
zyf(f$bAU*#oZ4KniTOySLRg~hj$#nPzLAL}H{VuG3$fN1OiNFi0Vjm#ftB%<1_TWi
z+H8gPrW%}p>a(z>@8Nkfn(DHF%+P$VN_9aFFb1AsUdV5+Zj!xK6q?~gYv+=3km=Zr
zFytudt;y%Sw=~#M9==9?3H~%4UGPNiUjj#C#<b>=T3p0oX>pD*F;-LUQ#*bV15CR`
z5nIqnY6}8Z8^8d37(e8eUZ0IA8N0Lu`yvK$k+8SmEupQwDnr{<by1s2yGQ0o2SFsP
z@{yP6|K2D!Mx3XeGQ|gyH>2@?*}e6$)Dmg`l`aC0B(1QqU}Lbkyo2k8+QPPO{3F5a
z7~sFBWXHWDvL)dsJ>T3`s>xr&RkzgxA;SWjKF`rg$v4e0Cu2yP9HTBo5S~T`8aN6S
zA8y^83RE(fBBpAJw=<9ia~Okt4guQ|)YP&8OLdAozQ*edW4KW+<X`!kh|hcciAy=N
zmVV0_jWWU|?B`?F^eoVcZk^9ErjdA}&!!qqlc@AGOvh!4aOgyZG+}Mv-c6}vBj!6m
zVgE;k$QY`r(i9tpZ}!hqeqAWY9Dozhe?5OEYJ8FxaWQ97(qJ+(Sy)<5D+l0{sesN^
z=M{E7Oc7XH1UPgAvB)?d+=@{n*p5^}UQ0v0V(Zx=G~6X9`@0DWq9ER{fI{V+jcAHq
zgwS)Y=`L1EHXjfY`4#Li%2LN^boWD(MdBx2z01Ar^Do`#mBtMwdXgSQp7#adogKc*
z8J3bIl|@#qYrEw0n@(r-1xP&~H(XlqE%xCp*?lPVv_Y0v1j`FcM|s9<gRQ}YAiazg
zlVHH{Y<sFMvnW=ph~L6XAnn^&YD7Xb)PJnJly1U<y9{I3-D#i!>Me;SA3AMQqaOUu
z0LhZMPMq`OiO|nUAU8!Dd6<0jS|N`|=5y`jC+F%k(haj7Hs7c}5+yQ&?YNPV(~esX
z)rY9ddeET}XT*#E<)E_*Zv({}jKlESm^u^_KV*L3nR~kduOC_Q94=5>Q)2Bt6*Y@?
zRWHtcYaoO%3*Tl`Px`f-Xt7(NO#7Q|%6U#SCcK|hF2o+Y_b^i}Cdig<Lit8kJ$GQH
z2Z_jOzL!xumpWg~=)pex62aZc<FWugH5nGw$xEbT<cjX_8BI2sQ~?xe?q5W9>3nC3
z2eE#+ShSaPTaBNS?C;_(^ce|QS<^Py);orsK@b6av+o0#ysoGa{auZd&5xwk7O*3B
znmgi@LNGC|d{yV)bIn6!WH;kIwvuyQ|33U#YG;PJpNxT}me#7~E1bAS%kURT5!f+$
zqX^tqT@CfTpenWLl^s9o3V8zfWuV6KC(mjmZby`aW8k2c;i@Lpx;~;v6Gmt&h!V)L
z`et9T_scAG9MiGN=6<nyCGpHca3+H#BX3*+xa!40z~Z^HJ~QB{@6l`Ec{o^t;(d`&
zSaVHRL*rN>*|yK8CDhaD9xgWn#NVaet?CxbG{WA~_If|C4cc{b0r}hNEFgNQOy|4Z
zk<@5f-ajxHQ5=`xB{S%kL<xg<nxv*jQgK><&+l|zXMo(veMZnpTqK3Mx&y#)e`313
zfgJe7FoRQ7ShO`LHqWDMGUZwc|A@fzW}#B9(>V;?2mFC7v;gd`if%C?S>Y51G>W;g
zr%0r*2J_e@TB9w@u#(W*_U8#;C=Ga@?CHvvShgGBvM7Am8dEYXFw0vAt^cFw^~R~R
zpH@Ia@7?lJrsksdX=fG%QRc(Ij}1{Q0WU<~db^hxmY8g$G*{V#Sz4^_U-LjhI|PkD
zU|k3ijZ~F<vw2~=9BR3P!Ei;JPY$n)tsy5NEMlT;2cxM$E@BP@JM$hM>o@K~7Y8bn
zR<qj=OE2*vivr#Y*Fd@PBfb**tGGw0W`uZ6`bDdE^GUo!Sb-8#J(_8={HX{c(?vb<
zjmB`Y&}`eUXsR@=+OnH+jL$W_YgP_Tu_e?qGRFs`ZISZWp|tqDCb=H^OIo9`g51*O
z7VwKc<A{a?>00~LG=$_sn?dg~j#lsbR3s!S?SWawfgD|tA}`QJrL7zwRDi49*-sfl
zQ%q+0M5HEoYTzYC#OI0D)D>@ua@?b=9DgtvWH=a}Sjvmx6h`qM!;6~7b>jozsFR-A
zCDjmeNC1~4vvI$MywGb_0-S0{M+l*<`}nY{h{XwKH#aQsG0JBUNk4_yOe7JG&&;B2
z56<O#vW)3`PEl8IiGGF6Tqe6CLO%^7N-3|TNHkVIVPkjEkXvE=vhPkE*G}c7&VU7p
z9m2%6Rz4;R&{K{U2dCGSYq1`_@+We1i#aqSdTPXDdfgFP!(A_jvKN8eUR{4Ng#|cZ
zU_?`%Wm`;kKEi$x<)0#1@D_)cCVX#g=O0bVo?$u~9sQ~t8vbd>kHGGnxTFFq!ISk$
zOFf>Vvo3EHLY{rm7*VfTDTSt&>oVfNgi>k`MBx(8N~yW&W(x*pb=|_qNQG((t^u!_
z<<o(WdQoC7uhkpFSWoBJz~3Sgz25CQ&pA@KV%o!&bw(7E<{m>H*|Yc1yOv=DZ=cs1
z!!w(4MF~>+&q&unw(x%tRC82I2in|(De&M>G%cqJFc3-A7gbmV&wP>(yR?i}sRyH>
zos3Ybi(*PmH;JjuU<{%&EqBoY=m_{YBEpm|?v5&G=J4pxz8hn;@3G?1ubpl&9^wsH
z&0z>!U~$xs>8e83MmV?LNkBcSEjDLZjp$|mtdBlSpPc!0YjUq{O6zT5uhF=bv0uqG
zBO4DSTv!7BJKG~mwL3^Rn$wrde^fj{i$#1YltbmlK$I~x&NZx%8otSnLTH0A5DMx&
z{Rn9L*=q{A`jv>=Q+YiOn(8|N9RefVX>YxQrK9?<Z?8#NzKuN<u|XpmN6MtvbZ*34
zcZdnvX^^=5S{L^lRHHqMEBvb<EqBEg!u@?tjZGHf`7+abx&S-kZQxZTXAnLrcJJW5
zF3e=bD9UN<u^Sw~YD7w?;=Mj@?Sz`h+BKSjAua&UyC+b-|Jn}soB|D#I9DBuZ66|}
z@_km*$`!4`r=KJ7#3&7Cb470_gzzOJn6L5xZv*FLXK)K-v98Cfh3m1}hLO644}^H+
zEQr2C90$oHtitPSzBfpRAo&#F-P^%jgK$_<)dXF~I9AB?wyFs}11k7!q;$FPdLbVy
zJ|8SoH7^(vTo_3t&Ij^`pg&Q~!90de8MtgrPGW#`G6`?y`u(`JUjP{&#z00)Os^ce
zJ~rpf@NRG{EoQZrWj2qS-3F_ph)Fa&jve&izAc1%lgsy<-2-6kSq37;eepb)y`@?1
zprC!l>CNu>VzFWM$S*I*)BnQG&CU}D^>I80mi*|n#~s)srj)lqkVYnDoE+oC_@4Zj
zk&5rUt5f5NiZ*L+B+6D|S`E#*&pG;SL3?8?wJ=;#YC?1p#Gi@}<onj}wI9iQg~6i?
z{*m)7W|hyh#GzrDPOX5Zbj3J!S*vZWw`1|~X;f3rCMOn{bfbh{Tw;~M3Y|PtO%%Ed
zNge-d<~-4YvPo_MwLOy@g?bIyS8#2l(V%_Y#kdPhKyur$sI(_NPAk8(E3Iccl03w#
z5X{=a=axd736=YwnNsj-(qd<&uX;iDbJ#C#a1TS_WAxR{_oNE|pGSLD-6vSI+?1IZ
z-^>qPkxr?MiYmQ3b2n!fyZ9oo_3;Hvupg2lm~LO6BJs5b2bFNF<$oZ^3{`1+4dtX4
zq9^p)Ce5oGc5g^j>R+{VcpM3oUQ}raB*lkDhDh8X;qh%`rxP<m`P0+PlXpMhW$^(7
zD~*XBkU^&`B@2ciR3j9T0OdKO%bt^%)T<2{@mu&NgLA-Y2=qUI4cV}=+t<`T7^Cvv
zpw3&`_+t@3)pEDdT5n3t2JNBgxh-i%NQccRH^V$ouL_;CFz=S+P<t~A($~J=2VB9L
zi0=j^>h5mCLY&`yApab@-Rvh=x*+Uj0d+7X_ZPP1;&{$LZXY`fV!pO@K^<bDoF5Y2
zRf5Gt!cwiX6SC>byUds;?d5_>h<b@-s|d3SE+i{Yc-YA<k4;+y3;2<g-|eiE&4}Oa
zH=RD4fb)J{E!hOrv47aaV?cVJN^J#GfpM6?dbp!U#~N&YTPhL3F2c~BZ-Q0}V2);O
zWdq_GTT^E2%AunrE9rLJ0Rl9+@f9mW%{Kcumv)zgoV%uYFoP!P71G*^nS7G+n_vfr
z!H2HG8DkKHluI71?ZcKGCq1F_+~*|dLPPM(eEb6;TIl_KQsH=Dqg0AUz_;C0R{26T
zF1jGtSQDI|Iv9@u|AIjpUyS^_d#>x{VoV@<a~!1R%l}Mi(5XugkE~cK#l)=<KR9_p
zZeNsA?exlWy#4+-J$B~=B&3R`(;mDrF<Q)}8m(yc7J--*(mGYBa<*m#ngR&T#CIIg
zm{b4nYe|h!7JQP*T0?{L2nv|I-jp1>+VY!AeDS<9igHv2G*A66Q|n6+r|k_0j$RCL
z+0;xQBl2R%0FD1FPdp(?DsGGT0xsR4i?Vv8*h9fYGE<nl#r+TvP7BJ;51wAsk#1*~
zp<c{=Eem?m+gsI~VqVu<axH%NKZ8abw&L**X|vh-C@Ez~+E46RcdIPq>vl`ZM-n89
z@NUbKIv8v^g*%<jY~f<;y5+;XM`*Z>7l<>W(9}}xHbry6uD)uS1j%cD%sa=wAZ^qt
zCxrF>)hwwx#0(s|%<0mju4LX&GscE(L_Y7|^rid8XFcOEw~c99dkph%pT6<$t1OP=
z@DiqrMdyJ|c&GGp{14^e^KmqVa<ORd!2UbJTKd{a%s{A>HP<ZAW|&o6p$3!Bd$e(7
z`AM#6Ss5Ng%Ugvj5ge_cLp}4XiWjSKQV;c<3y~P_wDl}Q<zZE<Y4MCb5D)yY*KoB{
z+K|V^C7ah$59ntNWIlaXYUP;L@=DtE$PyQ$>-k!Qux;QXN`$8zsthq(W#!e5${Uj0
zayJF*r;OlSPD^jwd~6@x^iUV&jih?OtGZ|kQkKz9k9$usjnMN3Om839F6+w6w){!I
z-ZX)5{d>qf04czh^x`gU`8x3Ea+y=Xd)rMhHKOyo=@F2$q3X39SF}8(2fLENV(*sg
zOsz4QsDuDG!l;=4o}fg!X2FQ-bVL43yf1U#*02v?Ya?C*WTV(P*P0vljp)^d>?`kb
zKGPZybf(J3%BJ3Vy9|VS%BmP%=)5REj!LCNO=CcZeGawooI_>P8pB$MWF<0Dcwm?5
zd{8svK-pK$f-9oVSq&An==|6P*iS?tZ9UcbC8(Q}KvuLsdCE(i1B=~2j0JFh#lU0n
zD3il+0K@s=$oS}6A>|rl?0L2#x`Ps{e0AUynSP22Snw(4J-<ma5Mm+HK;Lm8J((<^
z_|jYlQx9hPq)Yr}&mWCfi75eOxcHEA=U=!WruAZtXB|!nIGP8{!TF1P=*GO4rp4Jf
zv6;mxjJho@sW$Zq`ieh^^Dqa@;$*V2QSEK8=DG{5Wt9si8?wLAr~>B6v;pS0h^w@D
zczyFYI$G6!5t($L^4v0l*wwHi?oz)eAWvh5icb#Tl6eP@rcv(1%y+O)`N4}-GX8&&
zpt(_756wM3zx+GvM+r)J=|Xcp-ceP`X_*)52VHWV_RQF+Y{Pp{c%DS(8VCsj=yhGs
z7deX5WJd`&NjDU$@aNE^96VL~^$E(m{SyonD)+SC=Cz$45%zi<6t`6ki#MJ68!OrD
zem50o@DcMQzM=xhZq$m;;?1UC;}MZ!Xj|1NnA$Kz1&v8^2=~N&^|FRqku0p6RexHn
z@QaXtNpLb>Qq7d6X<X+V8R%5f77amHphhqa?^b41tz3ilIvBesq;tRyIGZcI42i$<
zo7=(%rNnv$E@{Rs4@21l3m>%R;p0Vsj4tBnrV>LSJ=gZ;*P!~<9)%|%L*jd?s?AKO
zlz3K^(@ZB0?wVVmY_h*TxWUH-SJ<4uVoQ8$FIA?5aP1O>{PVpb#pKWOt4_G@L1^3o
zP?kk*Vlf_rVBYKJS3u00u2+fWo!T;&<IUw~bUKWhp7U&C*l;i1+RkmFr26fViizuQ
z>Gl!f{z@%2!7%!G_tSF%4P66+t~|_vgB5-2nUb5seObqQfY@j?B4f)6)kC8;1HO@h
zKQ4#Ge5#LS($_11wxOHvr3JL3j9qzugxkngcBQ!-E$6qUME5Go=7xNKJ?L#<@PPBU
zQiogczn+zh$<21Ks!fJR`-0eRP}2&FlC6N-gX*heC}wqgzoy<zw`gCeAB0u;AE0Sp
zJM?_X3<ivh^$gxNDY{JsRL0dL;HC+vs^^!)``sU<Ql6`HWDQvh{$i<qjV0^pz&`E+
ztDn4lgPt3Hm`S|#5f^Rt|Gf(t<o~ZP0iT&$*5?mu3P#-!KfbA3nkI~C7xvy_L|iIh
z^A;-v#Nt(R1_^P$=mCr;g8)%1CHLlyyg8ePP*%E<u0;#f8w^SYS7?=!snNO-3x=Hn
zsknwsRCLBu(wT>vvGAPk1z+{Jb+qFyZPIuyTaQ^7&>sV`|3Q*gEjcDSUt%4{ei>1t
z_r>;#7yfvIn7a(!*CbMgZ0Lp#z|hlF(Z~{X>t6?qIv%Wn<(Fs`deHlb<RvGpG8hl$
z%o@W^aP9K6f=q*ChUnz$?TRbW{z90vi3SU+_J*Ar%Ir9Bb-j)a11=`Q1b)u5A<3{}
z^_9_dUiLy1$#Z%FT@||lWD&WmZIZ;doYYvGDFA}gqD=|}mcVICTqGFzIuy@}FgT6a
zs7I#I*IgCZ=1P04K0bDJPaZ7%HnGv*Ch>5dh=Jjbs4+4GnzRKXdjnmCG_jZx&yMRp
z^*p;<LK?y34;B5l?E70c{q^n~=CVNn&8FX^p9JS{`L0=&01W2K5<iWr@=2(|xnZ%_
zGXe*)bIay^Q7thG4i_Q3DgNFbnA((cdh<~Hz4*AgHGJ73(<m4Z!=93_*I6cxeYbbX
zmGOWb><)N+I7Gl+wk&Y-HI<iPnj_noOhLPnGe7Y%C)t<W4oE>H_13Q6x!Hu#ZXp_+
zLvL;QJP!d*FPD@&#v|+Nuk}&(7(qM0<ClzQ?gQDvGU2RCDT~T}m;T+58*nphBsW4=
z9TJ7qX6&c|*b!(9*F&gHetjq(+cZ8sA|#X%18%zJsHRnG!V{vp7(hS{mLNe978A5_
z=lT4B-JR?=uBo==Dv0O|$wW>jUuJ2bKZ>`IMF$hpP4ZV~M#sehN^6BB!+<4d4?M<#
zlKCm;$1|9_OQ0)PeA4<DL^$FB-?Ck~=Woy(E>r6agd5MM4y*;8Rj(SNY6rNmH!sTe
z8hnU7Yp=(fS>6+(!q`GXc(^s`HA*dqTI%}K&=Pe1rg9M`x=EqhaE&Dbq|J`g{<ypz
zs<XAdUhZ|W8bJ2ERCVpb3rj{fhK&!|t%oztY#fo%&3NynL`Or>&3mgte~%;yI(3o0
z>>oH<i=DtNlaO~)<O9FIT0hQ{Rp%Ei@o4sNKy=1a7w;5{SC!jbyPP}tZLjI5Ix<)z
z+XQmZz|aJF{TvI>EZQq6S=!3GTRJ25&DNu9k=_;0KIU<NF!-68#F<R9{_}V^86U-R
zg|&Iz-REsiCzw2o_VN**a^S7ZQD60*(I^gW615j^>>6SonXs?4Bu;Ol?=$`@4l!Hv
zc(Evg5d`1}NrO*{!G?rC_+Z4dGw~UFgzo3*X9Fu%v%LCFr*=$}d`|=3`;+9|oR;1Y
zuqb#UjhUnkhw5ahuhCCkZq$s=P5Ze(B+DEMrbHu_JL&0&E^vHeb~8K+qxSCmq$Hx@
z5BQAtZbhs!(56A+Wn(GrzRbna6T7-ZSZ`M%rrPKvTpom=WFgIKWt-x2p#@!T2`*P@
zu9J0}Wiz~-j`2QQk_{vhUAs~L$4VMB3tHJvcwW!jBwa)?V*H;$dr62D67Auxy(JfB
zgpd@1+p~Jk^8!Nu&a3(c(Mc>YufTQeE$CGN$tgD0N@mYoiauHj6Hc-C&sZI}I+FOU
z-=+>qflb;iu0dK_vj&p$B?<bmkB20n6V?}zALJ(5Qef!JT(0OCf99$6coM-{izItd
z$l*aPBB}%miAr)1w3~!}3*+iJiAM0*yo?dIQDiEUVx2jm)>YhD)#v8=7|$#WiMtYN
z9$gq|jbURK@<#L!sPV)IAW4o4OoBC$L1Vpk7z=_YPJqx^h)Dfy^l3MYy!s71CJH2~
z7C|zxjlcxznFPW2Kt~t`@z87j!IKwAzLQnKBJ!S4%Q?>if1xiBY;v;JDmX3ng#)P3
zLlLb}tL~D6r6PKySJ3Oi+#3~ih5jP4Z=VZI_{itw2;9Rg0t;;XR}l~fOi&v7=(HTT
z8g1%kB<;O`aRYit)caQ+qLALAnH4};?8PBaGX|HZ3YKT<BBtQUbEGwV$wcS4gG@Qc
z>;PoQ5ejdABRe;btB=lbR(Wr3^sx6YU<P3iiy(*hkPH(vF<D;+8I0!%m`nrVtiTCI
z@%9uXJjC?i7pINGJbQQ%xQonbKhuC?Ncn#EB35BrzsGbR^Q7h&rztr1)MJ@1v1J$M
z0w^wpJav5sJk)>uKV^@yij17S$~k*v@2u=S4u|91ap#1rtg^{oMTs&qvuE}uTV#fk
zl|s?)qkO;h|NehJkB2_*dwpK>`5K>>(HVSzr}1{}2Ks3P|C^{MlnGrAEG}}S)b=XG
z#%R3D;Z^E={dKj}t7U_YC?wNjRqP~d>B$mf+Q5ZK#+Hm`!;33wn^JTU@!u7LBkN4_
z^ZTK!lI|6|x(#Lm<*V1CGT*S%`deI^BzC;|sJQ2cb<nWw_(6c(QRxvYD!gJ*Ls?tw
z@k#6aC#nL|j4tQTACbV{9KN{zh=}m<O<mAij7zrp@wHAe3-0^ZH9u2v$XZ$`a$&Qk
z?MICcb%u9$)~L9mN8YwARpFz<AMX;Yv#l(4<yi?`xX5yM97}97*}(ptyng{)ye^Rw
zt?%M+WO>(W#~ts1zShyct4etO)2DJePfUHD3@fOrn}F+P>9c!3(!QsV@f$LTSN&Mb
zrr4)uEz?-~nw70W)7`HYN+&{VaPbXFs_P;o*PXiA$Yq&luw}^LP}_Yzd3~EoKq^}&
zP2=)nw)#UkQ{$Cth*xBT{O5>wmWO7&(@}H&0RiKEq|T2u?v@>XJ9NFX@l|l(84F#W
z!;u~?+!<Gh-=^S;qFvvQrCXr4oNEInO1k8jI`&t?&*VX1j&X7Z$h`0e*`tm*v&a5<
z9K0(W$EkETIs}wa98ZQp^$*=O)wi1EcfuIbALXoE=Hf~Dp`Nz9q5c@gTanG;rdIy_
z=)1ObfPnU4qqOkHpJG~FCHE3b+!=!6QYvVz!#pp3u-NnvzOG;NF;32dk%X2A_5!aI
zr9rB%FnMSuGPBn)FLJa^ISD;MOLTKv-n){?B6w4ElDx^IeemmLl1P1_zMiTsrjT1I
zKE2I1lsn3Y=qWL@bJ+(JNlTv>JDhKA<(|cu%IPlrR)Gd&+SgJfq00B=SocPPz7-
zQR&WZC3oR!h)(OeMQ1s`^7ULo_Cvn>h-=@jUotaQ!5NY(_cB#~?(AH<oE^elQbX;8
z?`N*t*NtDtuA%Z34Xm&V)zB=jT<*pssTc#*{4hU*Ax#&#l%4}P*xvvx+E<s0Qr~+w
z<1*up_el7ptb+0N3-4QC&Pf#<iZ2fGiJ(}1Opft!#P0p@r|-TeQs?`~yv;3R)%Y56
zghMB&qL_8B>Fb2=Cy~%G{vmfq?iqgl+=X7^2YLgqaU80c8beDOaAOoUPiE!%XWhVv
z%_WAlk5)fRWqhPgOPW5on}(X(vKu#B<5ti(U;ef)7T45^maEFX$ZR}4Rm2?lj03))
zKrFgcV9k3PM=O;#rq!oRaU+|AyN@&X+GkNStEweG$7b=@HOy?1PB=*lDd-Q$#q7i1
zU(gSXn;L|8cq+tw67avEA`sesLLN)>>BW^Ezd3~siMh3(>J3}ZyKAuIrC%^{^~oEl
zE%^J4_0A+u1Gnaj?r8?|Y3P%7A_wwnOko2IJ6-h(hz~n!F{@?$beB7$i0v~&2F*N%
zR=zY)EOZ~dG=0|6<JTjDuM8hutTam(5x75}PO!|dHoI|a5wi6f5-~Rr?$p%pUu=c*
zYNt~U&Tz6cGA{{yc_0)oOFMO1s*d@vHb810!>@rQ&IMnni+r5NoF^C*DZ;zJ@Z=fI
z!|{sU`QgoHTG@BEdsWHrMD3{*q)9U8kQYVadi&j11zGq$&u;!EQ(tOM!1)SU<y>$w
zbRbn5D+H>nYxcIsTSRLVO8bpex0xt8n%sTAmGqUK#L*`x;^`}Er%-;5rDq-Ze!DrE
zgfgdJF`Q%Z*Tfh$7J8ms2`@sJQ0#6L5oCV<PQwtojPJe7Bvg-wP=|ak7PiY1SEc&l
zqS2OhIoj`I&q9RG$EW@JPNd(gZon$tphJ8{b*>U5jI)PsurH$ZP25!^InI+-uDPV%
zL`X@>CduQW`wvO8cb((o4!-33mf@6527kbd-j!s@q8+K4jq;!3WBz$%?rFTl%(V-0
zHR7ai%%SV1I)|0Pvsc<*#W`#Gq4wi#ruA)aw-S?I=4Qcp?yhdRiBO{Ac43na^KVsc
zqyN~kh)XwZS<(2|!if5n<D2yUN>0!oV^yKG^G6UggyqnwgP8C0%LmAv{<StDLr;?a
z)vVok(D>t5xPDQOrYjgWBlYqKtFl=N94#WM<hR|ebsccah<F7dH7)+WDS6fRZ=}ac
zh$KDX@uXdFA|fN@7S&l3>Bd{Tum53b$Wr*H4PTRyg^H3(0LZ;_K@Kb0y_%BTbp6`R
z3h!o0KG!2{Z`vV?&X@QP5&i6|0gP$+oUpG$Ck2j*&zE}My69#d*XHJR!9|`2U*KyX
z_kzEV&UqI<o$omyVZc8|?BC<N*_-vs8LF6FA*nUAK`nQXcSKY^M<r7&+WWPUn9@gA
zJeq;arX<?!=iVdD;R_|iNu9f7zO)72HwU<IjEsCneCr=NWPYf*j%s(txVO=5Iw~dP
z#5%8h9|Z42HkVsGeFrO$U5e_Pmja7+M=z62n_Doyph&%LR5$^AU$ue1@rBnd|ILAA
zS*IY0w*-|}Nk$3GPi-?k!Pw|LTlF)^S-r!Od?d-{AF(H=Impelm=TD!jAk8=%!fa0
zyB$s>aVU~RVX7Xn5ze)!#ghEBEnMlAR{J_@-I|diW&9m)M%!ok3YHgmURLH$mKc-B
z#AePLd9<@i**Vv!dVIM{a;o|A)0PTt&TdfYhN$1zO>vUzi?UX<;Q^vKMc*%L8gz1K
zuTl28eva3Cc1cq5Qgc$P$WZFti4`nEHIAtWL68+4k(WL=zxd9mYD{64jtCESukiT>
zxlS-m5+z@GY^-wo6q2+<fJ8<`B@|t!<@%gtlva$A`$foX)1$ll86U<j$@1uZ<lgWi
zuZRlss@f5>Ya8tnqk4Ivn_)$O08i_V#CK8)-k58q$+BkVkxSLvH+QM>8~QBGzfsn{
zY0vFX$bw`%;-8p2dPaEdP1KHBF6};p#LX%_{Z|Xj0Xf^3d`L&TajE)PSZuoawFB`I
z{GPpO$g5QvRPO!|XummqNv+tzFf)Y5d<iU}9ZNwHx~LdoYvWU%+_>$N$1uoN0x=Fq
zt)iE!h;cEz>V8c)k-Oz%##c|Kf^2-MHJuG5jvy0wxG)VRCgnl|rE|AVLy*!>wT91Z
zvao&C&rFat8U5|g@dT@D0$8m^%SHYVCXL<)I^u-m9l`KB<E)L~*2ke4(cqE`l{=o~
z0j=;SFV}3$OGu^#x315cOw5}w>~iibEfl}+Gn)9hmE1nZ>m(u$?g=opAABR<7LA0^
zr6xhtf&aoaluC+h44gvTk$UENqX~9vSR+R@zGt0a8JZH_GGrR>4jP@!faC=o%JQyy
zch8XNOymStNz$&~@7A3&y52O*L@MB!`C^B7A{XcAb;M>vWq9&<A@^mcPRE(nx0W~V
z6RUAJF20FhMo-0F-c!1#X!&lu!Qv{tSrbd7<(hcdRTjrF-j$=<H7d%3E!R@Ba5S*}
z<hkNkE?X;JB#U)lod~s2BK(L^N-9M;IpNEa+`yx^w#G7=l7`)CtL}9Wg2z8+@}ldb
zB0VX8_If+vef|S~MKv5LI-;12o=nM0*@WvAkmgr)UtAnwDGzoc4+bBJK7q`96|nz8
zXKLJPe>AXMHTn&D*g$rSYk0vVp?XgWm$x)c!qjwjkn$>D!<D1;+AQWzO7Ci9H%O&6
zwK7!Y9=nI_XL^;H(WqW*Uq=6s1>e~YVB5Xa<F7=W^Ig1HI=R$tbE2+&SywwHob`HJ
zX8c0gn;#1!I(?<91@G_0(v49ZD<wYPDDlP?BEk=fL!V~Y<2h}-3cMjUyW4adsdBC5
z4#Rw7^yzro9cGK^>(rF0vX?woS68A&1cDCH_l@2V6$M~X6Cdq2{rsO6CgI$5D68s5
z#J7x5A(id#&|vT5>a9!?W)hnPe$1&sZ$!!{-?-TnZAme(ezTLzCowKGDQ>+;l!o5{
z#+UxSJE8`=FG<$)^tffaxXlq|CoQ<5(l~-qaIvwg)2EZaFS7Iagi=H_7~+u1h?=jK
z%1<0>%D2}--lOdI<YRk$QPM1hdG+QgtrcfZ3}<~0Y;?Z6w$;Gxy^xtu{Qx%2+~n~=
zOMt`|cZO3N`nZf~H<RhWr+vlgj!iIbW(Bs^z=p(1#-aME3wpWrG`_J?!8?0JJaW$m
zkQ0rCB<MZh{lVjAY>A9V&Ao;O=dz_=J#G((>X1p(g4y@q*=Xiee-la7H;Xj>q_OwZ
zXPJFCiD9~KmPDG#6?>|jbbZmTLB&1GKcqfw+M8I!PxnQs9JFSTBW$T<XwAP`pHAO}
zf{~i#DyLV<lc?>$JtPr1;g&K1@h3Wuz{fj+7|pVnyu!F-NoUqL>L?IgIJYgfgsb6{
za4O&AGPo8)7AzGB?Tqr^z8SCVlzOmcwahE+>gy}J$|eDF;;n<=daHc%zx)1E+P<T>
z>vZ%wHqbF@(@)LXK82LZ)&Zih=O?T+Dwiywz?70$vpMo@^d{y-cKh?MBnR|3tv8nX
zZ+Rk&j~Ry5^!Acl0z18d9~8z&_r8q}AJASeH2gNWM5ZV#;7idFp{ldJ<eqMOakS=v
zPCMg{lZZ2pM(kds<78^{W7(_thm-tDVcDUcZ|gN!?(7Nq^EdTRuH0WtskX3Kr<xIN
z=VEwZ0pYnl691zSM%OrP+-)H}6`>>&$aMRV+}ZAeQv+xJ>Uw4tr5o3geSOXFhLNIR
z?QuU>(bWPdOI6)=ooOVWg%2(G{(Il?XrgRS>ield@?RaI%O?d>Y=VEv<Jc`7*i9?9
zqECMg<w!PK<h@H`j5aP3Y;}EaWBD|8a`1xM0nc6iY`y1ZtLP@LgAi==%`#iVNG(2@
zZwUHW*@E6t=Ds0f7URxUI{!u4w^`zlhnF=2Pvk<I;_`NcD1s@7-K*~BH;E4(d3|*I
zJa~P{vuRAhLYJDecwZ~JRKu{^`{_}}UdmT@98P+gTqe(gANtq;8-wxJq|X&_fxqHJ
z$LXL*ci$!7UEu4ooX2pf9lJxbG+ou+_FGD{dS3TlzSbv(^w}R<*)QCrs0x4Hs7{_I
zJWawBaZx<@5c5e>Pnjp@@M)NA$%tdcJ+*lBm9Zd@?xfFKG2eIhx=7s0{Qa9=MqJo~
z&e16_bUV%tr$Q$8b?9N6--6vNZs_TYRZh(+rt+H6w_X<*UXZ_%Z7m;!b?ACdxpI*2
zmK^<3rd=UX*rVNy+^iUqAg#@dF{%`b_+%K9?sey7d0FM0?}T%3SYqY4lWuZWlKP#v
zCZ0FC&)p{Zl;p%2jGngE`(52p#NByBA*Q>^-hG!OBk_Ixrz;)eqP%&cQ$vn?94_U&
z)LEw^Ld}Tu5`(S*(zkf%n)g3`u1bGc+;n^@*J#38USU-%Ygc>8ftsfeOO2aF`r#v~
zM(g-ol2`(M1I$f3yMTiOVS2JOPGxV?FSXj4TTx<nuk*m3?9=ICuFzA52hId+!Z^~k
z2{}1^uPT1lu4&tHK*<Oc2e-VAgyfaf+xyYoY%886rLy;FN~Lc-n+@l5xIZv^!=<{h
z`dvau=bY5q8&Yf2cIdYLXD)-sg;DZsw^ctVBvOTec#lY@Ww;+vKG8<K*!C7BTEhS4
z6pNcvXL9?|b>Xb=d*fsYAw5DDDd)%{Tl)~3?{d4xRVrN|p2YOs8}8QiOB%E)3&kLd
z3|9Ww@s{mQKh-eyfn9!e6KMgl#aj`&qxq+1D??D_W&{~idX#{y3B%3!Cb6b>S&mH?
zq{ZAg-l)Tn-l6sfUmCTvEXD-KPa8g67ioldT~JZuG0fp5%`f-z)i&l)*fSh%@iPya
zc2Ki6Chs9JFY9HvVwtCJLnLfIA{C{zu90Zk{M}OTHjDeAa9IVl7(WwjyUF|NV!tj)
zw+$wN-Uz;sV&}xEtAnuDsM=e~RUe=Mjs)+N#0K|bsd;7U>DPHtqxKEc5js94L<#fS
z3>#`654Si|`5?z)WYF}^d%JoJ`Q@ErFDjgK))Q6Yz*gZU3%JH5&w7=VZYx<T_sx1g
zYbP<R%5iS9vof*}35#dG#P}Yosqq1;#e0R?=I%4Ft9{97R|h<`D@~)jjoU-j=U{z0
zgtGqVl+VJl6T>;~m}^2Cn1mZJR~WKEc27Z(^+|ZN1Z^v*@ki(i!Ar|XSre(T8U5C$
zQ;F^|dh|0S8xq$HdHix%BHw2v{0vmD;mZh?jY;e)rW6xwfJGS5Fy*~m^_@{Jx?ma@
zZ{$0ST^$u8DV<MB?|REa!D>d#xiy`_iQl@Fq9M5^yDMFaAd2`H61Oyz?7c3i>Empe
zqoZaa7#b9BxET^Zk5m9Xn_CdXTDIqraVyC#y<=tWd5Cr*u2UwvYwO4;W0)0ns5NC$
ztjnY++}NLIsI0F)k*u)jcCZ|>-0N<$k>B6gcZ*@092@*TPfqxQ_Nj&MH`lRsz7xEI
zB*a)0L((xd79Hf%;t%deY&214Ja9$^dSzA*ugjKtkKI!;)CnQ(&6TI+cv%+nMVHPk
zcPY0y0l9LVXE<@oJS@YjS#WnDr|cxEOIj3#n~kzNd=V(QtLF@T@_e1_TG#Hrc-Yd9
zv~PFfOxZM^8S-87f0qVZA2`W)7$s43-F;B!Ne?U~#U=@-`uX4sYMD=gk0)fW+^%!J
z^~$Q6kYoUnawXsrOcSoi%EfMuE&=C?^Ierj1^3Jprzi8zjk;O~)_>@$qrY5M+fhg~
z<h`3jHkj@j_5Oh#-<FNcjZaqZKSWbH1^zr>s*Fr>nYFn7{idh#!v_(BU0(<kzrO2$
zc-i4@+OS!r_gr(!KElpkQSGz1Oi82nh34%Gqc(mI((HR_1I+pgt!l6LFPHSCorVOB
zWHKKwm`V-kKsrGU6IUQKQH}wnQz96piYsQsCUsKx4!L$uw0B=+*p>bC@G$-Q`C|P;
zj_)~&r)jN22G+nkp^JHi>V>;@&ZYR2O0L%>C|%2_)kWMd!S;6)G&OP+US}moMpW<T
zc);oM-maN;s2cecJZ964#T}`s@)S>_*d>76`d;qG`2hS9L+0Eg+hy0V*Tv-LNZb`n
z2#2hMZRaY7cBT}GQ?o8yQXgQ6s+r!>2S2ouoeCq*aFisFmRIT3^9^Y3)Ju@6WH0ht
ze_YZ<huNQ%iKL^9wl2A57Rwrv>lFh!HRxPluBR&L&uBZ1P1wh8-KcZ81z8_HDC7wB
z8oJo6dgV|X?Y?e7L}K-lTaCJ0f=xbB;@dLYds^=SMc;?qP6e5snN0$$1vC)_d>$Dj
z!OiQmNUMMwJg>=C6zY867vA0RX`L1Kwes~Bw-WV0F6ega_eTIfu^#T<XoL*84YpD?
zNE#&ld~QBPP2;uU=A0yhttI;=+LMg|uV}tsL=C(i!bdHVHiukKcd;R1@5xCUF>R-=
zXO2^&Yz&+@nBm(Ks}8ZByO=E5vgw?MB&<2AaTj~NfzBgh9(o&P#o$gzgSRQoBlw0g
ztJ~_L{`1MSKH3qSc8t*Slh9-Aw<4P)jal8D33Il2DMuAuRW9A96Jhtii9{Mc!u=$Z
z0p55!Cv&@EZ(Z(zihvV89povF_xHUqMd`V`#?S?>ZV~zKLuR6FGL9wHCwM0bt#`vE
z^Wsh#QsvSs3ry}<i}&GC8Q_OGB(w!zi6|vWuLV(vYc67D!e5zKZPJhvDk_s~6xVBj
z`i@>gW!E{QZQ!_%D+=WXB#^U=tS{yy2?*k64(9wR_lJ9pc=HsvN0l>A+G8R73!{}E
zdxi4^=z7ROE>VS#WG2idk>+qW(FGGfWJK;7ZI{f?I}Q%is%<+}<C;5_d3N)S$nc47
zQaR~~7S`iz`+K%B^E-yexzP$_b1=d}5xH$OlBQJ(8Sx+@O{rQF{$AIt5A)4l3<xu^
z=N%*M>V6}u2@+vZy^m@d)qeV?=Z^PHG#Eu`QayWDuOuUUk%hsvromE_wWn9WBQPmd
zd!l{yg3~7+%r~-oyq>D5p2_2-MrrN@d*%7w!}tW#59yRsR4XULI?KaZ3BGi1y}7@y
z%OqcJ(r-c-gV8i7VO-#}e(mXctI9dM6@-6ni`x1GpAFoQyk4nG;<kAsg*)C1AB1X9
zM0u<CsOj<?#K9d1YksZ(e>7e{Y}5XSupHgQ#cpn+2lN$J{T$d&9zB-)OsgYKIi=ZF
z$TXQwV17)V`ufusVMaFMZwkD%9C7@?kyP7LkFT9tCRHCL%!_V<RdPVI<h&XNrW8Ie
z3?i+6bTnr4#kphZqAC5zw~pn0ytln1yHmDpc0aNAfnR}lz0fju1bv+j(c`-&S6F#4
z9la7<AWw(nwT*+pjR<^POMXUONi)J}sc<GXj@#qm@z%i=&)MS?we7BLaV-1uPfOP%
zbjItd9M}Y_AqwA9xzNq?aBg_@au&T!GM~uTQ9Cp*7<Baiq-UesEHK9=aCdu;w$=xc
z+7wMn&xfBiZXL6gUPXm-yQJoYy|8sYlB)3aXSd}GY>`&=%hjL8WKy>U=X0M6Hl`$g
zF29A+R?mDt30^%(eUM04!qO_tjWgzGaLnJQ*L>-6YN<(#`sLjr<1&JK<|c4Oigu=)
z{f7<-1g3kwk$$9O`g&{b9x996ZBpd^I&$yHU%9)csOb3U4Q{5U-5pI64(EN@5?>IL
zhh)zQDc*iUs^KQ>BikfRICz*wol$`+c{h&Tj^RX(@XLbGrL_oDldf+FEQeHJOKtRp
z%9VS|Uy+;5dJ1vjrLEcH9ep}QZAJ5L75j<y`huoEc*fs65`4m)#U&%N-D>&jiKc=*
zYe~hYwaGi4h9^MJUvC~HIK@~<BH<x3u=yfiDht8psHI_Cr*F~ZqocQz&DH1**u$(X
z64QD|7n5$k>2V}6=pGX1o1^8SHOr1)b^@Q8@3QochW@yBxuo;el<kKL0yn9Yu}n1O
zA_UHt6)m<?+2&%7jy;GZu55#fjJ+yF-AXovcNg;XsoYCzWbE87cPcaGrotcZdgDcf
z^z7eOFQn^^{K`y_GOdl*Ja>p2c$jz;MviNQZ0Q@VV%6fnqY|ji-H&&5Dz(KWz6c}I
z&ZxL0sP}}~jj>{AAJ^YLf2D<Qn)sQ1)y1!#@}=G19Lo#{q+{tW2Gu0&Z>$k=>d#e4
zpQuFq_@oeu@;B);7$_NB7T?rJ*a=pMGS_}7BkX$(7MC4VnBT6kYgMINhPr1x)Zw>f
zEYsh+ctBFpn%;HC9%>YQ;3se7sblvPjf5nNODRtGg^t|0t}fTg_Rh7FG{?RKFCaIb
zEmgsw?0#AXdKWbMa4s$7<y*5yH3bu#D~2G296ysdOzZUvaRH|S>DgWfN}a=<Ag1~7
zW2|fRT#?mVr<V$Zw`QBQwpXTbuCYaH5et`Lx%F70e3&cTwmd1>&GN_aB&re^Y)2ZK
z+YXlB3Msy39}c~bZF6TzX;lAkaws8hgs8k=k;V|1bFcr)`nU{{{DEW>8^`OwNi8yd
zCuA1ukcHKzfz4CLdz6r+9b}Qz%g>J=m}y;kbeN>F6gBPnyvkSdg#sO;oKZ_N=IPLW
z^cw7z>ib&x$tH5M;C91%Sncf#%avIUZAH<UbTLLsZyy?mZiI49pHAafGxY{^31vJY
zMx4e?LFT*0p9JR<8FXo|K6{+M$L4M>P`jd%#`d*V^~76jyS(A1bb^!!E)94{>bNdj
z2hYLEXeL6!Z(8>)ew;{99v$`DgQd$2vpN<p={=`Rs9tH+O^@V{xk)qDyQmRte@VUc
zt)WN6#A9Y@PDsFXg{!E4mfJXdzv%PRn^<OqOTx=hRXRrOgU;&HLL6q-cZJz?*hdEB
zm8hHT=6u&LwJ<j*_B;@v^ZMF)vYs-9AFMKf^+aTiHhM{-K0B__Sdi^=7)+QNd%?JG
zjE{`!%vBy_wSCoa(XX%abtnnn9TdlTPvwnt*6ddg@-ij{HPi5<EPT8^v`H7*uu|KR
zE2cq867l5{y4;t`rCb8{xj##6kv93(^Rkh&=a6Y{e2|}(p<}@-xAv&r4kpr1I29#Y
zHPTw#pZa!fNv3E*zJ}BaP`c;qjKo(YoSJ<q!%4qn(?vH=r$YQfs5{e_@uE-6hl_Ei
z?%fI9SD#EqiT%LvM<?*D#6k3i-c_sYE7ji9d)D+=)yq|(3l(ipd~vRbt^U)PRAP1Q
zZp(bM;XG$7V`xH0O~b0aSP9Z&G4+k%*81HCWD+#D`N>CbgXyv@Ck>@%ro|(#YT!UK
zrO1Y_c#JvTh*nu@)}8bH!YdhY)C#4C70EE<-LUN9>=<y^^jy=_A+1OdwV`i$^HpC$
zf{%yfBzNaB5$*zUPI_?Ra_j<mQP$dbHFS^drnW_DrKYv>3g>RV!41f9E;K54KTY-c
z{?cZ%)oGdM>oK7%l>5?6U(F?DQ@Pje1(FOe%u26Y7hiT9cv<>#(qrVxq1MvN7({5H
z&f}mY&wx+Ujg$9xi4ON~a#S+W_hSo=MSkF}X^ZMiuFC(!Q={Q#zGW|<W1@ch;^Sk*
zpaUJek3Ss{dbDC0!q+rED|{S#Xzr18iK7jrPLnwBNTww1&V>P5PQ#}ZKfpnwalT8p
z7@U`&2FcmOr<y+{Y8;$TW+-VItul+pWJ@*gN;NsbO!Ndq73#3o`>Un76$uCXDin7;
z7TWY{uY|&^6fVuE<769<y;Zxscf`MXyUBbG9fc58xN?mAsBPx1Mo@0Qz{5*lyuP^T
zlu4||b9es%g$eby&kxZai~Dp@flRz&sP#9(<<)qjpfy4k|CE~y7t$WZ+f6m8$?2Y~
zr(;U9@I<Hj92TC{E?I<k&L}x}UM{>%b1UdbWB4$uBT4t^fsdK}g|<m8%bzY$S7ai+
zyBzwP!+%zcKu!6tZeGt4j-QzT<L~*Lj3q7)a2>&N8AVn%Gfrw^+3M3MU(_C}Wk6VC
z4_l0{aDSM#RY+!3LPWFwC|KsIaD!MGsC9%tse44wz=C%HB*BR=Sqj$c8N++6Ztyc_
zF<tmR)D6`%4>lf)RWW6O=5mCa>@qCJW>Jfe>K58}6Q^7~W=L5|wBqb{9Uu_CBq3i4
z<NfJEW^7`^Yzk%a{a%CjWsgF?b8yrTYg$rO2ES3c@YPN1g*!n6qkHubN{n|_Vq1?M
zOUvE$Y7kGPR1J7L9hl^cC2Oi8z1sfq^tLYv7GJhU8UpzCTWp_*(xP5ZR6&`8jI*D0
zyp>ujm)aJ+lwx82;ih%qF8<=QD^Ze7ZkpK~<y?r}wR?B%BZxQlue?$ZWR{x=OUOn%
zDye&WV=KPx(lEIe+YBBhKX*so{5--58o3X=3-cugep!PJ>u&<LI^Nt=HZ?rxu1oXK
zPuR35GB2T1Up1DzsOyXr$_R?7)1&2Lk=;!g*S%DkqUynFMhG`AQcYx+U?OsmRK1V-
zNM62G?9?&2cX>BG8u&#L=+PirG52^tjx?F)9?hKWu`R47IM)jk$i8etd9$mkZG~X^
z`DeWNf!1MGH)R^-zJb}gp!zX?=P%Kc-en9gxZuGvC!beM<io%6zsqF5vm?A`{>4BU
z%p&GbaW|MVTl=Y~jrjK~VHX!lr&AxuJ%7HaNN9u%8|RAY^tpYASPoaDmW2$;NZ4v+
zh97Wl`F=`ZGWN`jB=b|<mQgryLpK`?(WlSU?%cY2I@VQfS)@A2`;GisaY@YKH`Pm)
zX<bdJ5MR9Zg+0)ed*q6i#*H-t^}u(5Y!eNLNmJsF!WnNkr>J5ff&vup6dwl*KGzt$
zl5w&0YJAzd29Ds-{5;yO{pfoVcJxN^*`}-?S=PTc`^J2v6Dp~l<qNzcQ$dVtl|5k;
zUfE;tL`(M6@)CCa2t{tVciH+xiPYw_^0?xpq<`UR+a39lYScygh7oh3AFLSTYbGlJ
ztnS%QA!-p$@q{%*%rEY^)pv~0ZIU+IT!_se!snNcPHMxxTMYOD<LrvBV-Z)<i_KTp
zR&-Lxz0o(Ej@tKa+L9vdd1P~&5Jy><x<92z;l~su{KrV>o|})>U`1gpM|4~2qwJts
ze|M+@*D?L;oW0u#v*MF{`+K!TZj`~J$qj9;Nw<1BUUhI&KBBK4?4wJ7{s@bRU9d2c
z(MXyKa<~xRe~Yww-o2hyMx>TiQqW@htHMAHxfHC)qsK()11~!fSO@==REq^p6@!Ok
zxdCa;#l*a8nM8AAvIbR7qfS~(UqtN1G>W(gqEsisEb=c{2ck0dU)dB4>640_{=o0_
zH|h_y_LTarcZqQ9L+Ks|kMm%3@^@rM4^vv+G9z8}QHiEr_DMsc`c;KjEVjw?fnr@a
zuQk|x`*!Gy3LM5hJ|W)!ia_$r-ASjo?Iatn?My}=EG^l@|8lbiG9^2(pvxwxquW7j
zu!!!q?MbwX$7Cb6P|M=I^`sdd{4xW)W7!mZycOlt?TlcGXD{Ex&@p=&KbQ{_5^IZ#
z9J_cur?=(9XlM>Hc-dU_mO~{8WpW_fM@u*Eb!K)#3e-0qpZi>MHFw|5wA`?2dvfV>
zAI>we+L$p8FL+xsX6n7ffcvcx!&J;#K(-eh>Yiy%m00L%hM~4diDwb3!~#+Dp=tLD
zJ71v{l7G89UJP7jarl(ZNBlwM^UB+|yJuSG5jPlrCNO)+Pu<Y>7+18ZivHH<m|R5W
z(hSNcpfu~UEd0Lv(P8OD{mF0;Mk9#d;#8E5VDxsEsTC{7;|PM5`OhQ-?b46P7#)~+
zDCumL#X+`v4yl{(e-x75^bwUYG14pueEU{c@a%2PGnMvLT^vx{?P=|oyVV5(rYb9)
za(m^x_sAiwlvh@khri;D;OmN}=q`}WIvid2@FV7`ZEa7d@1zRLHwV(2@>_|7SjGdM
zCbjSOG;D8Pf@g5P>;K5c;W2)T65vUM;1BAYOgERtN4<Vz>t%$~8uBgpb#F>ot>fZq
zpuM@gErYM1{@gHOvISe37R&V9cY;w_WwPE~QBr!+Xf?lyzSd`351FoVd8vibsDT3)
z$q#WBDON)|mR^q+tL&b<NU_cfl=|q1^x|1sZWv8ypY(e0x>I0>MB6da-mZpSU%@wH
zaw}5=MV^J(Z(lMZ598fDjaS48UMG#cnqs?qRU$f(vH3oISHR#i%t`tf1+(9a=P#w9
zSpDn+VwD%=C)VhosoB>yZ468H9<2i0-`3c(R$uo<PJ0N!wZXp8eQX5Pp*M7bOlEGL
z?ypY;WgJc3&cDq`qv*}#bW|op5juvC`93^SWnFGU?l@K4=ipTEN(PxiGA{R3HlYpj
zR`OxF<*4w(_4?PmZZggD0&9eYNv}!k?P!Yqc|$E0W?e{K4i%Yq3ge!dO9iwAzc0m7
zS=~G^>tUQ%PST=}rAAkAxyQQ?Dv)X@b7t!0-P5ppi2g*kJ-=kz*1Zt<?ji%ub|SwZ
zKf5h)L0q7qyOc(Z=!u@ptAjZ_iAeaAk7(S7Z|_u3B0Vbh?!f}J@8A8A=1V#pXxu&+
zWPVZY;f5s}gVr|DQGURP3}<%kX}wm~)Cjlu39Z&H2G<AP%--xDPit0e`LmwpnV5vH
zO`k#M#{GKw*aGi+Ac^nc{W`+1SK`FPWZz#Zj+ni<$L+QGNW>~ccE3a&tMk!7^~aA$
zld@0xr1n8nJ_y8g;-X5MS$gIrM<318HfDyUx`1m@sZ#HXAKN$`XT!Rcc@$;F+-Bu6
z>|c6Cjc)PuzF4!Tr99oxn69dG<&D9Vk-2{0jfs^yEeuvVbznVyXNjxq4_QAR@aQuL
zrX8{R{z-qJ5Ph?f%((fP@o7OE_M^>0wh`1WvX_@6vNCm<PxK}U3~7(?K*OM}XD5`c
zGYLc(Bmg=)+`0t<8+g0HKwuRV(gOt6$J%=QdU-~K08}^{<DuvTMFad`P3ZYqP(%a-
zHip}KIAK5#A>fuC30N6vhq8ww9YKFo{zD=GE5RKcU}zZ94u%1lpQ++UqCD(j4j`~S
z$^&X=2RpOv00%x1z$Y4N=K)6{L0~5=(h-Wrx+0)h;JzjX0mV3hz=&U05-`dUg@n0)
z!00ope>B>nVV*!du5ct4^B3974vIe0>2rSlbs;&kkAPt?fa&v-GvEN=m%}e_&fEb!
zg}Gy)zzL3YfFt1^-hc~$$S)!Z*b%_h1BmT6iyIW}0f!>&fp|e+4A#{Z`a2RgG|C=p
z2L#~;SaL)|&pdR;0-7;^^9UFT7~=-D186WWI|S7A{GPq-nR%?MEewr;I|9DjqYwxn
z6<{|Q`Zr`ySKx?&BJD|jYeT~woz9Fv0jv=)hkuUeWFUPoG#mp!=7u=K#tsGe2cYcj
z<^%(@1FGOCK+`$QJ}@*21O^}mfj#a5Cl4ny3^+TWuz&{+@UsXoa4+B=0B%4U_S-ZP
zekNdtLZFa<CYUSy*YP)E0Ml=j&M`X+6NdQ>5ESqg_|-@S5LgieRsw;Qf!YIsRY71i
z5LlfAd>aJT04e|k)&haGL0}ya_znoH3j*r_B@0wA2y6%f8v!*50-J!qrXa8x2y6?K
zkqZpS?imcW!1=sv&guVRVh4xY!O?bD0Cg<V9)Qyhg$6?Sm7U+h07kYzOaT5s1VHRS
z1_9Jaz(6X2Y60K`(ga8YAOJAA0ir-n0=NK>0V+XYPZ0Pn2n-+v6pqhtzhM7(g(mss
zCxHB~#6dBDHy&_=Jz(bF7E#8YB?6B0fB}UAnEF#UwouHOKL`}i0_PO~)XZ-x41s{V
zVc>ul_E1Mh7~s(F>#w5v9k2re26zRe-NOlmJ*y%B%=5Nz#KIB3O7pj)I?4d%t}sUc
z2{hCmZg=Je)D0+Vuk(!l>+F9If86@5>6i0>59a~fVi5=!0245me)TrY_4oB00}sF&
z%oU1u0cz>IZ~yh|TmuR~KnaHM09-?$d|*gN;I*ke9O%ej!T;;e@6NJE-35^P-8^UA
zfcP~#?6Byw;RNKz`>d30(I^)f()O&=FtGDc^hcu|%FX+iZD0bS?Hzy;|6TKEjv-*K
zXY=H&;t(juGn|0w@u$PkFh}^ARTzM@E7b1SJhBI}3IhhfSxcVJETG*m*k9cUc<`(3
z{=9_RVLgBdU4gm+x*Q5rgd0$azgzb#tl!G*VD^AB03<;B{*eXTbp<L63j__Q_}AZm
zf<d0;5)EkhCFTT00EPa``Z*aWrGNVM3`U?|&j*?OFQ-Vr@_(0={CNuH{~#y-ds53Q
z15N&~hsx(ll+TqY|E)y%kFfeVpZYlh>VNa8|KU^i0PxU0H?DIoq;oE$^S6-BKTLmw
zU9kucxElf}1fBC~VR){=@LYr8-x>`6NSdDCH*i9s0ce1Ra|Ig876E{EmLyLg!_adU
z=-(=U=fkgV2fFH?l7;?;2@2#0cru<R@~rpIl)%oF!2VVOF#RJ8Klc}Y&UZfBez!XO
zA3i{j2k<0Co*PG<3!%=1P=6~&{loN+Ffd8ro^X4>8Pxfs0(&lxJ(tJ+Esy;p=Y4)3
zm^Wu2dH;LIl0-&^1grpzv)^loHr&+~i_wQ7HE-&{9D!OtBP+o$c4)Yp2MP^>h@34Q
zAVBDRp*dSZ{@G_#AfR9S&ENO`-go37fA1(>9;*5tz!My(N%^zQ#10CAoGAv&3;y41
z{Gxv|3jLpqzy|e4pYZ?5CcrQ9i}CC#BlA~SXJMX)^(R)*|BWd2zYzuh5YM-~J5Wbp
z1p{lq?avl52n1lz16cyGqp-jV_ABszR>Z$Ff)&q}8~}H2EvSbZ0vK^{Tac#^KSY3E
z;0DOa!@~_D2?i$GKlk`iXh$9r@SR^v4=_*u7IU+A0G&OJ0W<%o09bz+0Cqm5vrKR+
zNeT)GiU>d;A`n3#K@mYd0YNSS0WO}uknuvp97qH}5OETLe?LH?A|gT}AP3ODGzfsz
z*#`)DP7@UpIjf!j&;*5mVE;=KmH7W?!Xf~o|KSrB18VQTGzp;I{vR6fLiE3O1q6hI
z|4W02NC^CwCLkavApReH5D{_l|Jns)1^+86Dj*>)@LyR`3Bmuy0TF|U{+AEHO6)Iv
zz_S>RfT4eV$)FGSfdSY9`xgp$zWuHuusYHK1%jNF&9BbGcmNNRUuhB*5)%?8;owl#
HQ6c#sOe0Nn
literal 0
HcmV?d00001
diff --git a/5_dpnn/results/gpu_16/slurm-25143894.out b/5_dpnn/results/gpu_16/slurm-25143894.out
new file mode 100644
index 0000000..af26cfa
--- /dev/null
+++ b/5_dpnn/results/gpu_16/slurm-25143894.out
@@ -0,0 +1,1270 @@
+GpuFreq=control_disabled
+GpuFreq=control_disabled
+Rank, world size, device count: 8, 16, 8
+Rank, world size, device count: 13, 16, 8
+Rank, world size, device count: 14, 16, 8
+Rank, world size, device count: 12, 16, 8
+Rank, world size, device count: 11, 16, 8
+Rank, world size, device count: 15, 16, 8
+Rank, world size, device count: 9, 16, 8
+Rank, world size, device count: 10, 16, 8
+Rank, world size, device count: 3, 16, 8
+Rank, world size, device count: 4, 16, 8
+Rank, world size, device count: 5, 16, 8
+Rank, world size, device count: 1, 16, 8
+Rank, world size, device count: 7, 16, 8
+Rank, world size, device count: 0, 16, 8
+Distributed package available...[OK]
+NCCL backend available...[OK]
+Rank, world size, device count: 6, 16, 8
+Rank, world size, device count: 2, 16, 8
+Process group initialized successfully...[OK]
+nccl backend used...[OK]
+Process group initialized successfully...[OK]
+nccl backend used...[OK]
+Process group initialized successfully...[OK]
+nccl backend used...[OK]
+Process group initialized successfully...[OK]
+nccl backend used...[OK]
+Process group initialized successfully...[OK]
+nccl backend used...[OK]
+Process group initialized successfully...[OK]
+nccl backend used...[OK]
+Process group initialized successfully...[OK]
+nccl backend used...[OK]
+Process group initialized successfully...[OK]
+nccl backend used...[OK]
+Process group initialized successfully...[OK]
+nccl backend used...[OK]
+Process group initialized successfully...[OK]
+nccl backend used...[OK]
+Process group initialized successfully...[OK]
+nccl backend used...[OK]
+Process group initialized successfully...[OK]
+nccl backend used...[OK]
+Process group initialized successfully...[OK]
+nccl backend used...[OK]
+Process group initialized successfully...[OK]
+nccl backend used...[OK]
+Process group initialized successfully...[OK]
+nccl backend used...[OK]
+Process group initialized successfully...[OK]
+nccl backend used...[OK]
+Files already downloaded and verified
+Epoch: 001/100 | Batch 0000/0010 | Averaged Loss: 2.3024
+Epoch: 001/100 | Batch 0001/0010 | Averaged Loss: 2.3027
+Epoch: 001/100 | Batch 0002/0010 | Averaged Loss: 2.3028
+Epoch: 001/100 | Batch 0003/0010 | Averaged Loss: 2.3028
+Epoch: 001/100 | Batch 0004/0010 | Averaged Loss: 2.3024
+Epoch: 001/100 | Batch 0005/0010 | Averaged Loss: 2.3025
+Epoch: 001/100 | Batch 0006/0010 | Averaged Loss: 2.3026
+Epoch: 001/100 | Batch 0007/0010 | Averaged Loss: 2.3025
+Epoch: 001/100 | Batch 0008/0010 | Averaged Loss: 2.3026
+Epoch: 001/100 | Batch 0009/0010 | Averaged Loss: 2.3022
+Epoch: 001/100 | Train: 14.76% | Validation: 14.99%
+Time elapsed: 0.07424663007259369 min
+Epoch: 002/100 | Batch 0000/0010 | Averaged Loss: 2.3024
+Epoch: 002/100 | Batch 0001/0010 | Averaged Loss: 2.3022
+Epoch: 002/100 | Batch 0002/0010 | Averaged Loss: 2.3023
+Epoch: 002/100 | Batch 0003/0010 | Averaged Loss: 2.3021
+Epoch: 002/100 | Batch 0004/0010 | Averaged Loss: 2.3019
+Epoch: 002/100 | Batch 0005/0010 | Averaged Loss: 2.3020
+Epoch: 002/100 | Batch 0006/0010 | Averaged Loss: 2.3020
+Epoch: 002/100 | Batch 0007/0010 | Averaged Loss: 2.3019
+Epoch: 002/100 | Batch 0008/0010 | Averaged Loss: 2.3018
+Epoch: 002/100 | Batch 0009/0010 | Averaged Loss: 2.3016
+Epoch: 002/100 | Train: 10.94% | Validation: 10.57%
+Time elapsed: 0.11137555539608002 min
+Epoch: 003/100 | Batch 0000/0010 | Averaged Loss: 2.3014
+Epoch: 003/100 | Batch 0001/0010 | Averaged Loss: 2.3016
+Epoch: 003/100 | Batch 0002/0010 | Averaged Loss: 2.3011
+Epoch: 003/100 | Batch 0003/0010 | Averaged Loss: 2.3011
+Epoch: 003/100 | Batch 0004/0010 | Averaged Loss: 2.3009
+Epoch: 003/100 | Batch 0005/0010 | Averaged Loss: 2.3006
+Epoch: 003/100 | Batch 0006/0010 | Averaged Loss: 2.3007
+Epoch: 003/100 | Batch 0007/0010 | Averaged Loss: 2.3003
+Epoch: 003/100 | Batch 0008/0010 | Averaged Loss: 2.2999
+Epoch: 003/100 | Batch 0009/0010 | Averaged Loss: 2.2996
+Epoch: 003/100 | Train: 10.43% | Validation: 10.13%
+Time elapsed: 0.14836958050727844 min
+Epoch: 004/100 | Batch 0000/0010 | Averaged Loss: 2.2999
+Epoch: 004/100 | Batch 0001/0010 | Averaged Loss: 2.2984
+Epoch: 004/100 | Batch 0002/0010 | Averaged Loss: 2.2980
+Epoch: 004/100 | Batch 0003/0010 | Averaged Loss: 2.2974
+Epoch: 004/100 | Batch 0004/0010 | Averaged Loss: 2.2955
+Epoch: 004/100 | Batch 0005/0010 | Averaged Loss: 2.2943
+Epoch: 004/100 | Batch 0006/0010 | Averaged Loss: 2.2942
+Epoch: 004/100 | Batch 0007/0010 | Averaged Loss: 2.2926
+Epoch: 004/100 | Batch 0008/0010 | Averaged Loss: 2.2885
+Epoch: 004/100 | Batch 0009/0010 | Averaged Loss: 2.2880
+Epoch: 004/100 | Train: 10.28% | Validation: 9.94%
+Time elapsed: 0.18542104959487915 min
+Epoch: 005/100 | Batch 0000/0010 | Averaged Loss: 2.2868
+Epoch: 005/100 | Batch 0001/0010 | Averaged Loss: 2.2829
+Epoch: 005/100 | Batch 0002/0010 | Averaged Loss: 2.2768
+Epoch: 005/100 | Batch 0003/0010 | Averaged Loss: 2.2696
+Epoch: 005/100 | Batch 0004/0010 | Averaged Loss: 2.2562
+Epoch: 005/100 | Batch 0005/0010 | Averaged Loss: 2.2477
+Epoch: 005/100 | Batch 0006/0010 | Averaged Loss: 2.2331
+Epoch: 005/100 | Batch 0007/0010 | Averaged Loss: 2.2193
+Epoch: 005/100 | Batch 0008/0010 | Averaged Loss: 2.2034
+Epoch: 005/100 | Batch 0009/0010 | Averaged Loss: 2.1727
+Epoch: 005/100 | Train: 17.02% | Validation: 17.43%
+Time elapsed: 0.22247667610645294 min
+Epoch: 006/100 | Batch 0000/0010 | Averaged Loss: 2.1578
+Epoch: 006/100 | Batch 0001/0010 | Averaged Loss: 2.1419
+Epoch: 006/100 | Batch 0002/0010 | Averaged Loss: 2.1298
+Epoch: 006/100 | Batch 0003/0010 | Averaged Loss: 2.1317
+Epoch: 006/100 | Batch 0004/0010 | Averaged Loss: 2.1325
+Epoch: 006/100 | Batch 0005/0010 | Averaged Loss: 2.1078
+Epoch: 006/100 | Batch 0006/0010 | Averaged Loss: 2.0945
+Epoch: 006/100 | Batch 0007/0010 | Averaged Loss: 2.1315
+Epoch: 006/100 | Batch 0008/0010 | Averaged Loss: 2.2913
+Epoch: 006/100 | Batch 0009/0010 | Averaged Loss: 2.3182
+Epoch: 006/100 | Train: 12.67% | Validation: 12.94%
+Time elapsed: 0.25950586795806885 min
+Epoch: 007/100 | Batch 0000/0010 | Averaged Loss: 2.2116
+Epoch: 007/100 | Batch 0001/0010 | Averaged Loss: 2.2863
+Epoch: 007/100 | Batch 0002/0010 | Averaged Loss: 2.1611
+Epoch: 007/100 | Batch 0003/0010 | Averaged Loss: 2.1856
+Epoch: 007/100 | Batch 0004/0010 | Averaged Loss: 2.1705
+Epoch: 007/100 | Batch 0005/0010 | Averaged Loss: 2.1617
+Epoch: 007/100 | Batch 0006/0010 | Averaged Loss: 2.1723
+Epoch: 007/100 | Batch 0007/0010 | Averaged Loss: 2.1417
+Epoch: 007/100 | Batch 0008/0010 | Averaged Loss: 2.1291
+Epoch: 007/100 | Batch 0009/0010 | Averaged Loss: 2.0929
+Epoch: 007/100 | Train: 17.64% | Validation: 17.58%
+Time elapsed: 0.29679980874061584 min
+Epoch: 008/100 | Batch 0000/0010 | Averaged Loss: 2.0789
+Epoch: 008/100 | Batch 0001/0010 | Averaged Loss: 2.0577
+Epoch: 008/100 | Batch 0002/0010 | Averaged Loss: 2.0503
+Epoch: 008/100 | Batch 0003/0010 | Averaged Loss: 2.0621
+Epoch: 008/100 | Batch 0004/0010 | Averaged Loss: 2.0633
+Epoch: 008/100 | Batch 0005/0010 | Averaged Loss: 2.0454
+Epoch: 008/100 | Batch 0006/0010 | Averaged Loss: 2.0264
+Epoch: 008/100 | Batch 0007/0010 | Averaged Loss: 2.0167
+Epoch: 008/100 | Batch 0008/0010 | Averaged Loss: 2.0152
+Epoch: 008/100 | Batch 0009/0010 | Averaged Loss: 2.0070
+Epoch: 008/100 | Train: 20.17% | Validation: 19.95%
+Time elapsed: 0.333870530128479 min
+Epoch: 009/100 | Batch 0000/0010 | Averaged Loss: 2.0020
+Epoch: 009/100 | Batch 0001/0010 | Averaged Loss: 1.9857
+Epoch: 009/100 | Batch 0002/0010 | Averaged Loss: 1.9832
+Epoch: 009/100 | Batch 0003/0010 | Averaged Loss: 1.9923
+Epoch: 009/100 | Batch 0004/0010 | Averaged Loss: 1.9898
+Epoch: 009/100 | Batch 0005/0010 | Averaged Loss: 1.9695
+Epoch: 009/100 | Batch 0006/0010 | Averaged Loss: 1.9722
+Epoch: 009/100 | Batch 0007/0010 | Averaged Loss: 1.9423
+Epoch: 009/100 | Batch 0008/0010 | Averaged Loss: 1.9770
+Epoch: 009/100 | Batch 0009/0010 | Averaged Loss: 2.0733
+Epoch: 009/100 | Train: 19.17% | Validation: 19.48%
+Time elapsed: 0.3709378242492676 min
+Epoch: 010/100 | Batch 0000/0010 | Averaged Loss: 1.9436
+Epoch: 010/100 | Batch 0001/0010 | Averaged Loss: 2.2055
+Epoch: 010/100 | Batch 0002/0010 | Averaged Loss: 2.4192
+Epoch: 010/100 | Batch 0003/0010 | Averaged Loss: 2.6073
+Epoch: 010/100 | Batch 0004/0010 | Averaged Loss: 2.5402
+Epoch: 010/100 | Batch 0005/0010 | Averaged Loss: 2.4502
+Epoch: 010/100 | Batch 0006/0010 | Averaged Loss: 2.3092
+Epoch: 010/100 | Batch 0007/0010 | Averaged Loss: 2.3693
+Epoch: 010/100 | Batch 0008/0010 | Averaged Loss: 2.2762
+Epoch: 010/100 | Batch 0009/0010 | Averaged Loss: 2.2317
+Epoch: 010/100 | Train: 10.65% | Validation: 10.82%
+Time elapsed: 0.407945454120636 min
+Epoch: 011/100 | Batch 0000/0010 | Averaged Loss: 2.2156
+Epoch: 011/100 | Batch 0001/0010 | Averaged Loss: 2.2069
+Epoch: 011/100 | Batch 0002/0010 | Averaged Loss: 2.2183
+Epoch: 011/100 | Batch 0003/0010 | Averaged Loss: 2.1873
+Epoch: 011/100 | Batch 0004/0010 | Averaged Loss: 2.2060
+Epoch: 011/100 | Batch 0005/0010 | Averaged Loss: 2.2038
+Epoch: 011/100 | Batch 0006/0010 | Averaged Loss: 2.1896
+Epoch: 011/100 | Batch 0007/0010 | Averaged Loss: 2.1626
+Epoch: 011/100 | Batch 0008/0010 | Averaged Loss: 2.1713
+Epoch: 011/100 | Batch 0009/0010 | Averaged Loss: 2.1538
+Epoch: 011/100 | Train: 17.01% | Validation: 17.19%
+Time elapsed: 0.44496387243270874 min
+Epoch: 012/100 | Batch 0000/0010 | Averaged Loss: 2.1574
+Epoch: 012/100 | Batch 0001/0010 | Averaged Loss: 2.1580
+Epoch: 012/100 | Batch 0002/0010 | Averaged Loss: 2.1418
+Epoch: 012/100 | Batch 0003/0010 | Averaged Loss: 2.1217
+Epoch: 012/100 | Batch 0004/0010 | Averaged Loss: 2.1192
+Epoch: 012/100 | Batch 0005/0010 | Averaged Loss: 2.1086
+Epoch: 012/100 | Batch 0006/0010 | Averaged Loss: 2.1005
+Epoch: 012/100 | Batch 0007/0010 | Averaged Loss: 2.1021
+Epoch: 012/100 | Batch 0008/0010 | Averaged Loss: 2.1241
+Epoch: 012/100 | Batch 0009/0010 | Averaged Loss: 2.1069
+Epoch: 012/100 | Train: 15.03% | Validation: 15.14%
+Time elapsed: 0.48203521966934204 min
+Epoch: 013/100 | Batch 0000/0010 | Averaged Loss: 2.1947
+Epoch: 013/100 | Batch 0001/0010 | Averaged Loss: 2.0808
+Epoch: 013/100 | Batch 0002/0010 | Averaged Loss: 2.0895
+Epoch: 013/100 | Batch 0003/0010 | Averaged Loss: 2.0978
+Epoch: 013/100 | Batch 0004/0010 | Averaged Loss: 2.0890
+Epoch: 013/100 | Batch 0005/0010 | Averaged Loss: 2.1188
+Epoch: 013/100 | Batch 0006/0010 | Averaged Loss: 2.0548
+Epoch: 013/100 | Batch 0007/0010 | Averaged Loss: 2.0681
+Epoch: 013/100 | Batch 0008/0010 | Averaged Loss: 2.0677
+Epoch: 013/100 | Batch 0009/0010 | Averaged Loss: 2.0452
+Epoch: 013/100 | Train: 22.73% | Validation: 22.24%
+Time elapsed: 0.519213080406189 min
+Epoch: 014/100 | Batch 0000/0010 | Averaged Loss: 2.0377
+Epoch: 014/100 | Batch 0001/0010 | Averaged Loss: 2.0649
+Epoch: 014/100 | Batch 0002/0010 | Averaged Loss: 2.0416
+Epoch: 014/100 | Batch 0003/0010 | Averaged Loss: 2.0183
+Epoch: 014/100 | Batch 0004/0010 | Averaged Loss: 2.0330
+Epoch: 014/100 | Batch 0005/0010 | Averaged Loss: 2.0273
+Epoch: 014/100 | Batch 0006/0010 | Averaged Loss: 2.0121
+Epoch: 014/100 | Batch 0007/0010 | Averaged Loss: 2.0357
+Epoch: 014/100 | Batch 0008/0010 | Averaged Loss: 1.9841
+Epoch: 014/100 | Batch 0009/0010 | Averaged Loss: 1.9791
+Epoch: 014/100 | Train: 23.10% | Validation: 22.14%
+Time elapsed: 0.556304931640625 min
+Epoch: 015/100 | Batch 0000/0010 | Averaged Loss: 1.9827
+Epoch: 015/100 | Batch 0001/0010 | Averaged Loss: 1.9509
+Epoch: 015/100 | Batch 0002/0010 | Averaged Loss: 1.9632
+Epoch: 015/100 | Batch 0003/0010 | Averaged Loss: 1.9444
+Epoch: 015/100 | Batch 0004/0010 | Averaged Loss: 1.9529
+Epoch: 015/100 | Batch 0005/0010 | Averaged Loss: 1.9441
+Epoch: 015/100 | Batch 0006/0010 | Averaged Loss: 1.9350
+Epoch: 015/100 | Batch 0007/0010 | Averaged Loss: 1.9265
+Epoch: 015/100 | Batch 0008/0010 | Averaged Loss: 1.9236
+Epoch: 015/100 | Batch 0009/0010 | Averaged Loss: 1.9088
+Epoch: 015/100 | Train: 24.94% | Validation: 24.29%
+Time elapsed: 0.5933115482330322 min
+Epoch: 016/100 | Batch 0000/0010 | Averaged Loss: 1.8987
+Epoch: 016/100 | Batch 0001/0010 | Averaged Loss: 1.9461
+Epoch: 016/100 | Batch 0002/0010 | Averaged Loss: 2.1959
+Epoch: 016/100 | Batch 0003/0010 | Averaged Loss: 1.9707
+Epoch: 016/100 | Batch 0004/0010 | Averaged Loss: 2.1841
+Epoch: 016/100 | Batch 0005/0010 | Averaged Loss: 2.1978
+Epoch: 016/100 | Batch 0006/0010 | Averaged Loss: 2.1647
+Epoch: 016/100 | Batch 0007/0010 | Averaged Loss: 2.1331
+Epoch: 016/100 | Batch 0008/0010 | Averaged Loss: 2.1279
+Epoch: 016/100 | Batch 0009/0010 | Averaged Loss: 2.0728
+Epoch: 016/100 | Train: 19.60% | Validation: 19.53%
+Time elapsed: 0.630297839641571 min
+Epoch: 017/100 | Batch 0000/0010 | Averaged Loss: 2.0520
+Epoch: 017/100 | Batch 0001/0010 | Averaged Loss: 2.0593
+Epoch: 017/100 | Batch 0002/0010 | Averaged Loss: 2.0142
+Epoch: 017/100 | Batch 0003/0010 | Averaged Loss: 2.0654
+Epoch: 017/100 | Batch 0004/0010 | Averaged Loss: 2.0308
+Epoch: 017/100 | Batch 0005/0010 | Averaged Loss: 1.9735
+Epoch: 017/100 | Batch 0006/0010 | Averaged Loss: 2.0438
+Epoch: 017/100 | Batch 0007/0010 | Averaged Loss: 2.0271
+Epoch: 017/100 | Batch 0008/0010 | Averaged Loss: 2.0121
+Epoch: 017/100 | Batch 0009/0010 | Averaged Loss: 2.0689
+Epoch: 017/100 | Train: 20.38% | Validation: 19.65%
+Time elapsed: 0.6672935485839844 min
+Epoch: 018/100 | Batch 0000/0010 | Averaged Loss: 1.9735
+Epoch: 018/100 | Batch 0001/0010 | Averaged Loss: 2.0019
+Epoch: 018/100 | Batch 0002/0010 | Averaged Loss: 1.9543
+Epoch: 018/100 | Batch 0003/0010 | Averaged Loss: 1.9863
+Epoch: 018/100 | Batch 0004/0010 | Averaged Loss: 1.9748
+Epoch: 018/100 | Batch 0005/0010 | Averaged Loss: 1.9776
+Epoch: 018/100 | Batch 0006/0010 | Averaged Loss: 1.9568
+Epoch: 018/100 | Batch 0007/0010 | Averaged Loss: 1.9432
+Epoch: 018/100 | Batch 0008/0010 | Averaged Loss: 1.9450
+Epoch: 018/100 | Batch 0009/0010 | Averaged Loss: 1.9529
+Epoch: 018/100 | Train: 21.91% | Validation: 21.04%
+Time elapsed: 0.7044015526771545 min
+Epoch: 019/100 | Batch 0000/0010 | Averaged Loss: 1.9344
+Epoch: 019/100 | Batch 0001/0010 | Averaged Loss: 1.9300
+Epoch: 019/100 | Batch 0002/0010 | Averaged Loss: 1.9170
+Epoch: 019/100 | Batch 0003/0010 | Averaged Loss: 1.9207
+Epoch: 019/100 | Batch 0004/0010 | Averaged Loss: 1.9321
+Epoch: 019/100 | Batch 0005/0010 | Averaged Loss: 1.9144
+Epoch: 019/100 | Batch 0006/0010 | Averaged Loss: 1.9169
+Epoch: 019/100 | Batch 0007/0010 | Averaged Loss: 1.9181
+Epoch: 019/100 | Batch 0008/0010 | Averaged Loss: 1.9202
+Epoch: 019/100 | Batch 0009/0010 | Averaged Loss: 1.9220
+Epoch: 019/100 | Train: 21.87% | Validation: 21.02%
+Time elapsed: 0.7415139675140381 min
+Epoch: 020/100 | Batch 0000/0010 | Averaged Loss: 1.8992
+Epoch: 020/100 | Batch 0001/0010 | Averaged Loss: 1.9109
+Epoch: 020/100 | Batch 0002/0010 | Averaged Loss: 1.8898
+Epoch: 020/100 | Batch 0003/0010 | Averaged Loss: 1.8905
+Epoch: 020/100 | Batch 0004/0010 | Averaged Loss: 1.8983
+Epoch: 020/100 | Batch 0005/0010 | Averaged Loss: 1.8784
+Epoch: 020/100 | Batch 0006/0010 | Averaged Loss: 1.8922
+Epoch: 020/100 | Batch 0007/0010 | Averaged Loss: 1.9031
+Epoch: 020/100 | Batch 0008/0010 | Averaged Loss: 1.8931
+Epoch: 020/100 | Batch 0009/0010 | Averaged Loss: 1.8864
+Epoch: 020/100 | Train: 22.07% | Validation: 21.85%
+Time elapsed: 0.7785535454750061 min
+Epoch: 021/100 | Batch 0000/0010 | Averaged Loss: 1.8998
+Epoch: 021/100 | Batch 0001/0010 | Averaged Loss: 1.8885
+Epoch: 021/100 | Batch 0002/0010 | Averaged Loss: 1.8800
+Epoch: 021/100 | Batch 0003/0010 | Averaged Loss: 1.8981
+Epoch: 021/100 | Batch 0004/0010 | Averaged Loss: 1.9003
+Epoch: 021/100 | Batch 0005/0010 | Averaged Loss: 1.8759
+Epoch: 021/100 | Batch 0006/0010 | Averaged Loss: 1.9028
+Epoch: 021/100 | Batch 0007/0010 | Averaged Loss: 1.9050
+Epoch: 021/100 | Batch 0008/0010 | Averaged Loss: 1.8839
+Epoch: 021/100 | Batch 0009/0010 | Averaged Loss: 1.9148
+Epoch: 021/100 | Train: 24.91% | Validation: 24.37%
+Time elapsed: 0.8155739307403564 min
+Epoch: 022/100 | Batch 0000/0010 | Averaged Loss: 1.8806
+Epoch: 022/100 | Batch 0001/0010 | Averaged Loss: 1.8975
+Epoch: 022/100 | Batch 0002/0010 | Averaged Loss: 1.8900
+Epoch: 022/100 | Batch 0003/0010 | Averaged Loss: 1.8467
+Epoch: 022/100 | Batch 0004/0010 | Averaged Loss: 1.8902
+Epoch: 022/100 | Batch 0005/0010 | Averaged Loss: 1.8821
+Epoch: 022/100 | Batch 0006/0010 | Averaged Loss: 1.8665
+Epoch: 022/100 | Batch 0007/0010 | Averaged Loss: 1.8649
+Epoch: 022/100 | Batch 0008/0010 | Averaged Loss: 1.8460
+Epoch: 022/100 | Batch 0009/0010 | Averaged Loss: 1.8622
+Epoch: 022/100 | Train: 28.40% | Validation: 27.27%
+Time elapsed: 0.8526908159255981 min
+Epoch: 023/100 | Batch 0000/0010 | Averaged Loss: 1.8387
+Epoch: 023/100 | Batch 0001/0010 | Averaged Loss: 1.8447
+Epoch: 023/100 | Batch 0002/0010 | Averaged Loss: 1.8615
+Epoch: 023/100 | Batch 0003/0010 | Averaged Loss: 1.8312
+Epoch: 023/100 | Batch 0004/0010 | Averaged Loss: 1.8395
+Epoch: 023/100 | Batch 0005/0010 | Averaged Loss: 1.8227
+Epoch: 023/100 | Batch 0006/0010 | Averaged Loss: 1.8133
+Epoch: 023/100 | Batch 0007/0010 | Averaged Loss: 1.8185
+Epoch: 023/100 | Batch 0008/0010 | Averaged Loss: 1.8084
+Epoch: 023/100 | Batch 0009/0010 | Averaged Loss: 1.7859
+Epoch: 023/100 | Train: 29.98% | Validation: 29.42%
+Time elapsed: 0.8899129033088684 min
+Epoch: 024/100 | Batch 0000/0010 | Averaged Loss: 1.7731
+Epoch: 024/100 | Batch 0001/0010 | Averaged Loss: 1.8072
+Epoch: 024/100 | Batch 0002/0010 | Averaged Loss: 1.7867
+Epoch: 024/100 | Batch 0003/0010 | Averaged Loss: 1.7522
+Epoch: 024/100 | Batch 0004/0010 | Averaged Loss: 1.7593
+Epoch: 024/100 | Batch 0005/0010 | Averaged Loss: 1.8207
+Epoch: 024/100 | Batch 0006/0010 | Averaged Loss: 1.7860
+Epoch: 024/100 | Batch 0007/0010 | Averaged Loss: 1.7413
+Epoch: 024/100 | Batch 0008/0010 | Averaged Loss: 1.7650
+Epoch: 024/100 | Batch 0009/0010 | Averaged Loss: 1.8222
+Epoch: 024/100 | Train: 31.78% | Validation: 31.30%
+Time elapsed: 0.9270869493484497 min
+Epoch: 025/100 | Batch 0000/0010 | Averaged Loss: 1.8041
+Epoch: 025/100 | Batch 0001/0010 | Averaged Loss: 1.7252
+Epoch: 025/100 | Batch 0002/0010 | Averaged Loss: 1.8063
+Epoch: 025/100 | Batch 0003/0010 | Averaged Loss: 1.8786
+Epoch: 025/100 | Batch 0004/0010 | Averaged Loss: 1.8200
+Epoch: 025/100 | Batch 0005/0010 | Averaged Loss: 1.8795
+Epoch: 025/100 | Batch 0006/0010 | Averaged Loss: 1.7825
+Epoch: 025/100 | Batch 0007/0010 | Averaged Loss: 1.7477
+Epoch: 025/100 | Batch 0008/0010 | Averaged Loss: 1.8658
+Epoch: 025/100 | Batch 0009/0010 | Averaged Loss: 1.7885
+Epoch: 025/100 | Train: 26.18% | Validation: 25.39%
+Time elapsed: 0.9641472101211548 min
+Epoch: 026/100 | Batch 0000/0010 | Averaged Loss: 1.8689
+Epoch: 026/100 | Batch 0001/0010 | Averaged Loss: 1.8437
+Epoch: 026/100 | Batch 0002/0010 | Averaged Loss: 1.7963
+Epoch: 026/100 | Batch 0003/0010 | Averaged Loss: 1.9435
+Epoch: 026/100 | Batch 0004/0010 | Averaged Loss: 1.9748
+Epoch: 026/100 | Batch 0005/0010 | Averaged Loss: 1.9308
+Epoch: 026/100 | Batch 0006/0010 | Averaged Loss: 1.8904
+Epoch: 026/100 | Batch 0007/0010 | Averaged Loss: 1.8470
+Epoch: 026/100 | Batch 0008/0010 | Averaged Loss: 1.8498
+Epoch: 026/100 | Batch 0009/0010 | Averaged Loss: 1.7916
+Epoch: 026/100 | Train: 30.99% | Validation: 30.00%
+Time elapsed: 1.0011820793151855 min
+Epoch: 027/100 | Batch 0000/0010 | Averaged Loss: 1.7815
+Epoch: 027/100 | Batch 0001/0010 | Averaged Loss: 1.7696
+Epoch: 027/100 | Batch 0002/0010 | Averaged Loss: 1.7500
+Epoch: 027/100 | Batch 0003/0010 | Averaged Loss: 1.7525
+Epoch: 027/100 | Batch 0004/0010 | Averaged Loss: 1.7612
+Epoch: 027/100 | Batch 0005/0010 | Averaged Loss: 1.7712
+Epoch: 027/100 | Batch 0006/0010 | Averaged Loss: 1.7260
+Epoch: 027/100 | Batch 0007/0010 | Averaged Loss: 1.7384
+Epoch: 027/100 | Batch 0008/0010 | Averaged Loss: 1.7952
+Epoch: 027/100 | Batch 0009/0010 | Averaged Loss: 1.6728
+Epoch: 027/100 | Train: 33.31% | Validation: 32.67%
+Time elapsed: 1.03824782371521 min
+Epoch: 028/100 | Batch 0000/0010 | Averaged Loss: 1.7504
+Epoch: 028/100 | Batch 0001/0010 | Averaged Loss: 1.6755
+Epoch: 028/100 | Batch 0002/0010 | Averaged Loss: 1.6967
+Epoch: 028/100 | Batch 0003/0010 | Averaged Loss: 1.7066
+Epoch: 028/100 | Batch 0004/0010 | Averaged Loss: 1.6769
+Epoch: 028/100 | Batch 0005/0010 | Averaged Loss: 1.6862
+Epoch: 028/100 | Batch 0006/0010 | Averaged Loss: 1.6325
+Epoch: 028/100 | Batch 0007/0010 | Averaged Loss: 1.6486
+Epoch: 028/100 | Batch 0008/0010 | Averaged Loss: 1.6259
+Epoch: 028/100 | Batch 0009/0010 | Averaged Loss: 1.6001
+Epoch: 028/100 | Train: 38.74% | Validation: 38.79%
+Time elapsed: 1.0751838684082031 min
+Epoch: 029/100 | Batch 0000/0010 | Averaged Loss: 1.6286
+Epoch: 029/100 | Batch 0001/0010 | Averaged Loss: 1.6227
+Epoch: 029/100 | Batch 0002/0010 | Averaged Loss: 1.5950
+Epoch: 029/100 | Batch 0003/0010 | Averaged Loss: 1.6024
+Epoch: 029/100 | Batch 0004/0010 | Averaged Loss: 1.5890
+Epoch: 029/100 | Batch 0005/0010 | Averaged Loss: 1.5554
+Epoch: 029/100 | Batch 0006/0010 | Averaged Loss: 1.5874
+Epoch: 029/100 | Batch 0007/0010 | Averaged Loss: 1.5484
+Epoch: 029/100 | Batch 0008/0010 | Averaged Loss: 1.5662
+Epoch: 029/100 | Batch 0009/0010 | Averaged Loss: 1.5118
+Epoch: 029/100 | Train: 42.30% | Validation: 42.31%
+Time elapsed: 1.1122556924819946 min
+Epoch: 030/100 | Batch 0000/0010 | Averaged Loss: 1.5414
+Epoch: 030/100 | Batch 0001/0010 | Averaged Loss: 1.5134
+Epoch: 030/100 | Batch 0002/0010 | Averaged Loss: 1.4737
+Epoch: 030/100 | Batch 0003/0010 | Averaged Loss: 1.5051
+Epoch: 030/100 | Batch 0004/0010 | Averaged Loss: 1.5201
+Epoch: 030/100 | Batch 0005/0010 | Averaged Loss: 1.4724
+Epoch: 030/100 | Batch 0006/0010 | Averaged Loss: 1.4991
+Epoch: 030/100 | Batch 0007/0010 | Averaged Loss: 1.5392
+Epoch: 030/100 | Batch 0008/0010 | Averaged Loss: 1.5930
+Epoch: 030/100 | Batch 0009/0010 | Averaged Loss: 1.7150
+Epoch: 030/100 | Train: 33.26% | Validation: 34.96%
+Time elapsed: 1.1493892669677734 min
+Epoch: 031/100 | Batch 0000/0010 | Averaged Loss: 1.8362
+Epoch: 031/100 | Batch 0001/0010 | Averaged Loss: 1.9788
+Epoch: 031/100 | Batch 0002/0010 | Averaged Loss: 2.0570
+Epoch: 031/100 | Batch 0003/0010 | Averaged Loss: 2.1727
+Epoch: 031/100 | Batch 0004/0010 | Averaged Loss: 1.9275
+Epoch: 031/100 | Batch 0005/0010 | Averaged Loss: 2.0774
+Epoch: 031/100 | Batch 0006/0010 | Averaged Loss: 2.0087
+Epoch: 031/100 | Batch 0007/0010 | Averaged Loss: 1.9838
+Epoch: 031/100 | Batch 0008/0010 | Averaged Loss: 1.8963
+Epoch: 031/100 | Batch 0009/0010 | Averaged Loss: 1.9197
+Epoch: 031/100 | Train: 26.28% | Validation: 25.56%
+Time elapsed: 1.1865429878234863 min
+Epoch: 032/100 | Batch 0000/0010 | Averaged Loss: 1.8172
+Epoch: 032/100 | Batch 0001/0010 | Averaged Loss: 1.8704
+Epoch: 032/100 | Batch 0002/0010 | Averaged Loss: 1.8051
+Epoch: 032/100 | Batch 0003/0010 | Averaged Loss: 1.7933
+Epoch: 032/100 | Batch 0004/0010 | Averaged Loss: 1.8120
+Epoch: 032/100 | Batch 0005/0010 | Averaged Loss: 1.7808
+Epoch: 032/100 | Batch 0006/0010 | Averaged Loss: 1.7903
+Epoch: 032/100 | Batch 0007/0010 | Averaged Loss: 1.7068
+Epoch: 032/100 | Batch 0008/0010 | Averaged Loss: 1.7268
+Epoch: 032/100 | Batch 0009/0010 | Averaged Loss: 1.7649
+Epoch: 032/100 | Train: 33.28% | Validation: 32.20%
+Time elapsed: 1.2235362529754639 min
+Epoch: 033/100 | Batch 0000/0010 | Averaged Loss: 1.6961
+Epoch: 033/100 | Batch 0001/0010 | Averaged Loss: 1.7131
+Epoch: 033/100 | Batch 0002/0010 | Averaged Loss: 1.6417
+Epoch: 033/100 | Batch 0003/0010 | Averaged Loss: 1.6740
+Epoch: 033/100 | Batch 0004/0010 | Averaged Loss: 1.6433
+Epoch: 033/100 | Batch 0005/0010 | Averaged Loss: 1.5953
+Epoch: 033/100 | Batch 0006/0010 | Averaged Loss: 1.6239
+Epoch: 033/100 | Batch 0007/0010 | Averaged Loss: 1.6217
+Epoch: 033/100 | Batch 0008/0010 | Averaged Loss: 1.6063
+Epoch: 033/100 | Batch 0009/0010 | Averaged Loss: 1.5903
+Epoch: 033/100 | Train: 39.79% | Validation: 38.94%
+Time elapsed: 1.2606492042541504 min
+Epoch: 034/100 | Batch 0000/0010 | Averaged Loss: 1.5928
+Epoch: 034/100 | Batch 0001/0010 | Averaged Loss: 1.5767
+Epoch: 034/100 | Batch 0002/0010 | Averaged Loss: 1.5510
+Epoch: 034/100 | Batch 0003/0010 | Averaged Loss: 1.5700
+Epoch: 034/100 | Batch 0004/0010 | Averaged Loss: 1.5594
+Epoch: 034/100 | Batch 0005/0010 | Averaged Loss: 1.5414
+Epoch: 034/100 | Batch 0006/0010 | Averaged Loss: 1.5479
+Epoch: 034/100 | Batch 0007/0010 | Averaged Loss: 1.5487
+Epoch: 034/100 | Batch 0008/0010 | Averaged Loss: 1.4961
+Epoch: 034/100 | Batch 0009/0010 | Averaged Loss: 1.5163
+Epoch: 034/100 | Train: 42.57% | Validation: 42.09%
+Time elapsed: 1.297670841217041 min
+Epoch: 035/100 | Batch 0000/0010 | Averaged Loss: 1.5223
+Epoch: 035/100 | Batch 0001/0010 | Averaged Loss: 1.5323
+Epoch: 035/100 | Batch 0002/0010 | Averaged Loss: 1.5386
+Epoch: 035/100 | Batch 0003/0010 | Averaged Loss: 1.5266
+Epoch: 035/100 | Batch 0004/0010 | Averaged Loss: 1.5929
+Epoch: 035/100 | Batch 0005/0010 | Averaged Loss: 1.4971
+Epoch: 035/100 | Batch 0006/0010 | Averaged Loss: 1.5529
+Epoch: 035/100 | Batch 0007/0010 | Averaged Loss: 1.5004
+Epoch: 035/100 | Batch 0008/0010 | Averaged Loss: 1.5147
+Epoch: 035/100 | Batch 0009/0010 | Averaged Loss: 1.4898
+Epoch: 035/100 | Train: 45.30% | Validation: 45.80%
+Time elapsed: 1.3346147537231445 min
+Epoch: 036/100 | Batch 0000/0010 | Averaged Loss: 1.4750
+Epoch: 036/100 | Batch 0001/0010 | Averaged Loss: 1.4717
+Epoch: 036/100 | Batch 0002/0010 | Averaged Loss: 1.4365
+Epoch: 036/100 | Batch 0003/0010 | Averaged Loss: 1.4365
+Epoch: 036/100 | Batch 0004/0010 | Averaged Loss: 1.4047
+Epoch: 036/100 | Batch 0005/0010 | Averaged Loss: 1.4270
+Epoch: 036/100 | Batch 0006/0010 | Averaged Loss: 1.3914
+Epoch: 036/100 | Batch 0007/0010 | Averaged Loss: 1.4259
+Epoch: 036/100 | Batch 0008/0010 | Averaged Loss: 1.4251
+Epoch: 036/100 | Batch 0009/0010 | Averaged Loss: 1.3988
+Epoch: 036/100 | Train: 46.74% | Validation: 47.36%
+Time elapsed: 1.3715450763702393 min
+Epoch: 037/100 | Batch 0000/0010 | Averaged Loss: 1.4162
+Epoch: 037/100 | Batch 0001/0010 | Averaged Loss: 1.4337
+Epoch: 037/100 | Batch 0002/0010 | Averaged Loss: 1.3892
+Epoch: 037/100 | Batch 0003/0010 | Averaged Loss: 1.4253
+Epoch: 037/100 | Batch 0004/0010 | Averaged Loss: 1.4107
+Epoch: 037/100 | Batch 0005/0010 | Averaged Loss: 1.4566
+Epoch: 037/100 | Batch 0006/0010 | Averaged Loss: 1.4809
+Epoch: 037/100 | Batch 0007/0010 | Averaged Loss: 1.4606
+Epoch: 037/100 | Batch 0008/0010 | Averaged Loss: 1.4716
+Epoch: 037/100 | Batch 0009/0010 | Averaged Loss: 1.3977
+Epoch: 037/100 | Train: 46.99% | Validation: 46.66%
+Time elapsed: 1.4085396528244019 min
+Epoch: 038/100 | Batch 0000/0010 | Averaged Loss: 1.4443
+Epoch: 038/100 | Batch 0001/0010 | Averaged Loss: 1.4092
+Epoch: 038/100 | Batch 0002/0010 | Averaged Loss: 1.4680
+Epoch: 038/100 | Batch 0003/0010 | Averaged Loss: 1.3344
+Epoch: 038/100 | Batch 0004/0010 | Averaged Loss: 1.3657
+Epoch: 038/100 | Batch 0005/0010 | Averaged Loss: 1.3275
+Epoch: 038/100 | Batch 0006/0010 | Averaged Loss: 1.3315
+Epoch: 038/100 | Batch 0007/0010 | Averaged Loss: 1.3407
+Epoch: 038/100 | Batch 0008/0010 | Averaged Loss: 1.3356
+Epoch: 038/100 | Batch 0009/0010 | Averaged Loss: 1.3453
+Epoch: 038/100 | Train: 50.68% | Validation: 51.22%
+Time elapsed: 1.4454755783081055 min
+Epoch: 039/100 | Batch 0000/0010 | Averaged Loss: 1.3137
+Epoch: 039/100 | Batch 0001/0010 | Averaged Loss: 1.2973
+Epoch: 039/100 | Batch 0002/0010 | Averaged Loss: 1.3043
+Epoch: 039/100 | Batch 0003/0010 | Averaged Loss: 1.2862
+Epoch: 039/100 | Batch 0004/0010 | Averaged Loss: 1.2830
+Epoch: 039/100 | Batch 0005/0010 | Averaged Loss: 1.3031
+Epoch: 039/100 | Batch 0006/0010 | Averaged Loss: 1.2665
+Epoch: 039/100 | Batch 0007/0010 | Averaged Loss: 1.3049
+Epoch: 039/100 | Batch 0008/0010 | Averaged Loss: 1.2686
+Epoch: 039/100 | Batch 0009/0010 | Averaged Loss: 1.2607
+Epoch: 039/100 | Train: 54.06% | Validation: 53.42%
+Time elapsed: 1.4824990034103394 min
+Epoch: 040/100 | Batch 0000/0010 | Averaged Loss: 1.2448
+Epoch: 040/100 | Batch 0001/0010 | Averaged Loss: 1.2673
+Epoch: 040/100 | Batch 0002/0010 | Averaged Loss: 1.2885
+Epoch: 040/100 | Batch 0003/0010 | Averaged Loss: 1.3087
+Epoch: 040/100 | Batch 0004/0010 | Averaged Loss: 1.2732
+Epoch: 040/100 | Batch 0005/0010 | Averaged Loss: 1.2661
+Epoch: 040/100 | Batch 0006/0010 | Averaged Loss: 1.4456
+Epoch: 040/100 | Batch 0007/0010 | Averaged Loss: 1.5739
+Epoch: 040/100 | Batch 0008/0010 | Averaged Loss: 1.6657
+Epoch: 040/100 | Batch 0009/0010 | Averaged Loss: 1.7690
+Epoch: 040/100 | Train: 24.57% | Validation: 24.32%
+Time elapsed: 1.519448161125183 min
+Epoch: 041/100 | Batch 0000/0010 | Averaged Loss: 2.1961
+Epoch: 041/100 | Batch 0001/0010 | Averaged Loss: 1.8973
+Epoch: 041/100 | Batch 0002/0010 | Averaged Loss: 1.9337
+Epoch: 041/100 | Batch 0003/0010 | Averaged Loss: 2.0041
+Epoch: 041/100 | Batch 0004/0010 | Averaged Loss: 1.9637
+Epoch: 041/100 | Batch 0005/0010 | Averaged Loss: 1.8149
+Epoch: 041/100 | Batch 0006/0010 | Averaged Loss: 1.8682
+Epoch: 041/100 | Batch 0007/0010 | Averaged Loss: 1.7475
+Epoch: 041/100 | Batch 0008/0010 | Averaged Loss: 1.7398
+Epoch: 041/100 | Batch 0009/0010 | Averaged Loss: 1.6963
+Epoch: 041/100 | Train: 33.45% | Validation: 32.54%
+Time elapsed: 1.5564038753509521 min
+Epoch: 042/100 | Batch 0000/0010 | Averaged Loss: 1.7370
+Epoch: 042/100 | Batch 0001/0010 | Averaged Loss: 1.6860
+Epoch: 042/100 | Batch 0002/0010 | Averaged Loss: 1.6980
+Epoch: 042/100 | Batch 0003/0010 | Averaged Loss: 1.6122
+Epoch: 042/100 | Batch 0004/0010 | Averaged Loss: 1.5950
+Epoch: 042/100 | Batch 0005/0010 | Averaged Loss: 1.6323
+Epoch: 042/100 | Batch 0006/0010 | Averaged Loss: 1.5922
+Epoch: 042/100 | Batch 0007/0010 | Averaged Loss: 1.5729
+Epoch: 042/100 | Batch 0008/0010 | Averaged Loss: 1.5293
+Epoch: 042/100 | Batch 0009/0010 | Averaged Loss: 1.5164
+Epoch: 042/100 | Train: 40.52% | Validation: 39.60%
+Time elapsed: 1.5933558940887451 min
+Epoch: 043/100 | Batch 0000/0010 | Averaged Loss: 1.5466
+Epoch: 043/100 | Batch 0001/0010 | Averaged Loss: 1.5076
+Epoch: 043/100 | Batch 0002/0010 | Averaged Loss: 1.5147
+Epoch: 043/100 | Batch 0003/0010 | Averaged Loss: 1.4922
+Epoch: 043/100 | Batch 0004/0010 | Averaged Loss: 1.4786
+Epoch: 043/100 | Batch 0005/0010 | Averaged Loss: 1.4620
+Epoch: 043/100 | Batch 0006/0010 | Averaged Loss: 1.4389
+Epoch: 043/100 | Batch 0007/0010 | Averaged Loss: 1.4070
+Epoch: 043/100 | Batch 0008/0010 | Averaged Loss: 1.4583
+Epoch: 043/100 | Batch 0009/0010 | Averaged Loss: 1.4194
+Epoch: 043/100 | Train: 47.55% | Validation: 47.66%
+Time elapsed: 1.6305038928985596 min
+Epoch: 044/100 | Batch 0000/0010 | Averaged Loss: 1.4038
+Epoch: 044/100 | Batch 0001/0010 | Averaged Loss: 1.3987
+Epoch: 044/100 | Batch 0002/0010 | Averaged Loss: 1.3914
+Epoch: 044/100 | Batch 0003/0010 | Averaged Loss: 1.4035
+Epoch: 044/100 | Batch 0004/0010 | Averaged Loss: 1.4250
+Epoch: 044/100 | Batch 0005/0010 | Averaged Loss: 1.3345
+Epoch: 044/100 | Batch 0006/0010 | Averaged Loss: 1.3770
+Epoch: 044/100 | Batch 0007/0010 | Averaged Loss: 1.4916
+Epoch: 044/100 | Batch 0008/0010 | Averaged Loss: 1.4024
+Epoch: 044/100 | Batch 0009/0010 | Averaged Loss: 1.4913
+Epoch: 044/100 | Train: 50.54% | Validation: 50.81%
+Time elapsed: 1.6674931049346924 min
+Epoch: 045/100 | Batch 0000/0010 | Averaged Loss: 1.3178
+Epoch: 045/100 | Batch 0001/0010 | Averaged Loss: 1.4365
+Epoch: 045/100 | Batch 0002/0010 | Averaged Loss: 1.3000
+Epoch: 045/100 | Batch 0003/0010 | Averaged Loss: 1.3968
+Epoch: 045/100 | Batch 0004/0010 | Averaged Loss: 1.3745
+Epoch: 045/100 | Batch 0005/0010 | Averaged Loss: 1.3740
+Epoch: 045/100 | Batch 0006/0010 | Averaged Loss: 1.4070
+Epoch: 045/100 | Batch 0007/0010 | Averaged Loss: 1.3143
+Epoch: 045/100 | Batch 0008/0010 | Averaged Loss: 1.3981
+Epoch: 045/100 | Batch 0009/0010 | Averaged Loss: 1.4003
+Epoch: 045/100 | Train: 47.38% | Validation: 47.92%
+Time elapsed: 1.7046887874603271 min
+Epoch: 046/100 | Batch 0000/0010 | Averaged Loss: 1.4626
+Epoch: 046/100 | Batch 0001/0010 | Averaged Loss: 1.4139
+Epoch: 046/100 | Batch 0002/0010 | Averaged Loss: 1.3790
+Epoch: 046/100 | Batch 0003/0010 | Averaged Loss: 1.2929
+Epoch: 046/100 | Batch 0004/0010 | Averaged Loss: 1.4141
+Epoch: 046/100 | Batch 0005/0010 | Averaged Loss: 1.3388
+Epoch: 046/100 | Batch 0006/0010 | Averaged Loss: 1.3471
+Epoch: 046/100 | Batch 0007/0010 | Averaged Loss: 1.2685
+Epoch: 046/100 | Batch 0008/0010 | Averaged Loss: 1.3113
+Epoch: 046/100 | Batch 0009/0010 | Averaged Loss: 1.2611
+Epoch: 046/100 | Train: 51.86% | Validation: 50.68%
+Time elapsed: 1.7416425943374634 min
+Epoch: 047/100 | Batch 0000/0010 | Averaged Loss: 1.2432
+Epoch: 047/100 | Batch 0001/0010 | Averaged Loss: 1.2449
+Epoch: 047/100 | Batch 0002/0010 | Averaged Loss: 1.2112
+Epoch: 047/100 | Batch 0003/0010 | Averaged Loss: 1.2783
+Epoch: 047/100 | Batch 0004/0010 | Averaged Loss: 1.1996
+Epoch: 047/100 | Batch 0005/0010 | Averaged Loss: 1.2225
+Epoch: 047/100 | Batch 0006/0010 | Averaged Loss: 1.2193
+Epoch: 047/100 | Batch 0007/0010 | Averaged Loss: 1.2368
+Epoch: 047/100 | Batch 0008/0010 | Averaged Loss: 1.2361
+Epoch: 047/100 | Batch 0009/0010 | Averaged Loss: 1.2276
+Epoch: 047/100 | Train: 54.62% | Validation: 54.32%
+Time elapsed: 1.7786788940429688 min
+Epoch: 048/100 | Batch 0000/0010 | Averaged Loss: 1.2916
+Epoch: 048/100 | Batch 0001/0010 | Averaged Loss: 1.3619
+Epoch: 048/100 | Batch 0002/0010 | Averaged Loss: 1.2528
+Epoch: 048/100 | Batch 0003/0010 | Averaged Loss: 1.2578
+Epoch: 048/100 | Batch 0004/0010 | Averaged Loss: 1.2787
+Epoch: 048/100 | Batch 0005/0010 | Averaged Loss: 1.2268
+Epoch: 048/100 | Batch 0006/0010 | Averaged Loss: 1.1963
+Epoch: 048/100 | Batch 0007/0010 | Averaged Loss: 1.2187
+Epoch: 048/100 | Batch 0008/0010 | Averaged Loss: 1.1636
+Epoch: 048/100 | Batch 0009/0010 | Averaged Loss: 1.1459
+Epoch: 048/100 | Train: 58.10% | Validation: 58.35%
+Time elapsed: 1.815711498260498 min
+Epoch: 049/100 | Batch 0000/0010 | Averaged Loss: 1.1479
+Epoch: 049/100 | Batch 0001/0010 | Averaged Loss: 1.1211
+Epoch: 049/100 | Batch 0002/0010 | Averaged Loss: 1.1114
+Epoch: 049/100 | Batch 0003/0010 | Averaged Loss: 1.1582
+Epoch: 049/100 | Batch 0004/0010 | Averaged Loss: 1.1489
+Epoch: 049/100 | Batch 0005/0010 | Averaged Loss: 1.0956
+Epoch: 049/100 | Batch 0006/0010 | Averaged Loss: 1.1159
+Epoch: 049/100 | Batch 0007/0010 | Averaged Loss: 1.1273
+Epoch: 049/100 | Batch 0008/0010 | Averaged Loss: 1.0911
+Epoch: 049/100 | Batch 0009/0010 | Averaged Loss: 1.1103
+Epoch: 049/100 | Train: 60.90% | Validation: 59.84%
+Time elapsed: 1.8528021574020386 min
+Epoch: 050/100 | Batch 0000/0010 | Averaged Loss: 1.0884
+Epoch: 050/100 | Batch 0001/0010 | Averaged Loss: 1.0905
+Epoch: 050/100 | Batch 0002/0010 | Averaged Loss: 1.0471
+Epoch: 050/100 | Batch 0003/0010 | Averaged Loss: 1.1085
+Epoch: 050/100 | Batch 0004/0010 | Averaged Loss: 1.0525
+Epoch: 050/100 | Batch 0005/0010 | Averaged Loss: 1.0805
+Epoch: 050/100 | Batch 0006/0010 | Averaged Loss: 1.0845
+Epoch: 050/100 | Batch 0007/0010 | Averaged Loss: 1.1600
+Epoch: 050/100 | Batch 0008/0010 | Averaged Loss: 1.2793
+Epoch: 050/100 | Batch 0009/0010 | Averaged Loss: 1.2606
+Epoch: 050/100 | Train: 48.63% | Validation: 47.92%
+Time elapsed: 1.889955759048462 min
+Epoch: 051/100 | Batch 0000/0010 | Averaged Loss: 1.4277
+Epoch: 051/100 | Batch 0001/0010 | Averaged Loss: 1.2638
+Epoch: 051/100 | Batch 0002/0010 | Averaged Loss: 1.5601
+Epoch: 051/100 | Batch 0003/0010 | Averaged Loss: 1.4913
+Epoch: 051/100 | Batch 0004/0010 | Averaged Loss: 1.3752
+Epoch: 051/100 | Batch 0005/0010 | Averaged Loss: 1.5334
+Epoch: 051/100 | Batch 0006/0010 | Averaged Loss: 1.4754
+Epoch: 051/100 | Batch 0007/0010 | Averaged Loss: 1.4698
+Epoch: 051/100 | Batch 0008/0010 | Averaged Loss: 1.4197
+Epoch: 051/100 | Batch 0009/0010 | Averaged Loss: 1.3706
+Epoch: 051/100 | Train: 52.74% | Validation: 52.32%
+Time elapsed: 1.9271082878112793 min
+Epoch: 052/100 | Batch 0000/0010 | Averaged Loss: 1.3449
+Epoch: 052/100 | Batch 0001/0010 | Averaged Loss: 1.4551
+Epoch: 052/100 | Batch 0002/0010 | Averaged Loss: 1.3179
+Epoch: 052/100 | Batch 0003/0010 | Averaged Loss: 1.3264
+Epoch: 052/100 | Batch 0004/0010 | Averaged Loss: 1.3036
+Epoch: 052/100 | Batch 0005/0010 | Averaged Loss: 1.2380
+Epoch: 052/100 | Batch 0006/0010 | Averaged Loss: 1.3154
+Epoch: 052/100 | Batch 0007/0010 | Averaged Loss: 1.2225
+Epoch: 052/100 | Batch 0008/0010 | Averaged Loss: 1.2194
+Epoch: 052/100 | Batch 0009/0010 | Averaged Loss: 1.1719
+Epoch: 052/100 | Train: 58.04% | Validation: 57.71%
+Time elapsed: 1.9642033576965332 min
+Epoch: 053/100 | Batch 0000/0010 | Averaged Loss: 1.1426
+Epoch: 053/100 | Batch 0001/0010 | Averaged Loss: 1.1841
+Epoch: 053/100 | Batch 0002/0010 | Averaged Loss: 1.1365
+Epoch: 053/100 | Batch 0003/0010 | Averaged Loss: 1.1116
+Epoch: 053/100 | Batch 0004/0010 | Averaged Loss: 1.1129
+Epoch: 053/100 | Batch 0005/0010 | Averaged Loss: 1.1094
+Epoch: 053/100 | Batch 0006/0010 | Averaged Loss: 1.0897
+Epoch: 053/100 | Batch 0007/0010 | Averaged Loss: 1.1280
+Epoch: 053/100 | Batch 0008/0010 | Averaged Loss: 1.0579
+Epoch: 053/100 | Batch 0009/0010 | Averaged Loss: 1.0696
+Epoch: 053/100 | Train: 62.43% | Validation: 60.74%
+Time elapsed: 2.001166820526123 min
+Epoch: 054/100 | Batch 0000/0010 | Averaged Loss: 1.0539
+Epoch: 054/100 | Batch 0001/0010 | Averaged Loss: 1.0312
+Epoch: 054/100 | Batch 0002/0010 | Averaged Loss: 1.0288
+Epoch: 054/100 | Batch 0003/0010 | Averaged Loss: 1.0291
+Epoch: 054/100 | Batch 0004/0010 | Averaged Loss: 1.0209
+Epoch: 054/100 | Batch 0005/0010 | Averaged Loss: 1.0436
+Epoch: 054/100 | Batch 0006/0010 | Averaged Loss: 1.0132
+Epoch: 054/100 | Batch 0007/0010 | Averaged Loss: 1.0450
+Epoch: 054/100 | Batch 0008/0010 | Averaged Loss: 1.0231
+Epoch: 054/100 | Batch 0009/0010 | Averaged Loss: 0.9628
+Epoch: 054/100 | Train: 65.40% | Validation: 64.77%
+Time elapsed: 2.0382628440856934 min
+Epoch: 055/100 | Batch 0000/0010 | Averaged Loss: 0.9515
+Epoch: 055/100 | Batch 0001/0010 | Averaged Loss: 0.9736
+Epoch: 055/100 | Batch 0002/0010 | Averaged Loss: 0.9691
+Epoch: 055/100 | Batch 0003/0010 | Averaged Loss: 1.0071
+Epoch: 055/100 | Batch 0004/0010 | Averaged Loss: 1.0821
+Epoch: 055/100 | Batch 0005/0010 | Averaged Loss: 1.1207
+Epoch: 055/100 | Batch 0006/0010 | Averaged Loss: 1.0429
+Epoch: 055/100 | Batch 0007/0010 | Averaged Loss: 1.1076
+Epoch: 055/100 | Batch 0008/0010 | Averaged Loss: 1.0571
+Epoch: 055/100 | Batch 0009/0010 | Averaged Loss: 1.0577
+Epoch: 055/100 | Train: 63.86% | Validation: 62.16%
+Time elapsed: 2.0753848552703857 min
+Epoch: 056/100 | Batch 0000/0010 | Averaged Loss: 1.0382
+Epoch: 056/100 | Batch 0001/0010 | Averaged Loss: 0.9644
+Epoch: 056/100 | Batch 0002/0010 | Averaged Loss: 1.0206
+Epoch: 056/100 | Batch 0003/0010 | Averaged Loss: 0.9598
+Epoch: 056/100 | Batch 0004/0010 | Averaged Loss: 0.9445
+Epoch: 056/100 | Batch 0005/0010 | Averaged Loss: 0.9389
+Epoch: 056/100 | Batch 0006/0010 | Averaged Loss: 0.9746
+Epoch: 056/100 | Batch 0007/0010 | Averaged Loss: 0.9416
+Epoch: 056/100 | Batch 0008/0010 | Averaged Loss: 0.9684
+Epoch: 056/100 | Batch 0009/0010 | Averaged Loss: 0.9614
+Epoch: 056/100 | Train: 66.98% | Validation: 65.11%
+Time elapsed: 2.11246395111084 min
+Epoch: 057/100 | Batch 0000/0010 | Averaged Loss: 0.9443
+Epoch: 057/100 | Batch 0001/0010 | Averaged Loss: 0.8937
+Epoch: 057/100 | Batch 0002/0010 | Averaged Loss: 0.8926
+Epoch: 057/100 | Batch 0003/0010 | Averaged Loss: 0.8949
+Epoch: 057/100 | Batch 0004/0010 | Averaged Loss: 0.9116
+Epoch: 057/100 | Batch 0005/0010 | Averaged Loss: 0.9124
+Epoch: 057/100 | Batch 0006/0010 | Averaged Loss: 0.9437
+Epoch: 057/100 | Batch 0007/0010 | Averaged Loss: 0.9279
+Epoch: 057/100 | Batch 0008/0010 | Averaged Loss: 0.9061
+Epoch: 057/100 | Batch 0009/0010 | Averaged Loss: 0.8826
+Epoch: 057/100 | Train: 68.44% | Validation: 65.41%
+Time elapsed: 2.1496105194091797 min
+Epoch: 058/100 | Batch 0000/0010 | Averaged Loss: 0.9016
+Epoch: 058/100 | Batch 0001/0010 | Averaged Loss: 0.9293
+Epoch: 058/100 | Batch 0002/0010 | Averaged Loss: 0.8912
+Epoch: 058/100 | Batch 0003/0010 | Averaged Loss: 0.9452
+Epoch: 058/100 | Batch 0004/0010 | Averaged Loss: 0.8287
+Epoch: 058/100 | Batch 0005/0010 | Averaged Loss: 0.8738
+Epoch: 058/100 | Batch 0006/0010 | Averaged Loss: 0.8791
+Epoch: 058/100 | Batch 0007/0010 | Averaged Loss: 1.0152
+Epoch: 058/100 | Batch 0008/0010 | Averaged Loss: 1.0328
+Epoch: 058/100 | Batch 0009/0010 | Averaged Loss: 0.9677
+Epoch: 058/100 | Train: 63.83% | Validation: 60.82%
+Time elapsed: 2.1867618560791016 min
+Epoch: 059/100 | Batch 0000/0010 | Averaged Loss: 0.9921
+Epoch: 059/100 | Batch 0001/0010 | Averaged Loss: 1.0315
+Epoch: 059/100 | Batch 0002/0010 | Averaged Loss: 1.0962
+Epoch: 059/100 | Batch 0003/0010 | Averaged Loss: 0.9601
+Epoch: 059/100 | Batch 0004/0010 | Averaged Loss: 0.9509
+Epoch: 059/100 | Batch 0005/0010 | Averaged Loss: 0.9799
+Epoch: 059/100 | Batch 0006/0010 | Averaged Loss: 1.0143
+Epoch: 059/100 | Batch 0007/0010 | Averaged Loss: 0.8966
+Epoch: 059/100 | Batch 0008/0010 | Averaged Loss: 0.9374
+Epoch: 059/100 | Batch 0009/0010 | Averaged Loss: 0.9072
+Epoch: 059/100 | Train: 68.55% | Validation: 65.60%
+Time elapsed: 2.223905086517334 min
+Epoch: 060/100 | Batch 0000/0010 | Averaged Loss: 0.9162
+Epoch: 060/100 | Batch 0001/0010 | Averaged Loss: 0.9033
+Epoch: 060/100 | Batch 0002/0010 | Averaged Loss: 0.9073
+Epoch: 060/100 | Batch 0003/0010 | Averaged Loss: 0.9295
+Epoch: 060/100 | Batch 0004/0010 | Averaged Loss: 0.9168
+Epoch: 060/100 | Batch 0005/0010 | Averaged Loss: 0.8869
+Epoch: 060/100 | Batch 0006/0010 | Averaged Loss: 0.8716
+Epoch: 060/100 | Batch 0007/0010 | Averaged Loss: 0.9020
+Epoch: 060/100 | Batch 0008/0010 | Averaged Loss: 0.8958
+Epoch: 060/100 | Batch 0009/0010 | Averaged Loss: 0.9146
+Epoch: 060/100 | Train: 70.51% | Validation: 66.21%
+Time elapsed: 2.2609002590179443 min
+Epoch: 061/100 | Batch 0000/0010 | Averaged Loss: 0.8531
+Epoch: 061/100 | Batch 0001/0010 | Averaged Loss: 0.8446
+Epoch: 061/100 | Batch 0002/0010 | Averaged Loss: 0.8192
+Epoch: 061/100 | Batch 0003/0010 | Averaged Loss: 0.8096
+Epoch: 061/100 | Batch 0004/0010 | Averaged Loss: 0.8235
+Epoch: 061/100 | Batch 0005/0010 | Averaged Loss: 0.8423
+Epoch: 061/100 | Batch 0006/0010 | Averaged Loss: 0.8254
+Epoch: 061/100 | Batch 0007/0010 | Averaged Loss: 0.8046
+Epoch: 061/100 | Batch 0008/0010 | Averaged Loss: 0.8436
+Epoch: 061/100 | Batch 0009/0010 | Averaged Loss: 0.8578
+Epoch: 061/100 | Train: 72.26% | Validation: 67.92%
+Time elapsed: 2.2979836463928223 min
+Epoch: 062/100 | Batch 0000/0010 | Averaged Loss: 0.7697
+Epoch: 062/100 | Batch 0001/0010 | Averaged Loss: 0.7942
+Epoch: 062/100 | Batch 0002/0010 | Averaged Loss: 0.7837
+Epoch: 062/100 | Batch 0003/0010 | Averaged Loss: 0.7761
+Epoch: 062/100 | Batch 0004/0010 | Averaged Loss: 0.8079
+Epoch: 062/100 | Batch 0005/0010 | Averaged Loss: 0.7379
+Epoch: 062/100 | Batch 0006/0010 | Averaged Loss: 0.7485
+Epoch: 062/100 | Batch 0007/0010 | Averaged Loss: 0.7355
+Epoch: 062/100 | Batch 0008/0010 | Averaged Loss: 0.7615
+Epoch: 062/100 | Batch 0009/0010 | Averaged Loss: 0.7769
+Epoch: 062/100 | Train: 74.39% | Validation: 68.63%
+Time elapsed: 2.3349552154541016 min
+Epoch: 063/100 | Batch 0000/0010 | Averaged Loss: 0.7159
+Epoch: 063/100 | Batch 0001/0010 | Averaged Loss: 0.7285
+Epoch: 063/100 | Batch 0002/0010 | Averaged Loss: 0.7376
+Epoch: 063/100 | Batch 0003/0010 | Averaged Loss: 0.7319
+Epoch: 063/100 | Batch 0004/0010 | Averaged Loss: 0.7506
+Epoch: 063/100 | Batch 0005/0010 | Averaged Loss: 0.7592
+Epoch: 063/100 | Batch 0006/0010 | Averaged Loss: 0.7739
+Epoch: 063/100 | Batch 0007/0010 | Averaged Loss: 0.8573
+Epoch: 063/100 | Batch 0008/0010 | Averaged Loss: 0.7700
+Epoch: 063/100 | Batch 0009/0010 | Averaged Loss: 0.8556
+Epoch: 063/100 | Train: 71.88% | Validation: 66.65%
+Time elapsed: 2.3720107078552246 min
+Epoch: 064/100 | Batch 0000/0010 | Averaged Loss: 0.8058
+Epoch: 064/100 | Batch 0001/0010 | Averaged Loss: 0.8052
+Epoch: 064/100 | Batch 0002/0010 | Averaged Loss: 0.8019
+Epoch: 064/100 | Batch 0003/0010 | Averaged Loss: 0.8460
+Epoch: 064/100 | Batch 0004/0010 | Averaged Loss: 0.8117
+Epoch: 064/100 | Batch 0005/0010 | Averaged Loss: 0.8214
+Epoch: 064/100 | Batch 0006/0010 | Averaged Loss: 0.7559
+Epoch: 064/100 | Batch 0007/0010 | Averaged Loss: 0.7540
+Epoch: 064/100 | Batch 0008/0010 | Averaged Loss: 0.7977
+Epoch: 064/100 | Batch 0009/0010 | Averaged Loss: 0.7502
+Epoch: 064/100 | Train: 75.09% | Validation: 69.07%
+Time elapsed: 2.4090988636016846 min
+Epoch: 065/100 | Batch 0000/0010 | Averaged Loss: 0.7133
+Epoch: 065/100 | Batch 0001/0010 | Averaged Loss: 0.7215
+Epoch: 065/100 | Batch 0002/0010 | Averaged Loss: 0.7149
+Epoch: 065/100 | Batch 0003/0010 | Averaged Loss: 0.6976
+Epoch: 065/100 | Batch 0004/0010 | Averaged Loss: 0.6971
+Epoch: 065/100 | Batch 0005/0010 | Averaged Loss: 0.6773
+Epoch: 065/100 | Batch 0006/0010 | Averaged Loss: 0.6961
+Epoch: 065/100 | Batch 0007/0010 | Averaged Loss: 0.7380
+Epoch: 065/100 | Batch 0008/0010 | Averaged Loss: 0.7354
+Epoch: 065/100 | Batch 0009/0010 | Averaged Loss: 0.7283
+Epoch: 065/100 | Train: 77.55% | Validation: 70.70%
+Time elapsed: 2.446145534515381 min
+Epoch: 066/100 | Batch 0000/0010 | Averaged Loss: 0.6481
+Epoch: 066/100 | Batch 0001/0010 | Averaged Loss: 0.7147
+Epoch: 066/100 | Batch 0002/0010 | Averaged Loss: 0.6522
+Epoch: 066/100 | Batch 0003/0010 | Averaged Loss: 0.6652
+Epoch: 066/100 | Batch 0004/0010 | Averaged Loss: 0.6590
+Epoch: 066/100 | Batch 0005/0010 | Averaged Loss: 0.6408
+Epoch: 066/100 | Batch 0006/0010 | Averaged Loss: 0.7283
+Epoch: 066/100 | Batch 0007/0010 | Averaged Loss: 0.7767
+Epoch: 066/100 | Batch 0008/0010 | Averaged Loss: 0.6939
+Epoch: 066/100 | Batch 0009/0010 | Averaged Loss: 0.7401
+Epoch: 066/100 | Train: 74.72% | Validation: 67.92%
+Time elapsed: 2.4831581115722656 min
+Epoch: 067/100 | Batch 0000/0010 | Averaged Loss: 0.7466
+Epoch: 067/100 | Batch 0001/0010 | Averaged Loss: 0.7290
+Epoch: 067/100 | Batch 0002/0010 | Averaged Loss: 0.7164
+Epoch: 067/100 | Batch 0003/0010 | Averaged Loss: 0.7487
+Epoch: 067/100 | Batch 0004/0010 | Averaged Loss: 0.7226
+Epoch: 067/100 | Batch 0005/0010 | Averaged Loss: 0.8311
+Epoch: 067/100 | Batch 0006/0010 | Averaged Loss: 0.8524
+Epoch: 067/100 | Batch 0007/0010 | Averaged Loss: 1.1661
+Epoch: 067/100 | Batch 0008/0010 | Averaged Loss: 1.2247
+Epoch: 067/100 | Batch 0009/0010 | Averaged Loss: 1.3185
+Epoch: 067/100 | Train: 57.04% | Validation: 53.96%
+Time elapsed: 2.5201773643493652 min
+Epoch: 068/100 | Batch 0000/0010 | Averaged Loss: 1.1693
+Epoch: 068/100 | Batch 0001/0010 | Averaged Loss: 1.1391
+Epoch: 068/100 | Batch 0002/0010 | Averaged Loss: 1.0741
+Epoch: 068/100 | Batch 0003/0010 | Averaged Loss: 1.0773
+Epoch: 068/100 | Batch 0004/0010 | Averaged Loss: 1.0160
+Epoch: 068/100 | Batch 0005/0010 | Averaged Loss: 0.9928
+Epoch: 068/100 | Batch 0006/0010 | Averaged Loss: 0.9589
+Epoch: 068/100 | Batch 0007/0010 | Averaged Loss: 1.0029
+Epoch: 068/100 | Batch 0008/0010 | Averaged Loss: 0.9189
+Epoch: 068/100 | Batch 0009/0010 | Averaged Loss: 0.9380
+Epoch: 068/100 | Train: 70.72% | Validation: 66.63%
+Time elapsed: 2.557389259338379 min
+Epoch: 069/100 | Batch 0000/0010 | Averaged Loss: 0.8160
+Epoch: 069/100 | Batch 0001/0010 | Averaged Loss: 0.8742
+Epoch: 069/100 | Batch 0002/0010 | Averaged Loss: 0.8208
+Epoch: 069/100 | Batch 0003/0010 | Averaged Loss: 0.7874
+Epoch: 069/100 | Batch 0004/0010 | Averaged Loss: 0.7885
+Epoch: 069/100 | Batch 0005/0010 | Averaged Loss: 0.7796
+Epoch: 069/100 | Batch 0006/0010 | Averaged Loss: 0.7924
+Epoch: 069/100 | Batch 0007/0010 | Averaged Loss: 0.7689
+Epoch: 069/100 | Batch 0008/0010 | Averaged Loss: 0.7854
+Epoch: 069/100 | Batch 0009/0010 | Averaged Loss: 0.7456
+Epoch: 069/100 | Train: 75.07% | Validation: 68.53%
+Time elapsed: 2.594539165496826 min
+Epoch: 070/100 | Batch 0000/0010 | Averaged Loss: 0.7381
+Epoch: 070/100 | Batch 0001/0010 | Averaged Loss: 0.7555
+Epoch: 070/100 | Batch 0002/0010 | Averaged Loss: 0.7245
+Epoch: 070/100 | Batch 0003/0010 | Averaged Loss: 0.6775
+Epoch: 070/100 | Batch 0004/0010 | Averaged Loss: 0.7283
+Epoch: 070/100 | Batch 0005/0010 | Averaged Loss: 0.6752
+Epoch: 070/100 | Batch 0006/0010 | Averaged Loss: 0.6822
+Epoch: 070/100 | Batch 0007/0010 | Averaged Loss: 0.6636
+Epoch: 070/100 | Batch 0008/0010 | Averaged Loss: 0.6643
+Epoch: 070/100 | Batch 0009/0010 | Averaged Loss: 0.6751
+Epoch: 070/100 | Train: 78.29% | Validation: 71.41%
+Time elapsed: 2.63155460357666 min
+Epoch: 071/100 | Batch 0000/0010 | Averaged Loss: 0.6698
+Epoch: 071/100 | Batch 0001/0010 | Averaged Loss: 0.6305
+Epoch: 071/100 | Batch 0002/0010 | Averaged Loss: 0.6316
+Epoch: 071/100 | Batch 0003/0010 | Averaged Loss: 0.6012
+Epoch: 071/100 | Batch 0004/0010 | Averaged Loss: 0.6093
+Epoch: 071/100 | Batch 0005/0010 | Averaged Loss: 0.6307
+Epoch: 071/100 | Batch 0006/0010 | Averaged Loss: 0.6269
+Epoch: 071/100 | Batch 0007/0010 | Averaged Loss: 0.6257
+Epoch: 071/100 | Batch 0008/0010 | Averaged Loss: 0.6037
+Epoch: 071/100 | Batch 0009/0010 | Averaged Loss: 0.6197
+Epoch: 071/100 | Train: 79.74% | Validation: 72.44%
+Time elapsed: 2.668581485748291 min
+Epoch: 072/100 | Batch 0000/0010 | Averaged Loss: 0.5943
+Epoch: 072/100 | Batch 0001/0010 | Averaged Loss: 0.5493
+Epoch: 072/100 | Batch 0002/0010 | Averaged Loss: 0.6114
+Epoch: 072/100 | Batch 0003/0010 | Averaged Loss: 0.5600
+Epoch: 072/100 | Batch 0004/0010 | Averaged Loss: 0.5628
+Epoch: 072/100 | Batch 0005/0010 | Averaged Loss: 0.5680
+Epoch: 072/100 | Batch 0006/0010 | Averaged Loss: 0.5734
+Epoch: 072/100 | Batch 0007/0010 | Averaged Loss: 0.5477
+Epoch: 072/100 | Batch 0008/0010 | Averaged Loss: 0.5446
+Epoch: 072/100 | Batch 0009/0010 | Averaged Loss: 0.5952
+Epoch: 072/100 | Train: 81.04% | Validation: 72.22%
+Time elapsed: 2.7056498527526855 min
+Epoch: 073/100 | Batch 0000/0010 | Averaged Loss: 0.5715
+Epoch: 073/100 | Batch 0001/0010 | Averaged Loss: 0.5513
+Epoch: 073/100 | Batch 0002/0010 | Averaged Loss: 0.5309
+Epoch: 073/100 | Batch 0003/0010 | Averaged Loss: 0.5275
+Epoch: 073/100 | Batch 0004/0010 | Averaged Loss: 0.5178
+Epoch: 073/100 | Batch 0005/0010 | Averaged Loss: 0.5556
+Epoch: 073/100 | Batch 0006/0010 | Averaged Loss: 0.5512
+Epoch: 073/100 | Batch 0007/0010 | Averaged Loss: 0.5260
+Epoch: 073/100 | Batch 0008/0010 | Averaged Loss: 0.5221
+Epoch: 073/100 | Batch 0009/0010 | Averaged Loss: 0.5387
+Epoch: 073/100 | Train: 80.22% | Validation: 71.39%
+Time elapsed: 2.742748260498047 min
+Epoch: 074/100 | Batch 0000/0010 | Averaged Loss: 0.5616
+Epoch: 074/100 | Batch 0001/0010 | Averaged Loss: 0.5575
+Epoch: 074/100 | Batch 0002/0010 | Averaged Loss: 0.5630
+Epoch: 074/100 | Batch 0003/0010 | Averaged Loss: 0.6553
+Epoch: 074/100 | Batch 0004/0010 | Averaged Loss: 0.6026
+Epoch: 074/100 | Batch 0005/0010 | Averaged Loss: 0.6990
+Epoch: 074/100 | Batch 0006/0010 | Averaged Loss: 0.5912
+Epoch: 074/100 | Batch 0007/0010 | Averaged Loss: 0.7242
+Epoch: 074/100 | Batch 0008/0010 | Averaged Loss: 0.6661
+Epoch: 074/100 | Batch 0009/0010 | Averaged Loss: 0.7745
+Epoch: 074/100 | Train: 77.13% | Validation: 68.85%
+Time elapsed: 2.7799181938171387 min
+Epoch: 075/100 | Batch 0000/0010 | Averaged Loss: 0.6691
+Epoch: 075/100 | Batch 0001/0010 | Averaged Loss: 0.6417
+Epoch: 075/100 | Batch 0002/0010 | Averaged Loss: 0.6827
+Epoch: 075/100 | Batch 0003/0010 | Averaged Loss: 0.6241
+Epoch: 075/100 | Batch 0004/0010 | Averaged Loss: 0.6379
+Epoch: 075/100 | Batch 0005/0010 | Averaged Loss: 0.6377
+Epoch: 075/100 | Batch 0006/0010 | Averaged Loss: 0.6207
+Epoch: 075/100 | Batch 0007/0010 | Averaged Loss: 0.6038
+Epoch: 075/100 | Batch 0008/0010 | Averaged Loss: 0.6389
+Epoch: 075/100 | Batch 0009/0010 | Averaged Loss: 0.5700
+Epoch: 075/100 | Train: 81.15% | Validation: 71.85%
+Time elapsed: 2.8169336318969727 min
+Epoch: 076/100 | Batch 0000/0010 | Averaged Loss: 0.5672
+Epoch: 076/100 | Batch 0001/0010 | Averaged Loss: 0.5557
+Epoch: 076/100 | Batch 0002/0010 | Averaged Loss: 0.5355
+Epoch: 076/100 | Batch 0003/0010 | Averaged Loss: 0.5610
+Epoch: 076/100 | Batch 0004/0010 | Averaged Loss: 0.5928
+Epoch: 076/100 | Batch 0005/0010 | Averaged Loss: 0.5361
+Epoch: 076/100 | Batch 0006/0010 | Averaged Loss: 0.5148
+Epoch: 076/100 | Batch 0007/0010 | Averaged Loss: 0.5056
+Epoch: 076/100 | Batch 0008/0010 | Averaged Loss: 0.5175
+Epoch: 076/100 | Batch 0009/0010 | Averaged Loss: 0.5313
+Epoch: 076/100 | Train: 83.06% | Validation: 72.78%
+Time elapsed: 2.8540143966674805 min
+Epoch: 077/100 | Batch 0000/0010 | Averaged Loss: 0.4875
+Epoch: 077/100 | Batch 0001/0010 | Averaged Loss: 0.5315
+Epoch: 077/100 | Batch 0002/0010 | Averaged Loss: 0.5021
+Epoch: 077/100 | Batch 0003/0010 | Averaged Loss: 0.4712
+Epoch: 077/100 | Batch 0004/0010 | Averaged Loss: 0.4963
+Epoch: 077/100 | Batch 0005/0010 | Averaged Loss: 0.4687
+Epoch: 077/100 | Batch 0006/0010 | Averaged Loss: 0.4975
+Epoch: 077/100 | Batch 0007/0010 | Averaged Loss: 0.4987
+Epoch: 077/100 | Batch 0008/0010 | Averaged Loss: 0.4669
+Epoch: 077/100 | Batch 0009/0010 | Averaged Loss: 0.4616
+Epoch: 077/100 | Train: 85.78% | Validation: 74.10%
+Time elapsed: 2.891308069229126 min
+Epoch: 078/100 | Batch 0000/0010 | Averaged Loss: 0.4361
+Epoch: 078/100 | Batch 0001/0010 | Averaged Loss: 0.4382
+Epoch: 078/100 | Batch 0002/0010 | Averaged Loss: 0.4289
+Epoch: 078/100 | Batch 0003/0010 | Averaged Loss: 0.4674
+Epoch: 078/100 | Batch 0004/0010 | Averaged Loss: 0.4417
+Epoch: 078/100 | Batch 0005/0010 | Averaged Loss: 0.4830
+Epoch: 078/100 | Batch 0006/0010 | Averaged Loss: 0.4857
+Epoch: 078/100 | Batch 0007/0010 | Averaged Loss: 0.4713
+Epoch: 078/100 | Batch 0008/0010 | Averaged Loss: 0.4649
+Epoch: 078/100 | Batch 0009/0010 | Averaged Loss: 0.4857
+Epoch: 078/100 | Train: 84.69% | Validation: 72.58%
+Time elapsed: 2.928560733795166 min
+Epoch: 079/100 | Batch 0000/0010 | Averaged Loss: 0.4442
+Epoch: 079/100 | Batch 0001/0010 | Averaged Loss: 0.4304
+Epoch: 079/100 | Batch 0002/0010 | Averaged Loss: 0.4207
+Epoch: 079/100 | Batch 0003/0010 | Averaged Loss: 0.4033
+Epoch: 079/100 | Batch 0004/0010 | Averaged Loss: 0.4602
+Epoch: 079/100 | Batch 0005/0010 | Averaged Loss: 0.4312
+Epoch: 079/100 | Batch 0006/0010 | Averaged Loss: 0.4343
+Epoch: 079/100 | Batch 0007/0010 | Averaged Loss: 0.4215
+Epoch: 079/100 | Batch 0008/0010 | Averaged Loss: 0.4265
+Epoch: 079/100 | Batch 0009/0010 | Averaged Loss: 0.4466
+Epoch: 079/100 | Train: 86.64% | Validation: 73.29%
+Time elapsed: 2.9658455848693848 min
+Epoch: 080/100 | Batch 0000/0010 | Averaged Loss: 0.3877
+Epoch: 080/100 | Batch 0001/0010 | Averaged Loss: 0.3884
+Epoch: 080/100 | Batch 0002/0010 | Averaged Loss: 0.4278
+Epoch: 080/100 | Batch 0003/0010 | Averaged Loss: 0.4127
+Epoch: 080/100 | Batch 0004/0010 | Averaged Loss: 0.3890
+Epoch: 080/100 | Batch 0005/0010 | Averaged Loss: 0.3832
+Epoch: 080/100 | Batch 0006/0010 | Averaged Loss: 0.3828
+Epoch: 080/100 | Batch 0007/0010 | Averaged Loss: 0.4001
+Epoch: 080/100 | Batch 0008/0010 | Averaged Loss: 0.3845
+Epoch: 080/100 | Batch 0009/0010 | Averaged Loss: 0.3994
+Epoch: 080/100 | Train: 87.77% | Validation: 72.58%
+Time elapsed: 3.0028862953186035 min
+Epoch: 081/100 | Batch 0000/0010 | Averaged Loss: 0.3538
+Epoch: 081/100 | Batch 0001/0010 | Averaged Loss: 0.3311
+Epoch: 081/100 | Batch 0002/0010 | Averaged Loss: 0.3476
+Epoch: 081/100 | Batch 0003/0010 | Averaged Loss: 0.3646
+Epoch: 081/100 | Batch 0004/0010 | Averaged Loss: 0.3377
+Epoch: 081/100 | Batch 0005/0010 | Averaged Loss: 0.3526
+Epoch: 081/100 | Batch 0006/0010 | Averaged Loss: 0.3644
+Epoch: 081/100 | Batch 0007/0010 | Averaged Loss: 0.3720
+Epoch: 081/100 | Batch 0008/0010 | Averaged Loss: 0.3739
+Epoch: 081/100 | Batch 0009/0010 | Averaged Loss: 0.3732
+Epoch: 081/100 | Train: 87.72% | Validation: 72.53%
+Time elapsed: 3.039933681488037 min
+Epoch: 082/100 | Batch 0000/0010 | Averaged Loss: 0.3513
+Epoch: 082/100 | Batch 0001/0010 | Averaged Loss: 0.3426
+Epoch: 082/100 | Batch 0002/0010 | Averaged Loss: 0.3483
+Epoch: 082/100 | Batch 0003/0010 | Averaged Loss: 0.3466
+Epoch: 082/100 | Batch 0004/0010 | Averaged Loss: 0.3646
+Epoch: 082/100 | Batch 0005/0010 | Averaged Loss: 0.3525
+Epoch: 082/100 | Batch 0006/0010 | Averaged Loss: 0.4008
+Epoch: 082/100 | Batch 0007/0010 | Averaged Loss: 0.4343
+Epoch: 082/100 | Batch 0008/0010 | Averaged Loss: 0.4652
+Epoch: 082/100 | Batch 0009/0010 | Averaged Loss: 0.3787
+Epoch: 082/100 | Train: 87.39% | Validation: 72.19%
+Time elapsed: 3.0770065784454346 min
+Epoch: 083/100 | Batch 0000/0010 | Averaged Loss: 0.3671
+Epoch: 083/100 | Batch 0001/0010 | Averaged Loss: 0.3938
+Epoch: 083/100 | Batch 0002/0010 | Averaged Loss: 0.3932
+Epoch: 083/100 | Batch 0003/0010 | Averaged Loss: 0.4264
+Epoch: 083/100 | Batch 0004/0010 | Averaged Loss: 0.3648
+Epoch: 083/100 | Batch 0005/0010 | Averaged Loss: 0.3872
+Epoch: 083/100 | Batch 0006/0010 | Averaged Loss: 0.3558
+Epoch: 083/100 | Batch 0007/0010 | Averaged Loss: 0.3748
+Epoch: 083/100 | Batch 0008/0010 | Averaged Loss: 0.4111
+Epoch: 083/100 | Batch 0009/0010 | Averaged Loss: 0.3828
+Epoch: 083/100 | Train: 88.13% | Validation: 72.24%
+Time elapsed: 3.1140260696411133 min
+Epoch: 084/100 | Batch 0000/0010 | Averaged Loss: 0.3450
+Epoch: 084/100 | Batch 0001/0010 | Averaged Loss: 0.3299
+Epoch: 084/100 | Batch 0002/0010 | Averaged Loss: 0.3281
+Epoch: 084/100 | Batch 0003/0010 | Averaged Loss: 0.3430
+Epoch: 084/100 | Batch 0004/0010 | Averaged Loss: 0.3060
+Epoch: 084/100 | Batch 0005/0010 | Averaged Loss: 0.3437
+Epoch: 084/100 | Batch 0006/0010 | Averaged Loss: 0.3419
+Epoch: 084/100 | Batch 0007/0010 | Averaged Loss: 0.3185
+Epoch: 084/100 | Batch 0008/0010 | Averaged Loss: 0.3585
+Epoch: 084/100 | Batch 0009/0010 | Averaged Loss: 0.3339
+Epoch: 084/100 | Train: 90.16% | Validation: 73.32%
+Time elapsed: 3.151122570037842 min
+Epoch: 085/100 | Batch 0000/0010 | Averaged Loss: 0.2988
+Epoch: 085/100 | Batch 0001/0010 | Averaged Loss: 0.2991
+Epoch: 085/100 | Batch 0002/0010 | Averaged Loss: 0.2728
+Epoch: 085/100 | Batch 0003/0010 | Averaged Loss: 0.2848
+Epoch: 085/100 | Batch 0004/0010 | Averaged Loss: 0.2677
+Epoch: 085/100 | Batch 0005/0010 | Averaged Loss: 0.3430
+Epoch: 085/100 | Batch 0006/0010 | Averaged Loss: 0.4022
+Epoch: 085/100 | Batch 0007/0010 | Averaged Loss: 0.4334
+Epoch: 085/100 | Batch 0008/0010 | Averaged Loss: 0.3750
+Epoch: 085/100 | Batch 0009/0010 | Averaged Loss: 0.4696
+Epoch: 085/100 | Train: 89.83% | Validation: 72.51%
+Time elapsed: 3.1882712841033936 min
+Epoch: 086/100 | Batch 0000/0010 | Averaged Loss: 0.3066
+Epoch: 086/100 | Batch 0001/0010 | Averaged Loss: 0.4291
+Epoch: 086/100 | Batch 0002/0010 | Averaged Loss: 0.3915
+Epoch: 086/100 | Batch 0003/0010 | Averaged Loss: 0.3974
+Epoch: 086/100 | Batch 0004/0010 | Averaged Loss: 0.4041
+Epoch: 086/100 | Batch 0005/0010 | Averaged Loss: 0.4631
+Epoch: 086/100 | Batch 0006/0010 | Averaged Loss: 0.5558
+Epoch: 086/100 | Batch 0007/0010 | Averaged Loss: 0.4721
+Epoch: 086/100 | Batch 0008/0010 | Averaged Loss: 0.5072
+Epoch: 086/100 | Batch 0009/0010 | Averaged Loss: 0.4348
+Epoch: 086/100 | Train: 85.62% | Validation: 70.65%
+Time elapsed: 3.2253715991973877 min
+Epoch: 087/100 | Batch 0000/0010 | Averaged Loss: 0.4506
+Epoch: 087/100 | Batch 0001/0010 | Averaged Loss: 0.3861
+Epoch: 087/100 | Batch 0002/0010 | Averaged Loss: 0.3442
+Epoch: 087/100 | Batch 0003/0010 | Averaged Loss: 0.4027
+Epoch: 087/100 | Batch 0004/0010 | Averaged Loss: 0.3844
+Epoch: 087/100 | Batch 0005/0010 | Averaged Loss: 0.3664
+Epoch: 087/100 | Batch 0006/0010 | Averaged Loss: 0.4287
+Epoch: 087/100 | Batch 0007/0010 | Averaged Loss: 0.3634
+Epoch: 087/100 | Batch 0008/0010 | Averaged Loss: 0.3660
+Epoch: 087/100 | Batch 0009/0010 | Averaged Loss: 0.3966
+Epoch: 087/100 | Train: 89.67% | Validation: 72.85%
+Time elapsed: 3.262441635131836 min
+Epoch: 088/100 | Batch 0000/0010 | Averaged Loss: 0.3047
+Epoch: 088/100 | Batch 0001/0010 | Averaged Loss: 0.3191
+Epoch: 088/100 | Batch 0002/0010 | Averaged Loss: 0.3289
+Epoch: 088/100 | Batch 0003/0010 | Averaged Loss: 0.3065
+Epoch: 088/100 | Batch 0004/0010 | Averaged Loss: 0.3311
+Epoch: 088/100 | Batch 0005/0010 | Averaged Loss: 0.3302
+Epoch: 088/100 | Batch 0006/0010 | Averaged Loss: 0.2979
+Epoch: 088/100 | Batch 0007/0010 | Averaged Loss: 0.3485
+Epoch: 088/100 | Batch 0008/0010 | Averaged Loss: 0.3147
+Epoch: 088/100 | Batch 0009/0010 | Averaged Loss: 0.3021
+Epoch: 088/100 | Train: 90.70% | Validation: 74.15%
+Time elapsed: 3.299565315246582 min
+Epoch: 089/100 | Batch 0000/0010 | Averaged Loss: 0.2795
+Epoch: 089/100 | Batch 0001/0010 | Averaged Loss: 0.2665
+Epoch: 089/100 | Batch 0002/0010 | Averaged Loss: 0.2796
+Epoch: 089/100 | Batch 0003/0010 | Averaged Loss: 0.2786
+Epoch: 089/100 | Batch 0004/0010 | Averaged Loss: 0.2505
+Epoch: 089/100 | Batch 0005/0010 | Averaged Loss: 0.2761
+Epoch: 089/100 | Batch 0006/0010 | Averaged Loss: 0.2472
+Epoch: 089/100 | Batch 0007/0010 | Averaged Loss: 0.2624
+Epoch: 089/100 | Batch 0008/0010 | Averaged Loss: 0.2424
+Epoch: 089/100 | Batch 0009/0010 | Averaged Loss: 0.2542
+Epoch: 089/100 | Train: 92.46% | Validation: 73.36%
+Time elapsed: 3.336540460586548 min
+Epoch: 090/100 | Batch 0000/0010 | Averaged Loss: 0.2287
+Epoch: 090/100 | Batch 0001/0010 | Averaged Loss: 0.2415
+Epoch: 090/100 | Batch 0002/0010 | Averaged Loss: 0.2243
+Epoch: 090/100 | Batch 0003/0010 | Averaged Loss: 0.2219
+Epoch: 090/100 | Batch 0004/0010 | Averaged Loss: 0.2006
+Epoch: 090/100 | Batch 0005/0010 | Averaged Loss: 0.2049
+Epoch: 090/100 | Batch 0006/0010 | Averaged Loss: 0.2578
+Epoch: 090/100 | Batch 0007/0010 | Averaged Loss: 0.2237
+Epoch: 090/100 | Batch 0008/0010 | Averaged Loss: 0.2432
+Epoch: 090/100 | Batch 0009/0010 | Averaged Loss: 0.2232
+Epoch: 090/100 | Train: 93.52% | Validation: 73.00%
+Time elapsed: 3.373594284057617 min
+Epoch: 091/100 | Batch 0000/0010 | Averaged Loss: 0.1817
+Epoch: 091/100 | Batch 0001/0010 | Averaged Loss: 0.2408
+Epoch: 091/100 | Batch 0002/0010 | Averaged Loss: 0.2028
+Epoch: 091/100 | Batch 0003/0010 | Averaged Loss: 0.2039
+Epoch: 091/100 | Batch 0004/0010 | Averaged Loss: 0.2107
+Epoch: 091/100 | Batch 0005/0010 | Averaged Loss: 0.2059
+Epoch: 091/100 | Batch 0006/0010 | Averaged Loss: 0.2036
+Epoch: 091/100 | Batch 0007/0010 | Averaged Loss: 0.2242
+Epoch: 091/100 | Batch 0008/0010 | Averaged Loss: 0.2134
+Epoch: 091/100 | Batch 0009/0010 | Averaged Loss: 0.2380
+Epoch: 091/100 | Train: 94.09% | Validation: 73.36%
+Time elapsed: 3.410648822784424 min
+Epoch: 092/100 | Batch 0000/0010 | Averaged Loss: 0.1948
+Epoch: 092/100 | Batch 0001/0010 | Averaged Loss: 0.2010
+Epoch: 092/100 | Batch 0002/0010 | Averaged Loss: 0.1727
+Epoch: 092/100 | Batch 0003/0010 | Averaged Loss: 0.1864
+Epoch: 092/100 | Batch 0004/0010 | Averaged Loss: 0.2162
+Epoch: 092/100 | Batch 0005/0010 | Averaged Loss: 0.2011
+Epoch: 092/100 | Batch 0006/0010 | Averaged Loss: 0.1838
+Epoch: 092/100 | Batch 0007/0010 | Averaged Loss: 0.1769
+Epoch: 092/100 | Batch 0008/0010 | Averaged Loss: 0.2083
+Epoch: 092/100 | Batch 0009/0010 | Averaged Loss: 0.2140
+Epoch: 092/100 | Train: 94.70% | Validation: 73.93%
+Time elapsed: 3.4477267265319824 min
+Epoch: 093/100 | Batch 0000/0010 | Averaged Loss: 0.1623
+Epoch: 093/100 | Batch 0001/0010 | Averaged Loss: 0.1584
+Epoch: 093/100 | Batch 0002/0010 | Averaged Loss: 0.1628
+Epoch: 093/100 | Batch 0003/0010 | Averaged Loss: 0.1719
+Epoch: 093/100 | Batch 0004/0010 | Averaged Loss: 0.1881
+Epoch: 093/100 | Batch 0005/0010 | Averaged Loss: 0.1726
+Epoch: 093/100 | Batch 0006/0010 | Averaged Loss: 0.1727
+Epoch: 093/100 | Batch 0007/0010 | Averaged Loss: 0.1824
+Epoch: 093/100 | Batch 0008/0010 | Averaged Loss: 0.1995
+Epoch: 093/100 | Batch 0009/0010 | Averaged Loss: 0.1973
+Epoch: 093/100 | Train: 93.83% | Validation: 73.14%
+Time elapsed: 3.484672784805298 min
+Epoch: 094/100 | Batch 0000/0010 | Averaged Loss: 0.1935
+Epoch: 094/100 | Batch 0001/0010 | Averaged Loss: 0.1822
+Epoch: 094/100 | Batch 0002/0010 | Averaged Loss: 0.1766
+Epoch: 094/100 | Batch 0003/0010 | Averaged Loss: 0.1887
+Epoch: 094/100 | Batch 0004/0010 | Averaged Loss: 0.1877
+Epoch: 094/100 | Batch 0005/0010 | Averaged Loss: 0.1552
+Epoch: 094/100 | Batch 0006/0010 | Averaged Loss: 0.1665
+Epoch: 094/100 | Batch 0007/0010 | Averaged Loss: 0.1694
+Epoch: 094/100 | Batch 0008/0010 | Averaged Loss: 0.1534
+Epoch: 094/100 | Batch 0009/0010 | Averaged Loss: 0.2112
+Epoch: 094/100 | Train: 94.73% | Validation: 74.34%
+Time elapsed: 3.521725654602051 min
+Epoch: 095/100 | Batch 0000/0010 | Averaged Loss: 0.1704
+Epoch: 095/100 | Batch 0001/0010 | Averaged Loss: 0.1393
+Epoch: 095/100 | Batch 0002/0010 | Averaged Loss: 0.1707
+Epoch: 095/100 | Batch 0003/0010 | Averaged Loss: 0.1697
+Epoch: 095/100 | Batch 0004/0010 | Averaged Loss: 0.1640
+Epoch: 095/100 | Batch 0005/0010 | Averaged Loss: 0.1596
+Epoch: 095/100 | Batch 0006/0010 | Averaged Loss: 0.1730
+Epoch: 095/100 | Batch 0007/0010 | Averaged Loss: 0.1462
+Epoch: 095/100 | Batch 0008/0010 | Averaged Loss: 0.1724
+Epoch: 095/100 | Batch 0009/0010 | Averaged Loss: 0.1851
+Epoch: 095/100 | Train: 95.55% | Validation: 73.49%
+Time elapsed: 3.5588150024414062 min
+Epoch: 096/100 | Batch 0000/0010 | Averaged Loss: 0.1498
+Epoch: 096/100 | Batch 0001/0010 | Averaged Loss: 0.1416
+Epoch: 096/100 | Batch 0002/0010 | Averaged Loss: 0.1733
+Epoch: 096/100 | Batch 0003/0010 | Averaged Loss: 0.1498
+Epoch: 096/100 | Batch 0004/0010 | Averaged Loss: 0.1408
+Epoch: 096/100 | Batch 0005/0010 | Averaged Loss: 0.1489
+Epoch: 096/100 | Batch 0006/0010 | Averaged Loss: 0.1696
+Epoch: 096/100 | Batch 0007/0010 | Averaged Loss: 0.1597
+Epoch: 096/100 | Batch 0008/0010 | Averaged Loss: 0.1755
+Epoch: 096/100 | Batch 0009/0010 | Averaged Loss: 0.1607
+Epoch: 096/100 | Train: 94.95% | Validation: 73.29%
+Time elapsed: 3.5958807468414307 min
+Epoch: 097/100 | Batch 0000/0010 | Averaged Loss: 0.1701
+Epoch: 097/100 | Batch 0001/0010 | Averaged Loss: 0.1610
+Epoch: 097/100 | Batch 0002/0010 | Averaged Loss: 0.1478
+Epoch: 097/100 | Batch 0003/0010 | Averaged Loss: 0.1495
+Epoch: 097/100 | Batch 0004/0010 | Averaged Loss: 0.1741
+Epoch: 097/100 | Batch 0005/0010 | Averaged Loss: 0.1771
+Epoch: 097/100 | Batch 0006/0010 | Averaged Loss: 0.1584
+Epoch: 097/100 | Batch 0007/0010 | Averaged Loss: 0.1922
+Epoch: 097/100 | Batch 0008/0010 | Averaged Loss: 0.1790
+Epoch: 097/100 | Batch 0009/0010 | Averaged Loss: 0.1673
+Epoch: 097/100 | Train: 94.42% | Validation: 72.68%
+Time elapsed: 3.63291597366333 min
+Epoch: 098/100 | Batch 0000/0010 | Averaged Loss: 0.1566
+Epoch: 098/100 | Batch 0001/0010 | Averaged Loss: 0.1675
+Epoch: 098/100 | Batch 0002/0010 | Averaged Loss: 0.1674
+Epoch: 098/100 | Batch 0003/0010 | Averaged Loss: 0.1599
+Epoch: 098/100 | Batch 0004/0010 | Averaged Loss: 0.1712
+Epoch: 098/100 | Batch 0005/0010 | Averaged Loss: 0.1653
+Epoch: 098/100 | Batch 0006/0010 | Averaged Loss: 0.1663
+Epoch: 098/100 | Batch 0007/0010 | Averaged Loss: 0.1464
+Epoch: 098/100 | Batch 0008/0010 | Averaged Loss: 0.1448
+Epoch: 098/100 | Batch 0009/0010 | Averaged Loss: 0.1620
+Epoch: 098/100 | Train: 96.03% | Validation: 73.68%
+Time elapsed: 3.66998291015625 min
+Epoch: 099/100 | Batch 0000/0010 | Averaged Loss: 0.1118
+Epoch: 099/100 | Batch 0001/0010 | Averaged Loss: 0.1341
+Epoch: 099/100 | Batch 0002/0010 | Averaged Loss: 0.1263
+Epoch: 099/100 | Batch 0003/0010 | Averaged Loss: 0.1311
+Epoch: 099/100 | Batch 0004/0010 | Averaged Loss: 0.1337
+Epoch: 099/100 | Batch 0005/0010 | Averaged Loss: 0.1488
+Epoch: 099/100 | Batch 0006/0010 | Averaged Loss: 0.1249
+Epoch: 099/100 | Batch 0007/0010 | Averaged Loss: 0.1236
+Epoch: 099/100 | Batch 0008/0010 | Averaged Loss: 0.1485
+Epoch: 099/100 | Batch 0009/0010 | Averaged Loss: 0.1383
+Epoch: 099/100 | Train: 96.15% | Validation: 72.78%
+Time elapsed: 3.707111120223999 min
+Epoch: 100/100 | Batch 0000/0010 | Averaged Loss: 0.1181
+Epoch: 100/100 | Batch 0001/0010 | Averaged Loss: 0.1162
+Epoch: 100/100 | Batch 0002/0010 | Averaged Loss: 0.1256
+Epoch: 100/100 | Batch 0003/0010 | Averaged Loss: 0.1446
+Epoch: 100/100 | Batch 0004/0010 | Averaged Loss: 0.1378
+Epoch: 100/100 | Batch 0005/0010 | Averaged Loss: 0.1255
+Epoch: 100/100 | Batch 0006/0010 | Averaged Loss: 0.1290
+Epoch: 100/100 | Batch 0007/0010 | Averaged Loss: 0.1252
+Epoch: 100/100 | Batch 0008/0010 | Averaged Loss: 0.1524
+Epoch: 100/100 | Batch 0009/0010 | Averaged Loss: 0.1354
+Epoch: 100/100 | Train: 96.05% | Validation: 72.80%
+Time elapsed: 3.7441444396972656 min
+Total Training Time: 3.7441468238830566 min
+Test accuracy 71.97%
+
+============================= JOB FEEDBACK =============================
+
+NodeName=uc2n[509,515]
+Job ID: 25143894
+Cluster: uc2
+User/Group: xk5289/scc
+State: COMPLETED (exit code 0)
+Nodes: 2
+Cores per node: 80
+CPU Utilized: 01:04:18
+CPU Efficiency: 7.47% of 14:21:20 core-walltime
+Job Wall-clock time: 00:05:23
+Memory Utilized: 23.93 GB (estimated maximum)
+Memory Efficiency: 9.57% of 250.00 GB (125.00 GB/node)
diff --git a/5_dpnn/results/gpu_16/train_acc_16_gpu.pt b/5_dpnn/results/gpu_16/train_acc_16_gpu.pt
new file mode 100644
index 0000000000000000000000000000000000000000..a94eb90716532ec66bacf1ab633c93d41e441afe
GIT binary patch
literal 1796
zcma)-T}V@57{||?(-~CSO==$u%A7LinLGPXvR-EHXeaYHby|=+OUD=vopXi-;co0A
zBCIR%B1S?Vkh05?h+SBCX<ZeB42+N<KkkB{dfz?I#&j-x;q3jL|MNco=Q-y+=S-H=
zg&e2Vayg@dE8!wxZ(zjZ9USzStR8=8qJGF5@ft(JL8n8*U1uX&aAS<qi==Kk${a*e
zwUz)lyB0-wh+xtn)VK+RO!I7}S%N_UFxL`*Dha`)N`PW4oBB(KNUTe!*vg|2vTvbc
zo<k6nB!ny*s94r?pjeN^0PGtC;JlxKN+iv{nLzuwxRRtZRhsG;Q02lO1%Nyc0kjP0
z09wx}03?<KXcD6oXgZl!DA4qlZ&85OHcWRuOMoJJbO=*2Kzkn#Ai=(f#QsPDg}g}u
zTyhWq@k0`zy){mO78T>AsBDHnkveApIu7#?x;oBg+E|+$2e>@2O@XF&XcSO7ZI~rr
zn?z;vC~SfJ?lmY~CuI!%U^6xBI+JReLTnWjS7{t!b{p6HLPOX47%1I?ApqAx65z@-
z0q9yR2XHMb0Nsyege}bKaS%{?jwk@#ZE2-_7DLPE(>wMVlwQLm!b?5EP88wi2ErE|
zfLj|wxR2G{k@OmW97DsU2n%ecccbsr3j}Xkxu-?t)e6D{8ul%&qOz_>csqkIhd%uU
zcoqF62?+J8c?LeCEEQ{)PTc+eK+SRNkn_IVm7F2>-M)D{;u{SQ`NB?z9$xO$8~*>F
zU?36+`dH=Z=Jkpq+6O!GS1#v-+{(9n;qk!eh||FzxNf>D^C4FI;}d^x0j<l9{I#pO
zn%vq(_DQQB_k{!AVBn57Li_LphMbP7gSEn2%~ru;vzcv;f>q#6yv1bZQ=@?@e1pK7
z`38%<k#)IK7C6rKl^wFAV$Ej0yxIF!!2-KR-tX{Uo9s=f+-32VJ`g!JnJwSzlYK)|
z&s)aJP5&-(d~gHzx@2E)b;g0gr~O>^-wIhODl%a5X?0!3hB~B1dahN$`J@JBnpe!|
WDSII4U}0I(q|OStLXNG6wf_K8bZt@q
literal 0
HcmV?d00001
diff --git a/5_dpnn/results/gpu_16/valid_acc_16_gpu.pt b/5_dpnn/results/gpu_16/valid_acc_16_gpu.pt
new file mode 100644
index 0000000000000000000000000000000000000000..da25b4718c1cf024af49ef3c470a613b4ec19e9a
GIT binary patch
literal 1796
zcma)-O>7cD6vw9@wD=W0cqnaR(3X^HOLx0;ftcD6gbh&Zg4RZp2C!fyB`s2Hs=cfS
zjV2mB=*<}PqCIGO@y^YQhiW|dm54Fv#Y<yi5B1^<@9l<<Vmit2e*1rK-g_`RgBosW
zCB$qdb;Ck-k;OzNos1=>r(>cVo5@}qOeXRPM|L(7@S4a)KD6ektHiF*fxqPh3bplU
zoD|yQL+JjFAkHE5{a_J1Mtp=I{idKu%Nl^~8G<NqfdxAP2=Yh~6uAe%9YT=qa1RTX
zUQmGHJsN^ZS3Va(aZhOg$NdywWENp`OasM}<d!8o(L@301cEPBAZ#eq--~c^nF~zu
zJSYNCO9{ZY6bJC@{RGxEP?Xm?oGpP8d_n<2Z!`dRgj<%G`U?QQcSQi@Be%>IoUsBZ
zn{j~Pqy`Y~vjUv%FCzSk0~o;}dkK`WuNuI3j9Zpj6;*&U2PrB_c~_Y0!UibgSK{cp
zoJF{=A#9ZZB6fY%N&%HqD#9ync`ZFSv_(;Qc>o}Ccmv@Ju8MY{65Urq*QX+aOc6G-
z08w!b;kJNaQxT$tTIGlyvsaJVpT(GSy32~IA}b|y$r{34ebuB6qk4>?EK>{)>hO3C
zm6#R5XzbWSRAvhZ#@CVJTuC|9aJzZ^?ae>;S_t7&ZFsgjNq_y>zH~jGn$IOuxq#OW
zhr4{j|NoIm=ku8quY7Cdr=^Yc!Dr*j9mH8*c`=n+NYBp&yuyy_>Rm~@dF}U4XC|6i
zT|OJv?j?QowH^FQ8(c``((JCgk;t<?V(DbSYui~Xf-K3-VV6sCd7QFS5QSk;63S!9
zBMiA@w@Y@qB|+w0cCD^<+gXQvs^KQyta0;J?wf@>sz$?mxHe7YCUjP%_{Ls{x-`a?
z*XF7GK|8AQoM(R*b$M_DYqL~-;ohn&AGXV~{Z^>qwzhs;e5hr>NW(Iehi;EH!Sj*{
aEL+#k={bKP+2C$9Y$`vsl2*dwA@(2r!Cr9y
literal 0
HcmV?d00001
diff --git a/5_dpnn/results/gpu_4/loss_4_gpu.pt b/5_dpnn/results/gpu_4/loss_4_gpu.pt
new file mode 100644
index 0000000000000000000000000000000000000000..8f78de3e06836c705760e0215a224de497f7f759
GIT binary patch
literal 39596
zcmZVGWq20XxA*ZRxR&Bp7@QU;R@^qO#jQvg+@(;wz%DIbpg^(y3k(n;332xwH{zZ^
zm?T8-kl-QV?EANd>v{3?&H9}08oMP)uXFXBnNzD(H#4hU+wA}Qr-|8DW}CL|*s<(~
zWox(X?7eFFugkyRwsF&hG4;$A{l9;>#nNBQx=&;l8-o;i35iX0gn=&10Q-VT!YGY3
z!p22kuv|AvzJbk52Ga#MXeODiyu-l82D(u?bHc_jM%YjxC(b%zl#RPoqs%q1$(K{v
zxLr=u!JB&-*f>a`vPm_{+E9(k#%!Z(IBj6#WJas?3lsyj1YzA<qpa)1sH_?FrCc|#
z#!YTacVF9+5mtKZgw<UIVbw}bSiauCvS1lZXZ?IfBdi!C2=fbd!o2=+nvR(ppb=(t
z<Al*aDTH1P1)<AfqqO>)qnMY+Q!&rC6ja=Xs|4F-3PC%;3A=NRvRBur42?9(!oh+{
z2PdP{=p}=EFtgO-RLnk|5mf$NtP#XSD&|vLs8lXwbHeI2M)~`efinS&R(o#fCXD+0
z(2|?LEZR;HK(#B+sZ@KIrctrGqY`T86D;18aTJTY{(=xP)j&{d1Al(YXys+C5cqAK
zu%nwwn0|{BTIXwoYT+tSZFYo0#kQCcR;8(gt)n^NrzMQgbFWUAw^}2xrYfOZvLH15
z%|K@}PNh#EBed<J5T=dggmGg8VXdo5*jsF1>}G{Z?KuoE&xqlK;F=oY{WJzx{E#81
z>FsU3&%mhOI+X$cGQ!dfh0wdDQ6~2gDCX&FG=ldq1Ai}&L0-`O%0f=K{@N&i3^dBh
z+d8cx;swDmPfpWWzwOirmv;-o?qhPARx4hrgc^Se!q2M>tXghh<12=hS=kvKFdzDl
zM);?lAbkHxPSZ>N&iJ{on9zk$Nxng=Y8}RND$m;)crsE@Db8esv_O^MJkG!^bA^iL
zET`!fYo6+aKCX;VFM$K*K{Mqvy{`BiMtFBjA=vq8gx`lSLQII9rVCD4A_#jtjk2Pl
zfp7Y1w5qjC2h5{|f%86`O7kHCsCIfL6^lcyWH4QMLx>=(+r$WmR%(P<hd5!LYLpp=
zbt>IwaYCz48sY03Dxr(<N-f466jUap$!U7Y#r0Jx4c{pQi)=xtJx?XDdpcqIGDg^$
zVw4f%jWR${urj~jfmY^mUk+a2!LhRVyS*Un?<A+`mA2Zg0M$A@AXr?PY+!qRhLy$W
zdvcl%p8UN=IDecI=47gbjShk^{G)-!1sauZ6$~&h?9B;RPH6=DBXXMFgbM|n;PvG@
zR_zR}EIMXrSXsC=7lcq3W|SZUf%zJh{SW0dJ?hwCqfBe0Qu*eXPUw`Q0Ol_*sRW<>
zI$>X512e}6D)vzd;rF(T;3YHycUKAjy%m62!4CxU#;G!xURTSrjPTtZg4ydU8j9IF
z;}4lx*j$}TS{F_zwlLs!fKl<9XW)hLd6_+EuF~qZ@zZ3seSv~v-nu#m%pUdB0P}T@
zI+fQO1>xeiRLpz5H;R5!!^+(Jv7GYZ&FmKn!o|-<x$#M-@<%<RSjRFdjT;F<$WJQa
ztMzi4u5<OAPUvu(5thx7)3mDPA_(|um`bw=j7rl!nhB#mr=CF>_1XK1LM3C7Mo9M(
zgh=`bMtxp)%qWLvYqXlz%_xWHx8JDG?pcgh{Z3)U%#JMAO@*25_hW=^AsQj%D>+S%
z(RewhGP;gR<<fqQ%7V=bl~JEiMtwfpL{Mq@o)gZEQwi()YlMM|6@pt2Ipr;?Zf?`5
zJnl;6b9knKNfU&WJZ9f*6lOMO3oz>QjC|d+VrH%XQcM_C6+Bp@vc6nJ`8?71c{l4j
z!zk;6bz0daD?nAiZ7NlPW&-8&%H|xXDqO9Yz|59(WG0NNDl1Y<U}in$aVlXCHQ@92
z;d07jKC@1o(7{3{47{rVRiEFgRGQqB({#cAZgVQL@)at(S}{W69!@ytt`V}od?hx9
zGAeui<pj$Uf-pTzC3sZR2|v~2gjuKLlsBz>^5x+V1yiXS>c+6DN;6Oy6RS{Bj5oaM
zXr4}G>tsRbU6TQo*%#zAy_}wxjPlOOD1QtVv|8LnBlKy>fXX7{Q?7EVlEHNPz#4+k
zZ4sgJ<Chm$?+3Xt9dqg>2P*IPV1#|=6~cp48c_B9FDfC+ixKXwlGAj0@3k7C+HV4=
z{Hv{jCk=I!s;-w5pfYJAGfEqs$~b=m^>#5Taq|UXzOP1Tb4np>EtXT>^iz0SLgfrk
z6{Yg|XF*6#VuX)v<TM>z;ZCKhsfC78b!&p0a#cC(948D~uM^t#G0I@wK$l9nF<r+g
zj{%io`xU~5fmEutMR1g=+$$<jc}}77$>vXnQWY~p0G}tUB~)GfRR+^zjM}RLRm~1_
zDoVIc*yF+oJvMOQ)9slC(myIF6)pn>@G0pep|U|SqjJBRflYZbm|jIKGoy?iYhdLN
zL91Q8I8ZS>QU^YTlnS7-ePg3EJZ#|6U$pu(_9mfX+Gt}nJl0J9BkwhClCk>jcTVN%
z89C*u;>%n2X?90rb;O@x^=bSY1*mY@!hy=RV+o&z7crEI)uX9Yg#Bco)nBywSXQi?
znQrm%Q!f>$SUa2%zS>LpxX8F*MU(3qRv%9V>p(?^^KzOVzI&hse0=vip<?T9K_#*t
zq2laoMy1n3o$z3!Q5H7PP%1hu7li)xj56F#NBMYZi=3uM-KR02qC*pfaPguFe9UKp
z@HtBXD!#R0RDx~`Lcv>&5PVK2G<wH@4~GS#(rmRt*ndk-dD9QuH)vEA>?2gv{8I+G
zsu-$r;A8zJDq(M!0#uBcE~wmWq*94*!GI68=LjG7&NX1>qGI*2|6n;yuXN#A4tzND
zff4-c=)n6I169IQ<B!YxTXSgj;p9<9t1}(~_>gdz5%P3Scx7kclJVL<UhooFRkY7!
z!29wW3h*&un*cuaJ0Yj(-LP|J!29@vg0Rl`W%WMxyhf#HxPg%eILe1X%MI9ECA^=~
zQ*Pu1-yf_}2_L-~q01masGlpRJm#I{bd5^lDnY1~WWYlGQk><+bntiA7{S$313uvG
zG2N@^Ul~ltJnEv-s@hfscwcR~oTg(osf^HL9N}HpM*`*D&Bs*U|1poFykGW}PB=6|
z0Pps+QwW7eG~j*9FE<UICWGnjBTG1~+PW%KKK;Z9H~*H?bWHqr8t^u<50&@nPgIn5
zg=cl(?ZbYY@cE1Y-i@7Sphkd3WxFc_-af7p!28X;2yfr~Zs5jdj@8=}J9NVJ7J_i-
zq@1Sb`Cm&;Shz+7-VW+6r(C^_+@b;RI@Kb)Evms#-oA~a@~-YSD(`-GRIz&Z<1h|<
zF#AmaZ~X%qmE&FHlvjR#d#Mh*%gp3dyp7+_@4tCTtG5-CHLTvI{7&WF$o2x|T^nbG
z@YgjtO^+J=P!J9#QhC??iB4sWHI=tDE*W^&R>SJu-uXK4Hsid3D|ZEz9+ez;op)TN
z5>}rPvZ{38P2^e?cvtfuPUS*{PKfnX2_YXf;NA36osheM18=Ru7~$3%IZdzPuM|O8
zR!Dd=eUDCM-w+PGE;hbwZ+?rBL7w%-&iH%rdg=r!uW$UUVfA)FZw|a^@~cs%e-<dO
z@7q#&<1~w*yitumH?Kd3=_s!~|5OO!E{rhq8v{+da4Nq|(ttM|w+kvh-F4vYSNr8O
zy{_S@3h?H#@ninx0jKh|jqyc#TWy0p#Po7znd`vYZjU(?eF~$}ai>n@l=05L`JulI
z@~qc?GeKC|NhO^7n-glZHPG=XLwPeS*C@Tu>nN|o7HL5F^I@FIx)myw2uA^wIesgr
z>0a^8HA1%sg!1G?9Hl&ZwgBF&=&Av)#Skj5yMD!}EFY#2Mzs<^dGmn`D62?Sf!B{$
z8ZaLzP|7PGGQu1i9Vl}ORe<u+5Kh?Ch!N7)%V~O4_o0L`+nF4teCPuzuj_uTp_I*-
z!h!NDS5-npJR{8dAP8~k20ou4yuNVIz?*5B89umb!&IO=^ctgb`V^Jd5&H~S#tW><
z<L7grEPt5>yuQLzDhC=Vgz{EYUPoQhP|B_Q37~B30S>(0(MeAEO3QtpYJ_QD$_Z6L
zDLd(>(rVl@M)1EUr#!1{$7d>K8^=+3Rq3zM>SF~5%9l=KR2(kp1h@X2(0P-brt92{
z(FtBZQ7Nz0T}65I>|2F!?35s+xlk$dyh^2PMs;pxy3VUcL5%R|dj%-3^Ou~aTUh=p
zfY+BUaNy0z%L)~*I7X<POy$i!OP$KN)*LABrD(t#i+==RU7-rR9$14Dw(S)_nf)+@
z@RxD-S5I6uDrb5yp!{oJ1CCo{kWW}zwO<3uOY#};s`#j!riZs)#erA$gH@nxiIYzF
z)RM}pqwxxruvv1N&I<B3O8IUT<yG_yDz8~nfl_uOLQZ*b`RjPiD5C`M>Q#{flv~`O
zQr`ZU45lmZxyXU?AJ6Jk7M*5<$S4C#_Yz)R4B%Lmcm2*-o!+gZl==B6gn(i>O;@gI
z$$-+L#)Pt)f68E5eWeh}yRGFYufBOnC`)z`R5mWvfb!v`gjWUY6_i)suV+B1d1ns1
zntfAFdF9eA4H%)!P7rz=(1BOW{?>q(4HqgvspTn^&|^52myc=^N|VkQ@M|s1Ojmx@
zu7*bF_)1PwS#X5`r7_(#DmN2!Ld69ElumD_QEAwT5$w4DN>wL?N>^VwO&6@J&4JQQ
zHK~;Tmn5hxjn;ry6D{R59qjw3AdK^(Qg+>nQQ4pp$}W%9s7ya3fU^4S<&>wFeZ5%$
zN?UK!s7#y1fLEOl%PEg3jZM>lmz#q)m1C8H%H$B0Fd<$})4|2#7?pn4sk|IFicp$5
z!N7r@OiCX7PaT!uJxoqnUhLm2fLFiW(t%QK4B=J8DGlZ2xdj4v)i;R)r5EZm!W%ms
zc<B^tlsfY?Dp~*IKxytVDlay>GAeN=2rrL^%OD@)`I^=$@N&Fyub07{X;s>DF~{mz
z%1?wBA$c0g^PCq9D6RjkAVeFtcz$)74D!k^cP`<;bB8#ij9y0N#oj2DR<n#R{|k#3
z8cJ!M)0|dww=&?#F)N+Q_WCMeU>^>=7~Y#v$&EH}?h@f?(H%~!UOhE}<zT`y=R%J1
zY}hFUc$peO<z-VVL8Yje%Jaj<F|X>rmK)RSYSKagrByq0;DypvA@rQV2oCji;MsOF
z6?l1Yv;sU&Ss<spPHBemahGa&0_Ejz|0;wPwKU+xnS+dAR;m;BFV=wPV-_+h4&wzO
zFrNXXjUUJ<Z}BW<vj)6adW6c;ukYw6FAE<l!1E(x<TSmmrBgVSo(BjozO&U(p7oxu
z0*}jtpmNY&PI=Z-t)B`!Z#q;1p5&J@;OWobs)S~4geTiJ3M!kLX@ozAGT_O@I)tZn
z{WO%P{r}K`=VLTM#mV?3^Xy$5LwU9=hE`81?h{IzY?T|+YftN-055!kR6+wg0X&;J
zPfpV@HM%k2`OXeTY4J=!c^dV*MwpVN6DAE6z{~Dq2v0t>S5cmR{#OT{{klm5Ud(AO
zr|JDk^iqIl@n7pyVr@AUmoGnq=Q~k(v9!9t>glPjRGzzzltEtS$&?*bp8OF{<#G4r
z46DaGyD5Y#Pc-1^!5;<ir2HrYo^>2X<!PHugy&~mHCowQa^S_EEp^~&qxK4wy}MN4
zS*f`Io}N0#f#<fG3OwH9M|fhsT?YB6kFCxsKuKnl3OxJT_-*_&KboOD-Fu$XO1&t6
z$0u5AR8r2!DX;S~-jDEN&vX^#*^_5fo}YZ7puBA7DuCx3ZZqJSOIHCrw+he+?HoAp
zZ2o3BO&^L)s17{QzZO)gpOaIro_{Q7RJ<Q@;PH!=R7z~75T2Yard9E!lkyOGR`J##
z!s7xLf%1eKf6bp>TC1QuoxRprbx78z47|w*ZJkt!)!pTrH@iQzCy6f&jB?>}DW7`x
zbi(88VG7FQbt6^a$r4RY(=ls|_w3n0<2Og~Xh+jd{=pH=Rp9ZiV{)1fwrHUPPkJvl
zaNA7=xq4hVkV;A6D~-y+p;R6xl^IC=nq&1O`&R*!G)*Bq{id-|@=`cfPZy0<2$y~_
z%ERS4%43&6IZZb$87Y902oDB4tufgsbDB_joE>D~PfOLzbXLjvALKN_t}7!<HlDCJ
z<32}u+{pN$De(;vbj-?d1t@O%hyz7g#*awx-arMVq$*edPde9?Q{JN3T;)KCeU=Io
zjepGun-0indW^AK1yCIOkx;a_P@(d%shp<kl;0ziELh2?^yfNIQrJczq#YN)<Ld2n
z!stMiFx8wB#tfuV{A9X@QoPFeC`+0(rd7$9cAQq%&NATf4>bgp!#0GHmYrpg5B7NC
zYZWN@<vXK1{!KwC{%W>Hu!vzmQR&}u%Cm}MYjZ-^uLMvyVgZ#$@uv;EIHb@jyh<b3
zwqwAf%q4`P-o_8sqt9nKt^N)ZgvMrs;@;;7g?n!>tcqOM8wd_DRtJ&@#by!Q3~zXB
zS2<0e&d%p5Q1oPqMtHD>6SN(I&~AYW6nFO_6egb4QA(;BbD(I&2!(LyiVhUp%@aW3
z@Crs{XAYH;I?Xvs3A-w%=?h?SmrCKB-w01{zENm3{<fT^&*JziPPmXkrReWY8cM;#
z#VSx3+Cc!1f-fk9pZBYTO?~8)*LgI=_(B%=57$r%20JsL@J$E@@~wZ<fMO?mDuw>f
z3=~Jojp;h&Jv5-O&1Vi2jSUx6Honmbx}^dXy}TrV;#aQ;MTe7glt&5uI8Zd9jUen=
zs1mwuCOmpv!cYoBwh5qU_5q!+Y(FD(-eZ(s+i56;%Lj4bvE^PBD7v#nC%jrHr+gKK
z&wnBm$FwFCy;K#fibE%Iprm)cN?7o{oTjr@e$s*B%f@RjoW52;d1SVRP%vboPNn87
zIptYJ&4LYFOcf}FD}Q2yA%Ae7s9_xeJSw=#s9aAW6mK@ZkdHb{6j(iKV0>+geqASn
zyhX{=?^Rm)|11c<_E8A!zvqNW#+z9b@Dr8724RGPv;Qb&rt3U%31>iFXgQUF^$9Y_
zRl%44&BUV`9T`e~X}kamws}%14D}{FN=uO&(*;{>RDlBfWeg~4Y)&YwIn=<4435>K
zkNs7ka9XIK^0XPDu*6zJ$zR=8PI=`bGvhB~L4SKH`L``~tn#&T70?e1H%g7=GRU*?
zTm8%lo&60oxyPtna27yb)Gdv0@C227iz!BVe38@Y#4<wul(qup(V78N@-AFsD0+)O
z<uttskD97L!LX)+$}o31O{>eN6(C>#qymM%PUBSe$1tE^rW~e&3u4A=Kz`mz20Z$w
z6_tWj=Ts`?w**j-;Hm(H-US-)$g;j5{PP<J3X~BBqA$x}dVc~BP|4>vG?YiD`s+Z^
z&`BzlOSu9lnE8=PLFH_XO5<k&$Pc(Br|Fq1?x;ZC#`6NmZ>B3$3hgvP^eGk4>(yWc
zw;OWG)AifN8=mhw+E|^j*Rjg`_md#JnMEjA`&dEI9~5#RugYAb65hcm_lgBd-jOh)
zd_JU8xj2E4-*q>mBAzOQXC0^%ENHG$X??<ghw&B7FLcvsb?=x0<U1s&gm%R`VgDuu
z<k=O-X?lNNDLR$gn+cCL`WPj(Ca2XOJs416<*Wja`W~h7Xz64^-lC;CR(Z!K%PFt?
zu<{Ba&+@g3k~e1^;r@=x97P{sd=K?{skF+ioUdS&mlviH+P2mKz0ijdo^%#K?$Ob5
znm(pe|D&S6Y^PHhW&8;16BT3iXt#=0UXPU;kl$bem4ZJyF_ePy`5eg0{g+B!=b8$Y
zkN0%Kx{n%=zwH7e)QM6Ewu7kXgZpxnM@{C-DW5RER-6jx8LOz|O)@?U-S-(4y@eMY
zlYeMDlS!V{-uQvlPgN>Z9?sT)yskYs!QGXh7w%#xdczM?9+sM`D7jxvVt{V3!zi^2
zs5~70o51S+f?aaTo94QXF!1bKhVroPOAb8vF<nl1%!6@3goiJS1WKMg<A8qQ0V4$N
zGfGY4?zt(4s61F$VXSWaD$Gn5%-dU`0=fU2%mBTpNC0`mqg6ovI6((;tyT&uha+`D
zZR4AnJE$>3$(^uUPI+ZL&sh-eIdULB{L6=X^$x8H%(+6V2E!Pj+t1U1ylM|rDksbY
zkU!dv5q=n{0eWUVDi6~a6S5a~CFEq(Rm@C}ao=quCp?`er>UeBP<c@My-}>XsaWa1
zgwpE%l}5CB_@<V?D$8%64(PixRUr3@vz+pR4?Qylpf^6lfCo0?<TSkrF8`@OZo}_2
z;Gw1Q7`av_6qNe|`U@c2^E*cPn8ShGtTcu2xV``$?)aNg*|wSR;Eb1olGFPs2l9sB
zV8BD4|0#gJatPtU!!s(Cr*q|$Pnb9Osi2bWqXGB7H+~#+m+3N?Ud5hU0?1wb9hHZJ
zX6Pt-<3Bi%mD`v~mQRih@~o^jg9({d#_#Z~&wm@Mw;3GA<PI@jV8)IPw0f{|o`zM{
z-YEje*=hU|&RYGFRyhUMgbb$z9LGEeFk^&o25Uf;eO(pExj0t=^y(@h$0AIi<eaRl
z5jI_sQ$BNU&>=?H5TXM5;$~E`E2?CW$J`&VUj;H3PN0&P>?BYgZtX=S`$IZM$;`ZO
z6y8cf$x4dXf!zEQl~Cy~r@V50^I}eA#U35FpO+v2eONP2m{6PWaF0fK&~La=rc~F>
zOxMYYDOZ5|4J#OsyJMQ1a+PuOxdvp<sbWB;R*zHJc2x(mT4oC%^VTv#_R4iK$kTIs
zCTT$Cy|$dtEI<Xa<Bc!2KIsEPx&L^*0zCM}cwG-lM+%gjY7Vrz?^r<PzF%{WRqmnR
z2@js$Bjf~{=~(5o3|E2dQH|x4&vSo-@eyRV52li_vp*qw*knTPY2)3<_+y14lYD>H
zk0(5^v1TaQ`_~KL{t)9A(Sw`*oXVRLIZgM9{9ORLWkUvJw|cJwxgI4(vD`_>v0g?<
zKklHLnNH6xS;m3%E<5CuCF_E_APhLjf%F?gR3N)uHkFJ;)nqUoT(VsNIS;!rAftY`
zoN|@3{j@^3S;r{Wrv#OPCRDP|yeHhZH9pgf-=7()>{uox|F~SzG$3P3b2&{{&Z^{u
zb){4?!v`^x`&TcjK(_x>9eB_$Un4YC2ss@F$sn(D|GftXvb#SJggkc!WaV0GK#t25
zPPpSm<$hT+6(xJwAu1XF%wZ^5JAA2RO`N8q<i?-lK>D!`0!Xg7Ovtdcl|ep6X77a@
z$XFjrCAD(Aj*_Y-3BuVfa>|3#z84Hgum0R9&KFcFK1Vr_dGa<Pz2iQCl9k(5PI-EI
z;!zFAc>kjSat3e)q)phR09m(;du2Oq&`=(P)?h%!`Y{~H^v@B%{XrIloYuoRO0KVs
zobp~db&Y??$(Xc^O4jCk0;}wO%XLDw!h!o)2?EFpxT^rU&$BfkXR!|lGCOAJK<*af
zdzfRXYA6r7M93*`nlb4G2kv`^2r9A5H6VS$OAe$A86v3cZ>mt~+?z^vNHQV);aLr<
z^n3MHAoc4|0i;#^DW|+w=BL^Ukbe830Mg!^(1FaazEy#Yh~sj~gVS0z7E~HfGfGcu
zLi)RHv`YKm2t_7&<&J-AK=#!Ygse>CTbJ&YMkQlUwt{2QHytp_kh&b@ejS~Vaq+W2
z$<dapKzc?GDjDfV8A^6d<DJiZ5}=`^&)BN~nWGy~$?Oo%sWjS2$c_q=!SuTP90-~D
z#`}|5a|s=j@nV#UV{$I-ppx4B2SQpk<E9zi-4z^@-hZS3QoAmolKSltD(M%`saU1V
zUPmQ$dbEO)krAl@X~ESvp(0NJsn?9JdU}Wb3?+SMJr&4o<EjA}BaH8SYRP9A<jYC+
zZzzBimvBPTN*^61^?o9ilvevyl#E?96(D)_YdKBt#=rqOkXdg&A*b~Zg!Fa|8Lgss
z3LtIVejUiM8o&u1z864NljaKHBj-T+j>&SGz8l^14LHwHP|_@>YJ`Zv9LU($NC(mz
z_^5=FB~+5{of9ZYZe}Ww*7vZSrq>=_LM7#LEsm0MW{(0S_L?CGKg}hiE{Rf6l5D4O
zAjA1G15$QglT%(W#o9~<Qok8RCHc#L3zqinV~$nQ<o687^q8&z$p>>dkg?T|kQP)$
zNR5aSSf&0w)mW)BsbqAUVU&Ru@(_9V^w$q`AnRsz2Bbx-(tz|+{|X>^+JBr%!)gpj
zUEv^rjE`XkwteI%nJpH{DX*Ndb%H|Zb%Kx;Jy1hQZD;)6NQ>qwN=6l@Rr<L3R8q6<
z$c?;CYO|qKQttQAP!cX1KLzQwZ#YWwJV!=&e}YQN%}6SVUd9(Yxyc}Th`eCpmiG#f
zKKiDh(q=CMk^}2fNsawpK}p+Te5MH*-_a^Lezx4m>m;@MNe5E?OJaoMo8*+Mv~{Lw
z!o6P@kkoT22U6R9Afz93kwG4l+QLZzQjUb_gtK2hrcvH1N_tyGPSe5Ny)_{1!(j%b
z9{os254$ge>6owoB4mVxag@v?M-50F@Jt8N{7wiUbG<bK(p#<7fsCV5R3K?W8!BnP
z_GDDzR&pTu&RB&Im8k<scY;+Q@sbq>Qap`MCn^0KD#_+^1XjuQ8!{m2+lCq;J)ILy
zl~76O^rH;&<s|;zOaRI4HfTUfvdV#^zx`E0W-tR1!e{9~%GDPfNdA3@f!7fNCDo>y
z4kY_Wt3b-0@f=9o8exDBVJOMVhHHeA>o}0m<)obQ!ICX|3m|3cSflhDp`oNjT~vUS
zMpNXJ2WM1W(ty-{;{}zjF9_+2yctT$vZo4=x#6G=B=^+>!68FVd7T8G7AlZ*uN{@7
zr_O>((<(yp5hq%u6u5A#;?KUL68HF}KuOq8hXeOoRi~1?bTO60-V-&fk`wwE<>3h(
zC8?XMAXp#aKw?S~l|-{`RAO6a%Z=&%8T3&F5))brD(Tndl&chGXOuc&RFb|eRA@Ef
zFar`7)R5EklJE9afYetx3`l&sP!K$;$tjOXt#w%glE+#J!ia7hNUcni({xk&0}7C2
zeu@LBZ+Z|CgO4aEDW}sIkoeJ41)?na$|+Be>S)e@_|Izv5OeproN^W2(D)@BA8~-n
zy~*1IR`H94ft}YFO60ISLhR&AGRV_oUSx9M-r@NQ5S@5k014~M<dnxG&e%mIvC{$#
zC9cf4Y2x<NoXX<8IuPwYmP%~OV}X*``aB_~jqx!h-g-gDB)oa8;h1|3j9<cu$AcM4
z;tF?K#lM}xsk9tr;8-<*lDcgeA#u6!dn4{e9K$N5=1>(#9v;boxWCrODPK;?HBKdY
z=MfbpbwpPkNF3IekWw~+QK>VPN=%n!3d+6ZeKa7t>3<xEs}rCC;l1MtQ3VwOC1L#q
zLPX0<1tsZ3e+`KK;4P<oo_pRu3Lv)2exrQ(Px6zp{B^A2>djDr7^^1&xM$a!6TUgE
z0<k?x84#CxTu##mS9M>f5|OF^@$JtL5+mMgC^4%Z$SDtwpL~)NzDpz|=6;evt|EhO
z2q5Y7ZYoiKOkyaB#}27L{1YaC*sj$Xkg{3Pfux>$sU**ztx=i$flBJ!nH(kg<SGFq
zY+s-PNyj&9K&<;(Li9=FYZH6qq=Hq<^jDnl_!9$S*Ef;V^m4X;BY^1ko+=P^-1vhS
zUwfep@|cL>#><KOHcdl`xp9mGaSIkwi8|Pr5NAD6Zsb|_wj5M}*nS-a5Lf%EoN^Vj
z-BAUi#ti0!!=ZA@RpNkqf)LW00STYCs6fnvkA&D}Habee(;x=KRNumZxbg7<h+lGE
z0b<wuK_$U#nhf$5@l#W&#Ezb%quhJ;j0176Qwg!R(g{f=Z!}stBrqU;P>KLjJ{!-R
z{Nqe2(Kb(XtfC{iLU`C-5aKNf3G+K~l$hcDsl;@srJ^M0SE+=5xW-Xp8|Dy_SL~9(
z^w!n<k&x75fsS%7sxu)f;Dn%(SW1W~kKiaVCqoHQmWP>{={m7JRYJ_TS^_0;(ot2y
z^k2$E`p#F-;%fQ`LdjMtQF9t{l=#mF<dj!RcpFUR-pzhgl7^%c;v*sr^z$S{KR(N(
zCXR~xpK;sR=e0GI#9vR*D)P`jI!fGw4+4k|*+58`@{SP6_9|FKRa_<{<P9@QzX%<x
z7`3OI@?H@M$2nof4F!mtn4<zAqgv@e)b=A95dLAiobs%&f6WO|ZT1^w^H7de_#xv5
zB5eJy3?(f2qz*)Vbx2Nmdi33H42aAeq7xDt6XJJ{RZ;F8zC|UfLzWEktb3QsI1qoj
zu100?D>>yVVSg<HBL*rcv8$g^x!1cwM~S+4mP(?xx3O9^lvZ(0#)IAar5Teu`3HH%
z7FxwWnj%mVx7Ff6jAe-m#O3->iC;2XK}n3ySAm#+t>u*0Ne*ygK=c3`DlsJ|HI%5#
z?->w&f2aaP{`ZLk@tv0v!g40dAg>eY=%xbE|0XgZB6bxglpDXCVk+k<D3LV|%PG%_
zK6_9FqOKJQAmaKUPO$Q%68FwHD`LbJxiOvP*p>tF58pB%M$}M&*jG0-Ak5m95dOz{
zjuKneRt4hg92G$9YU5Qzb}C{hQEL@B<t<`1I1?f*90W>4=r>fN`>Y{^mA_(GMa?h{
zj<~jxRxv#{s#wKX4N!oD=v*C$5DPgFy>YApB=!r`fqTDj21K5?XJDA|hKC)PDX@xO
z@vRDkTgK~z&pIKZ^bA7@Jy9iq@cL6VAS~XV0l~@r6d=65l@5gdIGzEK-R}}YzNw?2
zgp8ger+hb}f`3(kXovke5b@s(0fa0uzO)JUL#c#kTXI^}uC4%4r4!^dUFXJD1Jl11
zC=oGs8W5H7P)>PFc)#Bma4#&B17Ura$|+aT2YV7Cdfrk|qE5^bKy0cn17iBDl2aZW
zTJ1*;M6YqtfUqT}8I?H+9Jn{kT?Jxmos?6a6=%PZ10i4jEC~P63?*{cWDN*bXQ@C$
z^L-o$tIX1Xkc0*b712gcdBKP-wRIqL_#IAV*9e8M?Wmmc7!fyu0pSD8bRf2$B7mq<
z`BXw~#B!9-eH%0&tW}{(rS6|P5Y^U+0}**A1Q7n!B?bf^zrv{mnG+)L9++P8nHdbL
zh`GmU75$g-i!r(}S7`NohJooBbWBuWj@-!8LpI*gfQUbObAm&4LfDvTGRR}1v=k~)
zzS{*#gzJ4BNSqf*C2W~FmFTv{uiW6-|6@2N%E|alBkui5#{|4dqE+b75AqPxV;uL;
zfM7A6O7O`PD!xxc8CL$<O%(`AvKBz7%}$jtc9ER8PN)Dx+$kzzvhfq=w{Dt>Rmeg+
zf~Z`;QNkO~HQ;$$26>&xZg~O-Grz}ykn?>Ap&rIpJ#xWRhE;G$0Ii~W50b(37!f@g
z5L)s=BfLDX1L1jJe$cEHmCEkAw2BNiK9mT{c5)-H6Bbm;fQZ;wDv?XmG?b8Tg9%X=
zCoz=haRUiqKmDSk1p2JgfKYZsC9HcRfDrdBa>`ryABfkeq|{Lf(>id1{Qw;ZGwY%N
zQUAOaK)?$hDiJ}|7)nH|ml_bbYO_%m1q+mj1O6Ncb6F#&ylLpH!-Uv%#$VXrXOlIo
zg1PY#gj!e9G2y!#aI7NxA7((1Z6ksh^+ceA)_SW00U<VWnr<=1MgxL})}RtHO<^cu
z7s5FZ5qewzA(h7%5avE!1wy~>%z>Dm#{>{L=Zy*kzHBL{>C*{`(}7Ukc;~~n7$0|}
zpC!jCsKX=$5L%fEg!Sr9CFJXJff8(Pyy4*?->E1ej>bD5(!%)V8Mf3*!73;ziveNP
zQmF|0XN2JJ%N(oV?V|(`vZ|(>@=-<XZ&XCsB?TokdOnq~W^N2coL{Q|QKKHKK!n5B
z90(o#TuynNh%f(J&%lP~1+7j@AOt12=_pYr6Eq-n>{>bHS>exS>Oe%yC>029SDOK$
z{#kO$V}eib6hKJv3mphA8_R)!C#wYzR<|t!{QE}9DbET_>_a8AVHXu8^g^XVXmC~l
zUS&f$AVOkQ!29`8g3s$o9K~nidpYF={qv10dvEe%D1HOJyy307&@llX7YRX!yy=+0
zg??Nn`6`MV7`T$BqWFI}$AA!rU=9Q{Hojwi-K=O8usu-6D)^hfsDuu=siB1Y*qBOi
zw(%Vcw74L!itw03EAh+k8cIN(@tZcJ!*(6TZ|)%u1g|hn`Gms5_yPpI^;fA}sG?QS
z*Si=>=<Hoo0$v>;gl%l4U=`8kunL4-dnbU<pHFE(^kg$a)S0gsO7O}(DiCs@h6V)G
z>CA!f$E{Q#=!b@KnjUOM5d*>&-r_*WqADsO)3<6UB560le_J<>613e)PSfc_q8ODe
zxjGR3L!g|d4<-H#2g25GSAdW{#_5q|hiT<^##_Zo)M&2(k@elFgx6ajgXuapYp8^-
zJRne_-X76_kR5##Kunv*2^Vh*AfkFNg^E`gA!6<m8RP{+vho=a++(v2gwDStr(A^&
zF@91)3d|Tv@IEgU5J&EEAY^3|9SFW#Lj^(ucXJ^8zPSbj|FBpEM8#k^<&`~b)+#_i
z`f3#jc3mr|Oy5org~2LHSmIg+c+bicfZy)na?0z3_o$-+!fi4Y|HQKlCB)lO079?S
z34=#yK%nP!0fcs%B&X?3n0rnIgk47k@LiX$1A%9jsDSW2CZ{~ud$$V%d~N;~gqag{
zAly@EKuAJ~0*J_F0`OnvA*VduH(&-8pOptSl%U47<upA;uR=n=j8O_o=%TAsLM=E)
z2@X8W0H4$&0`Tr22qETA3BFz<jPgHw#msc&;L1M;K92P@l+a`KRlqO%xdH_Ex(mV<
z<JV5`FAb^q{W(#?%6GvXIpqcYs~JD1;qxmQN~kqcfsnh#A0Y3{89Iu8aSILvmc`3y
zdK0eoWmLja6u|4Arc&A2LjdC2OB&$Qb{YpFV<#}c?`c0dO^?y*X9W;VjejQz9yvjv
z1XasX0pC?B2gJA5gpe)@A>fyPj8#~gZiWYoo1y{Xe-$%8934Uk|2srS3Aua7fZ2SG
z5_bQL0)&Lt(twEnZBwaKzd<ErYO+G*xbedn@_h<N2@IS?2tF9dQ2g3tQVFVXAcS{M
zpp}3BMAgjnsA8tS4uq{ADyOMDb{2%IzflQlZrnX^!c>M;z}8x{3i)@Kj^cjxEy3^i
zODan6;E@UtczdRt^2#Ai?F8UEtFI0O-+nKr>891cQ-HvoZ>acxKBc3CesI));DI>|
z@IP5b#lIkn5Kwl|SUK*c5^_FAn3=8|eC{s}1e_V906qo#1>oPaAqRZZuj@eA!v7I`
zPaKv(o*uU57lOx3;}eXG-=JdUU)OjZVU@^G0v(ocKz#X6={@f@S5W*nd?x^pF`sq7
zZ%G3=<t+kw&C~!l&u=*(ruZ|!eL%PXJO^y#fS-R~IptYC{)q(Vu=XlS;G}s12smVX
zOuqeGXce%w8o|5uXl`aY-G8$2H^uMycQVK#;?C)S_dHjEZ(%!u;y>~a4v6^98sK*~
zmP*j?y#<wp#^>%gYng%~{ys&kpv(>o#miOWfNx2N079mXWPs1YaUAe{tP8+1+*|?t
z0$n)ZJ7$6ggw!7-r+j8l2R{{Xbu^Cg9Nv+m1e`c00H?S`3gEpap5SRag;S|rC;->*
zx~hQRC0jY=mHo!GAb3yerlWWU*irF(R>V;Jzna8>pj*`l{`;b2kf(c(E>;1*Ba5ks
z;&_fCYR=RMZ7&gg_jDF0L1}L|5Zv!C4G6jNMyIms9~JQa@_%OU-ea+X;uC*D0|M^O
zGvG3op#-&cqaxzA2o(Q7&Vb<AT{OUVgS`Mk9CCCZ>_(nOCG@ij_+<=YRE|CtKtRX>
z4tO-p&;jpehpG7f*+vHWCIn}NGQe}+4TAT%sWQlwhZ>~-J}K8!!0p0LD*mmxj^dvk
zMhFbE5hy-CR1my#jVJWD_Cm!fkVoi%=ZYST@cNbp1Y5l3fX@*J26$XB9xP;SR}IDY
z=mY_{oikSf_X-a=<)ezLr3~;7Xv2ZvK63=%x66x)NPogmywb8&AS9-N0DLze;Z)u%
zry_>7Cj_oBzUt!RDtU;!S41Ztf=_M{NAdq5MFsp~#%h2VvO)(!V#g?edr2H4+?&q<
zuTCL?&}h7z^2&j`4iG%oq;nKc&uks=s4p1c>O4|Td9df0RRZv>w~7NEmS4#!R{{NA
z2trIS6`zYQ2?4Pd1mAz>=u#6$d70g!5|qD#p@h_kRRHmDfdD*;XH)T4t}v9K)~g7<
z4}T%}1QiIZf}PLFX?i)US`tEzb{8m~g@+gr^4BRj<uT&B)*9gb^Hc>0a2>>fP+O)`
zv3|q}wan!-y}${3jIz>7Me$7@C8u14EK>v^Jd8Ig<XMG|5_H6dRvxZTIg0DkXBrS(
z{c9@1zZpaE>eP|o|Ia!Z<dyx0{;UCBZ|3TN&&UlN@Ez1b1-xDM68!5P;wa8{j4zT`
z|F*R9-Pe*<?g0;}gzT%Ync;P5lLg?juuKPhuYY2I|BICZ@JbA%5-{OA6(y|ddjW|4
z({;cttCRylom>Ur|0GdP)BE#-@!^K7IY_I(I)inrf}^i+z<YbSobuq{dS^61cy;1H
zXxJ{J4C_cmeAiKK<iUZ@hp9l&lwpF(^aKw0|9GDPV)kl++pBE^ua=txR$lem=zwqD
zI6377-HT>$z&T<mqmt{Q0wO&}03HR#EA@7qqEPvGo#6A8ahCTb<KcZez147xf1D72
z@62QlIJW(u06vpX%W1lMX&wgxx+W2P`o}XAm)6EF&(O|)2^8=7b2;E?lcWM(3swoh
z_4#Ny<pqQMR&&6sW?P-`$VmkP##U*7=&_RnUjG}<0PmRA8W6JlIF+D=pA{6Z2GMfL
zEBnpf&jInYOa;8&{7ncP(UjnGHc4RRe#C-`=b>f_iszfr0`SNvG|HN39L2qQTY_is
zavjCjs=ffcTRt$#Amfd4|K}CM$|K&B5cJDsjuPy!Sp)oo8&dI18ZJ=WuDoVIz^<`6
z;Inj$3V4UbX@Fm^zEnK3hEVbReGtdW^V%U=d0ah8#oPKX1uLHq9|Yie`~U-j&iA6?
z9TP;wbwAgz@+;ZO2{n8OJ}vABA#KMKJVx!%%uH{%$DxZ1@LHLx13_tK8sPDBDZ!<~
zNFBxV<YE=@SyC#eJl*w~g+^GODFE*U(-pvN^d<(l)}O2a9*)L0!EJ82j^fs}1q0l#
zrD=fI_F0@@vswXsnmVh1dv;F&cx9C8K)~`E4DcCNhl*#fngYdxUEqNG&fOXiZ1GM3
zf{t!rfP2YBD&8M-D&qGqe<t2+({YSP_6v=$s+kJ-czaWE`?*X8`6hVvo<qgwo9zmU
zf0vq6ysGt7Q9SBA6oBidy*d@!h752Y{EGlwZ*-&Lx$_r};x*AlPI(L8M#dW+xYdrK
zxX;?50M4IU%PEg>*xy=E(E|yN6Yp{qkG7`?_RTM=D9#Zh7~tBWgo=Anv5w+3|E>U>
zcl;x#yt4g*5gc&uQcVY3N6eH{t~}kA5WHr$AUN+&RIu{7(OCt&rreWL9_*4}e1fhM
zUr=#~ZKPx6*4$PBJgz4by!ThhU^>g|E(1Jn{Ad*K83cFRY8qDVSDbXf_m}+w@UEMu
z0v>heF~IFchz58^+sG-e?7wrB4g|DZqyoNamV(OY0R#_6<EP1e@@2V^X9Y$M=73+s
zXccf~ikzlL-M?Q0To*1d%B|f3#bea>1h=<o48^f{BEjQwbB^MgRF~kiW4S=_UAjyI
zyjqQ6fLp}{f_LO58RSjf8!Y93bH74@Q<%L#aXUInPI-*aT2}^m*7`~SPF)l^<;o%R
zHo@cT)eI$Iv=;|l=6$UI9zPkc)ThrKT6y+$mK%Axb5bJ)_%8d21MYRt>OeqF9SsQT
zVb1~kJ5F-Sv%Igg(EyM2XF1^MU#J0IP4@`Ey_GuyJXe1*%6?BOZq7&LMqbBdzq<}N
ztZ%^q`_N@7;2yS?;9T^dhT^R56oA7JPbzNX-V&VsQsqWo&~d<84p?qZq~iR23Blus
zLKQ2Q3n!?!o!>@q@3=@{<vi#a!K*<7TKV^TN5#FTv2v|re1b0j)#36W`Ac|8JOkXW
zjMf3)1%4dx+W(6n<ek$AeiJ$1rfn5KP;KMq)P2_n6~(9L&m0JzypG@=JlO=(e<|a;
ze<hQYzqkH0ertNwun;JL_vUawR9+zX^f@Ji>772+TmWuMH|v0V<v$wWllxEwoEQHg
zr##p#*i!&bAGR{UrLM*azgEg=x}fhS1#nL3L&eYY8-e0I(}Mx-C7(Iq@tg5?&yyV!
zD89>2GQcJ8UmfshaY;^jW!Ig?A1>bsJp`?yT55n-^CX23pQaPe8PDu%(TSrtfB15p
z5jh%)XT3JG^4VNO@Gb~vSb5Z#Dgd8j`xL<Y@NN!xUdR=IPwkg-$_Mis|ECW4-@30-
z$-Y3v`NC-fKfmW#`E2W^0N$;)>VW6{nga0Z7^DEs5d}KnTIUc0oCfzJICjb)xUNkz
zR%LxPGt*;uW;PIj!?so&aJThV0Gn}|09?Hbb--Ksl~XAnN5yCHV}asqUm>TwphMR?
z46wiNApj@VOaYu<)R0pi<M_EL!6VX(qqu}TV1SK92RTj0%-_HP`<~S`Dj$YWaXURy
zMR96-O-^~R`{5G;@NE7}0lZ>1Fu=L<eh&DxvX@gH?7HNz0Q}?jDu8ELss`AtY%Bnu
zf2VUGa7jA`c)x6}Q%P8(0m1doQt@0lpP{(xgE-)M<Bgp1UcR9%Rf6|@D(=-M2o&$S
z4pcn*eNd<@F@6zx<;Bv<Wx{8ISKn7Ej`4dr)F_9HFRk<M-wCX2zM9JbyR`*Wg7>$U
zK|Y2{i<1iAdfkiQ(JPpvxVZF?Qy$|r{Vo+>D<^^C`EZ{G-1)1Libq*P8RWt4WtVgy
z=wt^K@c1T~0|EI51mLB&*8zwA4dj$(dAi&ZfcJo94Dk5(KZ4KlaU3OZ#ajjN{kyda
zIM1$RfdBpb3gCNsu>ibA|G)v~Ce{q_JvdkeTu$j4aQEj`a>~1V)V!br?lHYN;Be2H
zisSSQ8RRi8#StptaXFWYXWthb<xY>93~>4JhzdA1T4NO3R|4g3@IE=^b(}nE=z!Ob
z1`Kek(_c=xa;rU`VAr^bK>6=~<^<;f-_gn@?LEWFro@K>E)RZ^Q=WdOyAu_MhrjD6
zuFvZl`0kECaoJ-dr|IP+8DGCU3xaeM+e+g)?*H=?C=R_hDS-P4TMh8o-C0g~x`T5C
z2i$A@M{sM?hT!>An2MEa?Dqoj82NyTSIZYViu31Na?0!2xwoL=)vv3H;@W4a04%#M
zXMlIV8&q5aj1R>&(D<9?9`aeEW6B!KX?k5vjNdJHTe@)+cLzlP&cDCc0iTriR6LzM
zsJLA$WLSCj7)UFxmVJ#9Jxs$&eEp9CICZW;@V?{7Q5-v`tANXk77TFyv6ce(-7?nz
z_x%<Gw;3A2yIqdJ%15#0fR}#OK-mW!#iz!41#oU_{B>}f(N{(Bd*P!20r^23u<w>a
z#rb|8g8PBh23}oI%uMf(yIM;HT>CmvxjWVPZa9Z#>sYzXFk^uIiHRI=NzBj*TXt#$
z|6Nq>R#cb4^v>6rssYY3U8y+7B{LMau@M5WX?m3buKz_6oEB|SQSKbcq2ju_36;CC
z;{;ajo7!oB{l*|F9wSv1#mR1{0=T^Q(*f6d`&7Vs>U<6GwAv^D`;Q9=)~!1$D0lm=
z<$z~u1_L}SIujfQZx<-u$GXUAdVzyKP;t(>M{w)^Ud77wZ9@*Yyqu^2K311Cz;3Pa
zbKw-0$Wa{I#WKJ*?v?;-L(Hi-`Yn|~-s{fXivqA;VElFPYOFF8_sw(VG(C@UTmx)8
zlNey<nMlR1zVSZ0|FS^A%KE|vIptZ7Tb>KRHY1V))(aPEfZN$;RP5_*Cfs>;L||o^
zzE1%>s}^y<!T(#r-Jgu_{GBU<<VN1Y?ufk#`1vi@0GEG)2=?C?pSz1k6r<J5zd7J^
z=%faCPTfMqefK5>#dTO02Do1~9^UQr84bnu>lqwyzMe_&Y8*#!*|S`3<ny>MKTGh+
z_$*L-hc}_(vcdTMeCMq3?-H&HS}`2s*-ztuUHAJc;N03x0B*CtkyBpTX=6=-#}C64
zqudsN^T0zo;F#D;1Dv>}0Nkh9$|+BGIntB?w)09=!0BTY2VCdQq+;)yCQ!U@#4x~R
zZ#y~V>GmxrtAKakg&eS*JWT)|V@}FxdhHErQE^~3RTS5d>r`BeFUnxL&X_4G;I_?C
z0Cz2N6yU~;mke;4xSRtXH{%50Y0->|^N*DZic_yV6>w|s%&0v0S^<0p8D9Xq@tb6j
zH@$iCTLIX%@1g^4R@oe|Ij~WslCXndIfuz$ders5nI!*(!Mdvk*!(_72OLhisKA|6
zR{_{MCn$i+ppP7IHhW6NsrnX%a_8bXIpqcIsxEWDbxe^0*z|ucr(E6sF@*!p>&B{p
zeUnEFxU>B~0oaZW(*W1WM|9xM9$z`->9);(Q~;;<Cm3Maai##Q#t-L!{jwlA<-tx)
zO&H+Zq=y37g;fw-_nuHutXg@?DG#>%IE#vddj&&r{qo=MyY4v8Q5>dK6M*w?FR1uT
z7*EB1YG;DG^$)bNZ*<2@{v+RK>p)s*|6ZWt)^iKT%H>WU4X_(ljf(xR@_#m+<vQ*c
zf=g}V3+a6Hjljx%(RdxOUY5lOvr81fzTO%^i0!TbcWP8nan5SNQC!_0$Z5Jodn*la
z4iF5uv+W5LcguY`iq(bp47gLiO8|EHpXHRN-?5HW0q2gv46rlXOt>2qEl{k-*CW_i
zrsycHl_NP|?fgFh*vI`Nr@Wxum;b*Qws{vgibs!Ew7R?7_^o9VaX`n)>bMsNZ2vI6
z*w!v11d7ARHFBC>;Ft#-u-kf#0hUXX<&-PujsK{?-Co_PIJR@ss5tpjx%FT$;qJBi
z0xREHX&ku5<2Ar`@m(FTnm0fNymB8<ad{k0#bNYVg4;wZGx?AFm*hB&;1G6Apx8F-
zq5zJ8%^2YI^G^bB-a1GHY|YaZDto^YfNO`ga++T0&NCcvxnQOM?%$6kSn^60#XfHh
z!S&i~fnu4I!vI%*<9q1n-j#}Dr}qS_3610-@)p<s>TQ%A&j@#0^r6+=FaKZd-0qz>
z%8x_n7|Wgag-jDi-CePVR+f7nC@2nIO$k<)8*>!**&PIMH?a)^ZWoskY-i6RSb6T(
zuyRR#tpj!|zGr}Y8CL+e@0-ghZ|d->J_9WG7IG>-UL-gs?G;o42NIl*c`y{a$-5Q6
zG4L!W96Y2Ec6O9gUfC&gj0$*%t|ZupOyOpx)$K}O4X}9?&Hy*7nN*x#t`{iwmHl)=
zq4CwSdu2R^ed|jqRyLNQ0&t5ytN<Q=^`zqTTbK;;UbmXJH?TQFMX@_z!2rkM#&w)?
ztaTL2=}%O^+TM)=_ABxjU^B_Mh4n9IWH7yTEvwVY{o=0z#j36GxjWyyMdgmeA`L5N
zkM%lWSIbgP)BDreTm^1!x2B@)SVC|NY@=i4e&IO-?yi_6fLo&)DS(}Aww&@hj^C{y
z-0V9|pg6s8VZhD7e^7C6-GifenA^)~y8G*f0&qL)L$I%Nh2V7N2S%&Q-4wvR<4_%N
z{4bFcf{b_G>3RtjPq%DAtJ%$T!1b?#3<wxrBmnn~OLf34E?Ncr?U!)C?!-s|I22qa
zSPrSDptxRmp#ye}Lo~qQMOQiHtGH8lHo>ZFp@MR2!ar2*oL(hREPu^c0NcYesaX5$
zqH=TiBMqxt;e%DcO8tfdH#<e@fF`=hDev`P=13}5UA!2|jl4x1xbrku2X3{mDS$hF
zz0d${!D<Dty;_|Emu_VWz_#&lg5&Uw4CPW#ZxyhsH<1HD*MDJv{rNy0`0w_Q0<b;M
zLj&A1Z*sspeF_6a?OQr<W8gqJ<-=Rm_#glWr#T#OjG3zfE;}_1u+6W=fSdiSIAFb@
zrUKX;|5Z+T9lL`AIAKo&18f(&$!WU9q*N-_X<HN&=L<tPV6$(A0B*P3NN_H9=O{O?
z3kF=TY%Bm5m&tO<3)<H^LB+Z8365eJvzY;R`#C6p-NV}gaE-j80XA!<bKvHRwQ|bq
z+`RNF1FVmAAl$y)NJDYmxmr%s>l%7Z0bDMA`P#JUsG;1=x95O$k4+44UYDW*_Vb%7
zfMfa=Dn45~G8CtXX$03A#)sn3#e<6Vf~N%g%4mXJxlXWYv0FDYJ<r{OPXwzYa|O!H
z)i30ftE=v_2-mB1)ln>G+j77vaTo(^mvy6J!=ec;W5ZRfw72sGq54OH&8JB!%Izu(
zIpr;!uZ3%X%S)yJw;Hx$fZaotihV$Rg3FjfTDew!;$}E~iMO2cIxbg->xA)41+=PV
za>|wc5myDUvUy0w)w#WjV%KLj13bSreka)UR&*3wn{%{s{BtzHHm{4o%5ue3g1z-@
zj^eoeTLrMY^OTC~u(|@})|6l>c27>~D0idX8Q^q$A>ppyLXKh=K3N59R}avE+w(>d
ztakrSxVB@Zf|X<VGYz<IwNL===7Utg>iRbdaOcQ&DlSuh;3&6O|E2>jSNHv2Y4;u-
zMRmskd_^8ns}zW_dMZvTNO%(h38=*3SunN)p|wFF6ClNem_R@f<&*(R0ii;$CfPi8
z@9fTF-`OPuLD62YK|nk{0zR;|gHkP`f(S}{rFXx-i3dFGA8q%X<@d?=cW3UM$8PrI
zBnJirQ=QkKzyIeV5NAw}JgNewTd@J_u(LvA)O>1%1`n0d(r$ZRRIybktxg3Z1Dz9<
zm~wzotK1o!@MOzKto90O#b=Ksgy$|49kHLF_qYlqhAx+Y*ui%>5bsf>0+ChDdDbQz
zqa-wDoJ6gT36wP5F_n<Wb5<Dt<8LVm3~&*Q6+LZNa#F8<n+yb|FCoNTU)xA(?`#!S
zj0r5*Bm=RGE`;DuI$x)YoSWAFUJAn)Z*H0j81_9Jur_2Wgu`cLpuTj84Vd$0F+gvs
zkb&6#EjAF%n4<!|^v^^U%dnnx&Y`hdrB)+bDxnpe^ehLgu^*{`sreNka<Sb8jFly#
zCeQQD1_!K>|CE8yo>Id0+xig#8!w2C80??<r448u)f}j~HceE~@^yY{GSk~tBo=>$
z0ipBXQQ{9wv61-M%arKtnIaII>u;Zv0P}=S(C5xzNId0j3J~i2j~rqPM{p$ihC&HQ
zyis5S(M@MK(0HAjl6cE2l!V4Tr=k_=TTV#a-@%Y*U542}@Z(!V73+yjKQ0622eWM;
zT>5JT(C*tzNo3b78HttF&Vj^>_b5Pkw*&K|N)?HI<5~t7d+J3MQ?S)FDiAugg#kv#
zhc=-9b1MgGUhZ7{iop_+hBa;(FbjMt5G*(<s@S2q^GlhqVle|E3kGpu`?g~?5KGLI
z0i)t5!8|ZU1Y&cK{s$G%S3S)EO&@Jj@|m-h|M){Pk_NMv0(j@UIiU3)&Va^@R}~;q
zIFr!y#7Pl|De4Noqa<+k9)dRL0SPUuZI!5EjN4pG&>m^#NHq5p8Hj{dJJ^0kMWRpc
zNr+~yBADKlGFrj>4;7$kU7nM$5J$3O@*zU#SUy1?&t<e&ZXYN4e6Nkf@|6(6_y2_<
z*^%`bC83S8RZ7<Fk^uc`=N+=8ew2(P`d}LenmRsEfOy$HLg;7Coe&#-fuR+7q)}9H
z65>C0-o653%M~OE*T)=))!(E7){EaN02^K|0g0>nGeEn2qz!}~^vXbV(@QF_Go=p$
z5{1qQZ%ThpMq<_%h$?1pTDL?2TDGKdAlCU&#Jpym2*enx)wF@Yed!#C*S5)knRrap
z<WW?GDTy>JSCGU?HZZ{0*=_??da(lR$adb)Lh~m|NVN0c%79+gzyV*yBo*)vn$7`F
z&cg~Nr+ZTpIasG6F$O##0c_<BGSE<6%>m<q6j8+r{UwKGApB?NlZ?C2S&#AYej6<#
z<G4y=vaXc@b60}|7;nr|fX3P!LgL&}6^XY>p(OHig(C^BXqAA_i8>izXG1EWcZ{PX
zcC*EiG~Mcv0AJox8PJa{B80D-tRk^Ce<5n}JbPLcpe3VS1}yc*q9zY;@^dzjcq?B5
zLLZElfpDKwD$vsTd)%J4&T}M*L?&T}R3szuhhJ6z_vAw=V6eFo5HPAl6>~Lz@+cvC
z_h}UgOFzN@E!s|r{_e99lG>>QZNQp#JqL_=<pkf(uVf?*dNBhMc{4;6Q)r&~Dq!Xe
zq{KY-Fh}Bb^-uxL*H;G2)-?>UdR3~x&f<lH*j@Py$&Qq>Hqh*Po?r|gAtABa?p6S8
z^?p&s?6Dd7HV|4Gl>wu!kO7{B&g|j78$}=n8<P)kU{{%20fJ?3N`SY(xdi?L3uGj|
z10AA@t%Aq9Nr2V3g_8K**9ojgUurdOyO)VbeA8^^q9*4md|U;>2Vda;yJsi^tT|69
zgx9BwDh7K`WT-%><_MuS{XGRq_{CC!FLjnJlH_m7G~}c+LGc}a+GY+kyyv{WwQK)Q
z$)=XBDq5_$KmvRp#R<VZFDgj%_J>3j`|+G8vVlNqNCx~LJV$6=U%-&)gMFfk!RFeP
z5>Ov~R0g#5H*mtqW(f%WY`Lgnuwj3t0!^KNbs2BlE+a9Qe<uO4wL@&cce+#sLceeh
zI+1dgA+e4Plz|;HKC^*D&t^&zYnBj9sZB*oU-vmBk@0gll0df^3J`y_QdF_9NUs_h
z2%oPev`pNqB5AsjQ(|V%B*Z#@4v{$bB*z%5|2k?#$K4<yY5Ie66mfe9wGyW`6PmO3
zF^mZu`a31T*^6u>@q*O~5Z&LM1HrLdCBQm4Pz3_-_UAz4t8)bXfm#uWqwv45Q2~Na
zFXoh-tfwS4cf1J17+=XN1as4;42f0wE+yKh368|q_9GdHt=}!G*oyUSuz`Tvxs)+^
zt%5|qe~=0U*FP_+7;IkORR!WB6a~<y?6v`~^oR-=M+a~qcJmAwNPO661OEE=Rls-0
zc@7x;iwT~gPpC+u?>Lvqn43>6cE-8q-TVGVW9kl^R4_O`kWr$wKE#k{D<(>SIq<9u
zxTi1UK-0u*1*pppsDQaSi{P%QQjo;Emy0Tn)FY?ZfN^9b2izrJDujWDY@p$`dItE4
z`iPpm1S?k%*!7z@6823OLbX&$unwM4(egc$D*?K%*aoco4vQ-GW3)Q&+1gX%RU`qe
zL!s6US8z&h-zx(_%WG3|?>ih|;fV?m>{luSTIcU{=)De8;xDLG(TaZ_CdAUdRgka^
z8z?bKkK0K6?^j4b;*+V2lJpW$#ZvvJ&&hyy-+W497qcZK=AujmV7FY53F9S#-|O6L
z>?b}6t-9VPsTDr`1xFHkEzE$%(Yqud5Rb}$zx6Ezu$uouN%-nT97*+_%>>W1EES1&
z+>@e;rPl5GB?l5$xfFn1d6f!8E_!UhJTz2PG1%%Tl!3?{3lyN?$nQi=9`Y|kg!<S2
zp&-%!wp=3Q$~NFh-NXRxuKOwRdYzX8n_Z=%<@@3r21F<9Rw%jiBOCBea^Cd3sZna_
ztrtW`%pRV2)&`7{yA+^t(jR4@di?W*V9Rz9h{4{>Dh{w`TNvQ^G{pvtwYQ6!Je&ur
zB%p3*JqK!6IhRsfrYK0XrxwaU!>{X9N}8RM;2m}cC5>O65FN3@x|^CQF>_ZjB=PB=
zO2Cs1(^SBJ&2bJ`kLO6hPC1_t8N7ibsVY9L0=}(jGGGiZpd>VDD@WoF?oj}}pGN|M
zn}#vKd}o3Rgof;p0e_lvb%O3i9Eow=DN)6mYouce5Y3FsfN|4C6{yMW;$VN32*hCD
zH{~2)gPJL6)Sgq3gx}xGf%t1-8_)~9l!QE|7?R-A2Pm;FIyZ?~f7(XNnl@2XF@@>A
ziv!x%ODGAy^m~THmphN(cVBBG33j`{0pEf(GU2bzy<tAR*+vrSJDUU1e%}$y8Lz5H
zw6ejJ=x=W()Gof&Mk_GuH!7grS1PJl7(1HEfzUT;Hc<a~j1sH&T0*q@DMFL`CEJzU
zN+0744g}VSAbAeg{8R=aE6WI0QJaD!TDF&x#y~^_VzAdbr4ss|r$lRd!$zY2;~q+C
zZ=S=Du+R~O&~v;E_{y(Rf!LUyGN9$gD2YE_#*qYP43GeA`wI*Rz2)3dwFfUy;+{20
zK}(M~7p3XUHW7#whNu3VQ2pwgDw5jWaSnJNxyc5$o$pVH_&JRT#8!3PoV(XMsV}uy
z-fcEo=8dHa;K}n5{I6ulNP?x05j?5R%i&M=&f#cjV_zV!i^ViXyG~WnvNk+q1FWUH
zsA6_j`5Ol4J)e<*>I&!nX|S9<H%>gSh*t5i{Jb71zkO|A>zy?&SMr&kcV$<XD<wrV
zfAC0kO>mV}EL~bSs&L`rhlkHwxvaRNazSzBlnJ@6+5gw0tYq1;vf|`Q){lGt#^g~X
zpZ|F!y<N9ox{~F^l}k%1%BM`o{vX$ob#-#?*E|b<)kVykeE#RT(_O<aoqO1VqGd(H
zmlju+6qS`cTC_~8qp)PblnI0Xo287%&dMG&IwvP9XY8o#Q6n-(jLys&fyc;EV>5Gd
z#)yALkIosBT;-5m8MkJL6(*ngc~>MC`~B6r^g8xQHZFTya_y3f9=3Q<*$;Pc>FH0s
zY~wBB*0^K`-(UP6?%~pt*!!|Qq}5)w2j`;waQjOSFa5IZw_m>fzd!N$dEL7YaW?CF
t<<ReUAZ}?qI&1GM=<6|Vs$#nyN!?vXUnkr7dEM}W>gww1N{*+o{{ng!mYM(n
literal 0
HcmV?d00001
diff --git a/5_dpnn/results/gpu_4/results_gpu_4.pdf b/5_dpnn/results/gpu_4/results_gpu_4.pdf
new file mode 100644
index 0000000000000000000000000000000000000000..c98abc7d5ab4b1f3184bf406dcdf601409311a2b
GIT binary patch
literal 88265
zcmZVkbx<79v;Pk#gy0TAf(CaBuE8Aw1Sh!r;!bb~u8X_7yW8R}!DVrGUDzL=d++nq
z{noAd<5c(by!+KXRj1CJIaEKzC7IZmIgqF-*2v1+kl4sr$?T1+kOTzCSbljpn3A!G
z89E!<*qf8F{4_K-bs}Sb4^SZ!5<)VyGx<*-&;NQr(%#OQ?E8NOuxKh7S(zF;lX3sI
zDdFxcrRr?xY)Z!Q-;SS#&d#Qec4VCYF_BnQEeuU8?aaxz{^zcvy|JpPGnvkNT5-vD
z6jOI+G8P%zcLdS@8O8o*lp*{5zi`<8Zvp?|{tvyQ={wzj<FlxkI@!B88o$@`AO83J
z3Z^EOhNAZF?-^O&FE%y~4rW#!G7k3t8h*dG;k~wZ0Vguv|0Wc%v$KD<*!~m!zx(l@
z+yB1?%A4AmJ6n*k{ZCR!OPlwS$yg+9-uobKYHV*}`X6#9XGc>*TO{|4@lLrI`?2JO
zCE;xHbTw4h!Iv55Wqia|W7T-VVN_&els<zW?E0Hv?)}KMR4@H>FQ%`}vW0x+8A-K^
z!X!6>Qjm4O2Nvjs-z&@89`sh}?G4=hYVHSFpYnUbgC4~^U)kUKL67sjA%w4+es3kD
zP;ux(^&9PTGxR1#8}z6g{r0r!`*vgT<_A@e@qPnCuRx^G7gE26%|mo(Le?vdAB62C
zhS%!k)wqJR@q({9B;ERz*0{l^y9attxHcNo8>SEE9;<&%DzxEe>9+Rjl(D$<l+yT`
zQV6UaD!a+QxXO=lm0T`wj)6Y12)`Y~_&r>?K^5lQuE&H&V@fw6Q^LnF-|tusz<x6a
zQ2Xa%7rD;z)9)Olo8~svq*51FC#y;?0bLib;Oy5cVP6P*XBDUcQXZ3CJx}Qm0m8Qn
z7azrRLtmLTvnu)ZArH?0=+nC$D*SfdT}|rp){@`A#WDK$7z%xjS@(@(2US+ePbk0U
zh6@{K^4)`kA@+U{g{RdRzejFA2-xp(3A&VWCVYDXg-*JzOL*mHy!z1i1ViJW-oA&t
z2!Xr14`1f|_A{2${N{bMj}9X5X{bA2)h0Z3UZn`ZP0zMzU44Ed6;OhHpX|F${oB_g
zVW-8_`L&Eivoi$XrnwK~JvRFUgnw}CqF;Q(00E9aXS1z6XhB~kHnZQpn|H>%8fPt>
zFSnlyvz?q&TP-#g`c^E<y)p-O|6gGct?)gQPlY<PKvURHtzJlKZmzvTpD~E#QlI_A
zZ^&WO{Ph9P$M1P6<8Jvprqwg`U`?2A$@(e7)$fWF<}r4zJA1VH@~yVZDu)HSr~1jK
z_xK6032b^ZmhK<F5QB>Q5j-9EJrX_A+`8zZHoq*Sz6o98Jp$vpA;78JZUcQ#*YEc7
zxBD6WE&ZXlYkeWmTYwKi`@)^E<$>?|wnaS4<wI@v8&Ik9=JN}mYxfHBYx0ZFDa$2`
zvELI*wy|GtSF}&E;z37u%9Q^0|CRv0h!NgKU);QYc@w%{f?k(?>3Yw0{u;{lmbq!V
zsqqM`{SQO!e;76%KqL=udE>r$-I0(RVQ(}D_A~oA?W<Jk-o}O33j@SwxB9(0z5ire
zUyKpbp##7C``^UGI%beeEXhrpT?WY5KmE6{gMa!p)yA85`3j|f{!8Ki?`1Kbq7UcQ
zNZqvUt^S|dWL`hN(98MC9;^H560Getlu7jBt{s5vjH|4S_aPQ?zN&`l--@AE(7Oxh
zAoNP3kW84^pcT!rp;+>5Uj!NimH+YtW#786kblYc#GZN(e)<UgEanZ3-K<~C@R^9=
zbwN8=W_K_{2i}16fq-jXwMIPK#BZwiFS~xd(Aj0E>9FueC?B*NA_1Kvy#iMJu>FMH
z`B1m<ar)pbiMcmvPrb<Znnmss8yW%JdWrRed9y|Ag8o-99+SU3Zuq^AIxm(>(-)|=
zlE08@w`rDF7fUZ5gynaDeh38nvRdwPO#fE+71M77t6{;}w*be)ZkRrb>XAY4bLW`n
zVt7&yqzP94vfCNam*85dr<=K##cu|BD+QsFo8Dtnn@IUpmlGEB+L#rfydOhVo(pkA
z>e})<!h*2xX8-Yf&+?Y{K36?PBya5mX7K5xGX%y7D~a$_>UU`lE$bRz-j@RX;Qd}!
zSd>^wGxVq3Ax@`)e#4qNm41W}n&*?l%`7M{q}uP2_bsllM51a_`0d^M_#f}vyH~X`
zksA9ZgT1?R3mKBG{dA8ftbetd4e<``P}W+kGwkjL(?C}FyIu!@Zjfgh;Z2b|sit0`
z&CM(9w=?3+x1BM5-`9lfi<FfhiA{(96J1L+32f?n(D2Ud?%Y1P(bu~iBZUw_$4NUc
zm|1pLk$wB#{j2)k9)s0>*Q!dg1H`YPZ?8YS{r%<=-o+32eP7<iEyQ__tv1nz@E)7p
zyk8<%E*|fce7lT?T^QV!E+Vcz|Gy+)sddNa`v*5jke?T*x%=@tS|7T5z_0jPYu$bD
z()9`hy^pr{yV@h}ejAx`TchJ{PPzbH&^)gab_qfEK6gWUl?L^z1a9(tUe7-3Z*Ag0
z-$kDOZoa*Xr2SGA#OnXQ)z2z=9^={G{=X#CV!zXB{eVaB8XH$<EL|PoCE=&1XdmB1
zokf;_W90v*uK+erG{jW<i$}!&g{QQdvqAdXY{hML%fEXWPCx6-h|t>$0%gciHPL#_
zrw^3%n>mjE!ZQPOkrABwZfA|>bO3dx3#>@<d%NO>UN+Bq*14WntkjuL#dJT#_`Q52
zJGrsQUw=3Bp?BRcOVD%u*E5i?m5y)lX<y?*<r&!iEz}R1{WiBbY6J=Od)A-fI7=xy
z181+uR#;N_a{uTydDPlqlhYy+^T~SqO*3^^^-VfxlK*b;&uT_HytkEg0T@>4^-b?n
zSET_i%n3!a8(MqVAGc^9H+?~a*FGV|9*JWs*fmacs%%_~+7YiN?^Ob0me)%nFWU*_
zh~jn%hmbqtU1ZRllXM}x_jv~mT&kKYuzUB7*7P<huz}FEdP{rt9Du=ryx`kv@an9E
zaOF(wbNZ*#nK!KK2$dbS0$)6!_+<cT@sDkX+P@W<kJhNqXqwbI^@1*cZKgTp#@!ga
zB7+*KOuRudf`v=CmnDLDn(eGMa=GoK#N55w#oFpJk5(T+2%1La?5eZWXNgV4h7;Zl
zh!0MSHX%x_#+iRKQXRj-R+DPz%<Zd|6!>WO6dstL3|PY*a0;vnq48!){&8dKYPY1A
zP}lCMqd$?99j(w}x3TMav1<xcY?nmnS3O*Svq_`%06!OCe-9X6+Q$z$X<gdyFgP=M
z4G;o6H^!LdR4%b|4$T`b!i80z4}_g;A3cl7O>6gAM~xuPeB973c(~nIK})lJqDiWr
z<j%_P&l4G1*yTvC6Lncz%=FENUAYGEiODg6LPGDZv0vmZs#OL&HQ9}Dm)8Wha*cE7
zJ{8GtSNt&tDwMY-UsX&9gQLKp_~N4ayfa$AyVH1MQV-kDwH1MdPf;#J!vB1}#*POS
z^wu;|GSsv3t}AA%8w>?Zke($EJ@LGNit6^@Dnt$z=Clh{?K=gxv~H<$1IeB%!eu9~
z0{zEM-J<@u&7Du1SZ^ke?IUUKsH>iFTCJ8Y23ql$4{gWFm8}yw(AXpB-s$|kcj!r*
zV^;phQm*VZC9>GrIK(!6V9op~neAb+P3U<@X`Z_JrY85+Ibrvj-7Z{MbQO0s11t!$
zAKN5AWIUBrQ_y>lXwBGmFV|x&r^$lO-hj9B!qF>b1ans<$UKN$UR6)WR4?GXgvI{?
znErv+y1X}UH`zuir0$*}IYPfIxd$i9m=YWppES<QX-L=1S>A1|gqTfG!-f3)v_2x}
zQp1}2M3vXg6CA6OMX~kwK1w9o#3zyoQTs17tBGRfwK2>hN#AtsN`hi3jRB%VWt5-E
z`?|+_?>zGmeRDkq&8zD=QeC=?g?J(lYMA}|jGTa!dmyYK4}?KK!VRyEJ3~}jLBI-=
zX!(0Kx!g|9y*K&(TTVuZ#Ybx=rgSURR!%vzyc;zh59I1CBR~A)TSH0rd@ivL<$w@z
zjy#K;+YruTgVd1{EkVZL4$o-6cJlKR@z9--ug9o2PeQ!un+LCv^rYa9@4cl5$6OFJ
zRW2QRD6vj%bMY9p=qK7@mXw=*I@u`J1Q}cqrHKZQgrQpshXgTcCu8T<`tZT#TTIsl
zcqqtc@3#M9e?ix(Vd_Mc^H+*a83-n7Ftxw0Yaf!xB=Jk!<L|xjgtJnO@V(9Tffc}{
zTSul&#UV*CsFVbc<2VKu?%`tz0rMMv^71X~!L-{oqLHG@q?#_%v=En(bZRtH1~!&%
zQ3U(1Bu?<vWV-ceV2_rz9&WhU$PDL~)lbB8p;qgD<W6O#DTpw0U*_qu5VnoyDe(f<
z3_6T8Rpd^0J~vWOTgAuI-Gu`eNGdv~m5%j@YZC$wDP<&vo)!$-eG5%2l03Xn$6PIv
zmjjgdzPkqc%4_+f+4`R7Wlp%#5wSBi^2(yDQznsgP~OK3vR*&J(sH-vc?R2Q&2S2f
z4L$H52U6^B<`HXqf}IV@M*4*O6`7i2IH><-t$#eZpdS5ICe~(^<o#Uncc=N*>5FKO
z+li7Gl0IS{4@q5JgQTkt5<{n4-*vRq*IcRlvyG1@=1v<a#Z>U*df|J^$69nPl%5(r
zazo!&e!<1#ayws8O?lhJeCGEI@09IbVSXjZjW1W!ltwoW4aQ&e2$RRsqFPQPa<xQ6
z959UdS>x+mPd0n5DJX^SXnNnok-%=t85&FBFjF*)OuXRjKIS0#fy`zMjq0i}3*oNv
zW?ry-YvPNk`a&I+XoSDpUpcE1uMtfhqpiBgm$1wxms>`Io<;<Im*R@7V^*9N`Ci9r
zzY^#1+K^8`A5NQz<4ZY*>c`2y6X)NB0XTQ23;-XkgAz7G>3<I$&ij&q<P?uNp)6h^
zZ0O)gNT<z?kmzSdEsADpMh5+tVbyBE)_WP<Dk(6IA|iK3!ed6rvct?^JUt{D(#n@m
z7sWNJ?InhEIX*x&#bif$v)7t+pOwwt!JFGxm<)FJS9K0aOIGu&?cLzPDnkvfxR@~`
z1XD7Dd&$i$NGEY|^$KlFM*>Pnq^=7xv)q%`U5$u3oh##b$O%Nq6SVgwyZHgX>`?Pv
zEYg?KTGk?+#M(xQ(r&h|0S#(`UGg7W6~f2R^s@XoO2_AWIzf!_CJfp1pE<Fr;|rs=
zhAk45b<K9zS^g1v7k4Bjt^Vy4ED98Vp5y9M{{*COHZbzeL?RV)fY~Ji{MFW~deF1P
zP=emF@rTR$q1-;z>~T5bQcsRc=DUsLt0<C?TVnGv^PA~MxT`DaUhdvLOa$F-V3|u!
z-?h-DRla1I_k;hqbzoX{r}CUHqq!A~8odsf+O<CFgU8Uc?H`#gazL!?u^M*!!5YT?
zi>1YSxzs&(N&l8wDC-tvnHT$NwT4UW@QjwS9#1280R?MU-#;yp49rmyhFvXgJUXKA
zAU?mA6zEv(u;bISi%ns}4O_0dbK`+?!KsN{Iy?6}!j!xXc(V4Kh9lqCXmOQk8uqki
z63HIh0Zhk_1A9|13{-&HRrS6!3=}>VBdBhLLaTB=isdZAwak!LB}v^bk=kOAuWaAS
zCLZD(z`XV@%DVqBW%~@B_*tUhn*rT=6!HSq-@61A)}KoayXx+f%=_|vSbVs}X@LO=
zp4dQamstMwyqvp-j)JAW$(c)=e-AB+(7{Q=kO<1fep7O-XRb4sR9iuzEH|02^5$WP
zJNuXOJF=1kJjGxh$|@NFSR8H82=M{65G6TA{ikwCNjrs%%TGp9*)nh*T~~NXzGLu;
zUWMaXEzU&Ioe&<|WbF;>a1D9V1Rc>`4PUp~!!z1OT=lOlZ#LRES23K;2@a`g((<;W
z>-{jW3yBQ`WAw%OLs@vtOpFR(f;n`gc3AIY%5Sl~rUCy>J}!lyA<&|-41SRi-f}$R
z>WwUVjog*hq%Y%3nW@No_G0>mpGPo}<r}0%SqXSyEN*PY1B7I@$XA90%?6D?g=_7D
zPVD8<TVD(Q>FlWH5irQVRImXrJW}hE=T2!dO+t-&r~tav8)1zjATRLnR#5{A1chm)
zpHm}o?6@Hy^mjq#Z6FyIZrV_H+VVfLzVgJkzXPA(<T6aRZCPZopMLXoJ~=hEMjKP1
zR7ZIUhXM`tjIXJFkW+rDFs;upYnS~Hnp~=$D$f)_>)`@=`TCTX<Jxb<>;W<z_^@Y5
zyph@62qxb7U^$nwc;Pvvs;?*@Zolw1rX4GLb{tmxt9lFsn9zdq7GYVuS%%HOPv~b|
zgt3{F1MAuWIMQC$9}wq2Tw!lK9Vp-UYgznoY%g2Y>KxF(xTCo_0%uPH;7yui2W>aE
zZ``c-1KknhyeyX>Z@n=~U008=EhOH6IHSk)u{y_+`0vjO_^znkl^Z-eeTbjg5vU7x
z-R0bWKDNH9q+*#z4$xfn+3AwHh?XG(1^R_%h#C|ys0|5}KD|tk{?rlby5tR?Vh8;@
z>7-^p@>C)+=dGeyQm3!J7A6TTi&_IBlkOSlUp?vMM-5arH1MOKjeWm<w1+3Pep^dz
z5^lNIICx<Yfm;^1@Sxj|@+tJ2e-v~p;&)}7=kdh*L<_Lp0B_Kx3Or{_UCreG(-`(e
zgzD;G!G9JDD%0iN=C$QpokgT+suL&#kZxq)hpElLF`{4WV#0zODmidq+r7DiiZVK}
ztpa^=Wxzy^x0E-lin>Mmy?UY!^Y#6syQ?LhRaX!|HY!h}H*qnJ?*0!ZCcs2_A<nZ)
zc6_<0yGU7~Cs*U#fe(0^<p*=%1oN0s2H7@jR(P@W^C&Y3ILZlkkM}ES`;5z7Ut<9l
zoJ;V0Vg~P(-uAT~lfH8p1Dt*zd;Ut)L=JRQ^uwY4ZWpr=)9^^;SqB!-Jk;vetjRE&
z#~&3Iedw?J@YF%|{Ytp})xciF!w%%w&4;67tZmA@21lxz?d5vXY#4c)SW8_nQlO52
z4<VNdh%Q&Gxe@*30CSnbp;^-B<l0c6jkAlKOj|~mwPbHo@R;#ISW%3cS;N(Mjuq7r
ztq&e+12LVm`$4|6Hx!-D!kf<KqfF%G!q?iOJ`w96HIt+PdL$aEKUV>(Ibh%%^;4A$
zi9oktAg{1CA6CfH93x%gzH9$S8}`qy9k9-07k-(%5M6?83LEy5Mg^<DC9CHp_ZSx0
zT`xxD?SO=qjR6~VrLC!%8BTU-_i`f87m{-R4&ve7J>Vpg0$oZ`@gO);HI|FtO0{dQ
zWC%EUWEVcQ&%@e3hEisaU#f83Fmc689GupYSWSJ9PdpN>BdVr-U|;uHK<H54my|9-
z!mfY>6c=AfsQ(>cj1K=bM%D3BNcX~uZ`!+Xe22p=c1F9OM2;$-aox{4rC*98ep772
zLC!mgDn`1+WKJPp{xUjV#nYip_2<X257aiRVHJ9nj|vbYnCi#cm080(BLJ#+Dl+dq
zxh;MTEDM_L=>--6+U54$98N>$CPpIH<Tm<zOC0}48km(=D!u^DM0yn&&=m^$@8Hm_
zFIA{{OlO<I7kUrdCN<AjJn=&ekdNED7{`}4);{Mj3S|`l+W|>jRawC$d$5OW&Leo4
z$fJPv7v<OYulPYLe<|5UiKwxmJ*#kQcFL?fNCOBTNwmdA#*sd&j+)>;3wS^`LErHc
z>~e!Z@->)m<{OL#Xy12hwl-pgSS6ck7O{u#-n261MaXjpuU@L@kFXT3p}0u7g0bx#
zF)}kYcQbbK2-WhyR8mJ|3UkYbOR>IJ-FLWOis9qkF-^#rds#HNEI8}587XlGEDp=L
z!{6y+IQ*C@)KDDH>|7@o&G1>ccJ~7~(WkJ@!cC|hK*{MLdu9tSAq1W*M(6)lNDV0;
zc6L$q{mxFJSib}QA@$OCCXa>#h=}0bS_Bot_wU6qK5KM;6B9bzwR{fH)Ika}y#V4y
zI1WEhZZz7IP|b!EC`;||oh2=&az!o$E>YJUEEm{N1mNOBR#N)&#~_&i6~5BQfP{mQ
zi)Qg3vMn8A$Q+=_#zFg?%JZ&Q<I-GB2Ez$!)5qWlUrCRcG1QC|X7&$Mr(RhHipPay
z(<>ydgfA}{fy8Qpir=zav6c*zpmrTear_wO(46>Syh*)?ShPyFVJZdcQZJPn3c#!B
zUQU8)oWhs*s`}mnEAw&cj~fpikROr1%WfK<<J3Ot+bUc42n@Sim{Uz=kj*{5$vAQX
ze5j-0B1$@2E(rDQ7Yt`^C<!;s<2}0A2VUzJ$$%*>8=)-}bak@YNYQ3-*H?$vR65L6
z&SPgmqMc!UKel&}fm(fR`s)#2WFs*o1(RKo<!*@&qNC_*t0eR{{Nr3cE?K5gWOXM^
zB)2qhTo}#%{Bjc2|0ao~!dq+oRRwlEIXM$hfK3_Mw|q{y$JGglZ(q^p3@n^S9Zs+U
zMW`WOb^wOo<mRn6rM)vt5=zjPr3a5ajLxL<(^_x>*W{qg6J`-P<(JfPfrbSK#f)_h
zMEr}(iKq7+YG~(PdVMoSsuzR3LWoLFniPxPnERgn(zoQc1#i1vQ7Ea-Fiq+6X<Tew
ziUjDzo`yqU>6S;e^zOocA)v;_jS3my7X_mSMpZTJD-!a8#%EYeWj9tz0~w(;dPJ(T
z@fStfiHI@luw@?=Ubj4^+m5H^QIyQ=7;B-+zLf8;9VjzcG>d>fJ#F#6bL>(0ub2%y
zPAo1ijtr_^l;|bT1~NT&Z$q{9ULAO&bVr7mxD3B2w6*1Cny!Xu>i+(j$<k&IoFtSY
zFEnOT#`4c5Y0;%#ZVFQvtpVA1<glO}){^BaTqe(xd=sqk$L00f0fmD%3_e-jo0{BT
z4aH4_tYrkUcQ%xM9)qq?ptUjn#ien^fhZ+soO)+pspQV*7<{fgH$|>Di%pE5%a@_R
zDbK`DV9||-gC-2d`{QkU_RdbG1VvH)owxnx^0q3X7XR@2m!8(ZvA9m0aK_VhOS)t1
zOtc6!V|VVjpHoS5e`{9WT4;?kXhP0W2+Z8~ZJ8E8qV3ZK=>fHC8rhOl+atFd<aBdb
zd-_U89)7+NWRZe%0TE%p@u34z4+2hPQ>D<M{<IBi%#mp#2!>MrY>z_+5Q||{nJ5gt
zpY+ms1}8P4yW^n_l#Fy#VV7y^X=LiI)6xn7oor;w>zNKo-$U3cKqJ<nH5gS2;?yB^
zm*t2kjZC^bq56F$R{dvBS(}5wxD6Y|4QCJ9ys1I<1fhI!SlLQDrJDG>JL|Kb@it0Y
zclk%cGruHPI1pov{~iKn;_P*8!H6ASwvP^1$^^)`n%vkl$qsP=!g2T)C|4+x$$xMl
zR8N7+T;;`iH<+%<cq{dE2<{n(zV;^iX1Lofw+n0HlCxF9%?!7tYGE>~!pX%}q*e@a
z-;0#mvYe%HnA7;u(G%-w+omVwGwMpd73N-COe)d*>0r|SSzrPpVUB5X;tM?+Um;aA
zCgDQ;S|1hrCvjT&49FmXI4SFGsB0xr6T$;VS6U1YIRpZHH#J7hcL%IPrb8HPJ~Mq<
z#^hp;f~)TaOS*Sz?+#OUteC%U*S*n$2nbc+$tmj`6a}b}=0<1Sr}BH*Bw2|tkiP>)
zm)mHxjMuJ0;^<U(_73e~@Y%+Dc?V{DH5OUFNqy$Vy5v_pV`h*~6A5RPpky3LneU<Q
zkk&#|5*0I|?qK2aC#BKw+mvIW<C+j;PTHsD=;l%uQ9_^c@rh=l+cOg!H6)WtmJ3cO
zK=pvAi%!sq7Qn&{%KI`cUp)_{1~uJvq`!_mc~@?XFyGCOVd~juh~X#sCY}`tmrNq#
zez^}Uk~I4j{^9SXh(U?Y?xCC)5kMQpS+)qC6yXh?MU)7I0^qG^RYS{mxBV@7wq#FQ
zSN)G0+lJYN;WColBd<0s=)g|jKE>aht)oVzg5VC1x!=!#HfNG7g*O~~wU>-cz_X4K
zMb|zPA6Ap~8ImG5(9>{lX6Ia_h(J4gd83K3ZRc7S5YC;rd*ya~oAqy~Z|{6t>bu~q
zJij8-EGlB*(YKrpt6}3Z1Pmn4j1(R&nJr@bV4vP-dMDEz$(9Q8)_nk<xeR*7F+$}c
zSA}h=3;H2!QE0P@!tF6l$+W%@CC-vbDSZhFtCme4@xFzjjH0aVO7uDXYrz*~WXnNv
z@UNi?j>fFReBln<$Z~u=v!5O<T{{Nq;yB*H0mWpoP2viL{ln_!{z|)IcL$F>0^SOV
z@eX;4sa|3-)yDpTxbeBaHo(_+CG>U2bb^5z39i!7>D-JmjTwY0<b(0S({cBvB^^hS
zD(vkDRT3pf)QW-H=rZDqRh?IR`D~3FmRNhO#(1CEP;asj*sQM|&b;yxqfU|IR<c^6
zr_7$i2`rp~+bMPX$%<tf<%EZ}@#OXJB66^s+`p!@ef6Nxgr)SMR1^wbTH#`ZAqR1+
zgQg_thRe-eMk*#Exe=B!lzTC;xWvQ~Q2s{uhbxfdvG(D*YqM`^lq4{G&N}cP=*h}h
zl`N?gAEe?@8^(Xkxbe!z{@`sqByZp3?{POoV`<fIq#~D;%Y`!k7jTqa=uw#)L=z{D
z%E2T@g|+!G{Edv)M^F$853@fz?8bHdE>xQ(RbYTE!B2Fz9uq2PuRIxJvC@P;LvJ!*
zr=s9yFun8G-gXy7K`bE3Gn&`NiwvZPIUY@&sBA^t<0bxVq7ddPYZ=N$r@8oYcO|tJ
zveE-Mju_`w?KmH?3O@SD`guGm%!7sr$RlJClljA3^07C<S*|z3cCx=vqOAsgYM^Gz
zSkg{o4X?oI&T{da(#IBrUptWJVpo~eY4n!B;5V`;=81~Na5}P|{vWJJ-mt2Cz-z<h
z>BgdCcre**?>jOKoPQC*!YT%+dYJp7IV1@zVN|pja|1g?2FGOW4VL-HzAct4>nwXA
zF&F-fL&8C}a1ldb?qeY=Wg<+<OD6<{BJ?!2&>V@(+<GzWohE@|BHSo?#ABrL6~m}X
z#kJJ?ewAU-=MK}Q|N29<VHLqkjY5d!$_Xq2|7u|k=+AChV-Y|PX)Qd(u<SEUuEb@~
zDKzO@)G>jgNy_J6DkYmsZ+y|>BqW+dzm45u{dr74PCBTSkJ7rM)-;y0XqWh@abTu^
zA#0PJ@w$^(4oxXc>%^-P-u_AG?io-z*uKd+OP(CLV3~%MMUqvQk%6fra+8Mndvu;t
zi;1$j-+=#W2<~Jf0M+e<Qb;z%*w7~^Tohk?$9IVFVf!tp$L#?7eW_e!?-+J|d}(j5
znf`FWP%5Bfm}}9dbk1?acV=&!FKc;~BTNGj7fwdr3;gFJ2spB#{L24x|F+VesMWb4
z;wfdW;vvf|ckAkA7HuX8Me*2ip9iZm_w`6#oRu?%f`MT!E8_6B5rJ`Duwm_Std^Sb
z>+zj*mSBc&IvRImQ+~93iOjZBdq)oMps%o$enDqjGmI_b0cw4zN7v$~Oxi;)z36Y|
zUGdTsB*gEl)!&_L(Lo2PTCDA~IN}B>qDLXA;#8SPxEOnOjv0=jJloPcjcpK5^LPNJ
zR0j>MBV~Y&M~o6*WV2E$O5C_J6SlxOcDzR3ce%qL<6LPoc5a`@$knP6i5_b@SYmyG
z6-RD{&ss_9_`D&QNR;Y&2nLUx&suTYn(n#tV<GbG3=~9td+aI>E>AlSH4Z^cBky
zOauln#t6;l)@f(t_^$GHvAdb6Z{kma7HXD(YotLKQ`jF|DXh2Dv<>wx9HKUzh%5!m
z_<*45=X_nKYhkJ&>Ya!h^_?n4y)3vccAIF4woK01n|b&(=lh^JJnS^FtWgy63mR*{
zdQh0$txAB#lCF9&Dd;aJ9wVmsR#x-qwqETgZr7fG=qf#6o+*OLW8xBLGf0bIE5}d_
zUd=9U;3M^H7=HR?1c3{LBa<F8fdXT5xC_Bev~z$68MrUXUAOCl?V!*Bz8SEr;wr$B
z5;?_1I=sywoJC;28oKk^`;igrACz{7aIZ3w<^Le4{WDdM&G%#mh{Vj_Z+p)rr=-4X
z1A=RK5#&$1_$87)@RQCgwJc*5>+K6-$WB8HVy;DJd-#}6VQ0Z?KnVd#sU+}^9JI4|
z*DQ;>BXNdkL-~25A)ia))Y<OIz%oxIDB&=@uyEd=#G}Koim|j-uuaHO+&yvy0?t0q
zEb__Lmr?owb`PpKVKC3mRw?QZee{ShQ*!VqXZP;pq+3KiNv4KxRYHK{ra1qV*k{^r
z)W+U5p7b^RRUxmK$hJ^cgJHFRiuooQj;Wfj(DcGXH|<L~^O$u|j-%JtFMzksdGlaV
zH_!%Ixvphc--hR7hwVCHxpXO*Fgr)UF`=qUdhoO2sxG**Um@5dWLSXlt&rq&IQ1IX
zfkH*IevvuJMpf0lKy>YNa`TPr2n`QUZN>61*+U6Eo^w4$e<ek6C(WyA#g&gwOoqdX
z2ycKblDcspGn1!1Z0_rNsQnl-M6WW=fiHI|-CI7ppL-xrH8RB=<iz_{izRu|ad1U=
zv@Cr8MQ-4Wj6tMgT1usGtvpHBA*;ZfRszPXg0628thWulkJ2dTs0rE^YuuOlrfxuq
z0S{nTnmFkBAEk7BdeoxzXcqQ`<ByX@g!bYMoY;atx;~b@t$L%HkAH)Zx;4zqcRiO&
zX59<DR%#rEydaZ{{qk%VmIA(ON&!6US|sz>#Xyi5uB*5@TWg@tPv&7Mm2QIG&uzaO
z7bG(nJzP$@47nOJSEp1_lW4Tt<lnZ3v73j}DHXTBXp2H=zoiq{c%JONZ%D6E(pNqW
z&9FMly-fcE^J9@_j1tI(g<J#`!j!x<HtAITMA98XlPQT^<6pP!%H<I`mg4Yy#N!~)
zfRIgT0~7FIVtgTg4VttG+D$UT=V*R=8!KB(ASTJR%1HUT+nUOn^du_2Zbt*q8oGEM
z5D3%1R-aoY%n5B?M(}-Rd2<X@%F#JLPDo3^qqsV`3#SvJfZ?f#5s<B5TyZM2FtYqT
znA^`Gz+H!{IQjA1w4QYxzrtAZ3-xoemTQnY!wfc(5M4d~{^a2`a-RDicC_O;(J!iQ
zPb9^0=9$xn&Iz3$%QnJVv0bF(Hz6tWKlC1?=CiO@0x-YTkFO4-qfEXOmGf{(j1D_I
zWeW}r`v2^2HFK>iaQaJp2)8DJBe$R<NgK1Yz4Dw#iyuP^=#!V0YCC9-Hupx?wR&#S
zbVM=|9VKymT`2pI7P2@M$XmfZX`X9*O|9s6jkaA*j3c<Zwelbsgbex?RIw0XZDeDz
zX_EsBECzT(!XL&fYKr-lq^hq4L}ZsQ?TMJ`gO{43(nS|J`S=1BvZs<rTMawC${(mj
zMmv`X;hk{2kP0iZub=ch7~#D}#fg6T;1~D!X#WHJVNLSfaa0HHNdO2>J*VUfw?H~V
zCpZ}o9l9BIZI#;?tMZy8w?i4C;=@(pk1!oDeAkPn!^i0~Ve_iB7`Giwx+{bCjWrZq
z0F$cq0aNQ~2~+r-Ot&9+Xlw`A-;)%pvn~t3WXayWw{Ijs+!@NpRa07XuCyY{r>Nl}
z1w_%zGoM0}uhHqh697s(8&X)-zAU2jZanc7i(kMP?W5SRQk|56iw>=e1$zhfQ;C%n
znvzy<Y1<^RdOdu7YO~@JiXS%1db1O}3d|N-qy2@<+)s?nF36&4o8}@nb0h;L$&7%P
zSkt+~MuzQ6SZyD~y!yBfA4ZWz5k3^9to7pdY4Zyw)DnSFxz9o~uWw|30~l^G7BNGZ
zGF|O6d=O1^Vq?kt`iI#XI{J}|=sA5CT_zDf8o;}nl%=f*YGFnktbeK%%Dxyl32{*s
zbDSCo7x?xO0WS8zk0&C6n1jvH`fKH5JVHG<UoP|aKh>XK*I<(V#H{(aaN(g5yfRFw
zuDt$jGvK6UzPx<6s1Qc0tz&Z|YR{q>r$D)<2JLu@W3-yzVZ(l?dx^tGOUlRr)}G~q
z;M-xRz;~C^!V_l<y)E$Aiat}uMn-U$Y<y0^e+FLPCz9k(@-*Bdd*m4+5tS4S_2yC3
z_YG^HCU40`DQyd3X)NZ>Xn<xLWSu)YL+})be)V!`6CA2^gK0Hw*D+%{s*-K(*4$Eo
z*s=ZHXcQ~8V5F92ZNv*@YNCt5f~Y6dJM_08x5O9*|5Msy{*$%txPG5k5dC%<+ElF!
zle>H6udzN0W$<oSoo)*jQscu_67aSX5MP6QP81qvQIZj^p>GHNl&dw$b2zn%(c!c*
z5khtNCC8_0BlXk8RTPVPas|`T#0jPJjs`<?qnO_Ht-+_#pNtkrwXSBvP@a}~2Oa<T
zwleON%pZFwmthIHd}z7Px_yE&t2~xY76UTgM>R$l(`%5?73Nc_A`2I^-$)DxDFN_0
zu0cI9lh8m@LRG%q4ickpzH<eU2@`vaV*1NpQ9}D0nPNS9oWpZXr>`3{9lzrBMUoZK
z)&@U`qA1C7IU{`@N~0Oa!@_8Y63pm8E6&~>(d+#fut{eeC-A+V712#l_fyUw@%7T<
zw*`X*Aa<zMcRFLgy|#P(C9;xTo&|5Ck6A?Z^t{dU)<cU19%TGZuY#x|mzPoeH4v}W
z6s=76nMFK8*7#N?)K_G!Xs;G>n2!3Qbmjm>e=DMI+Syh+j?c8#Zt=jPgolobYDrB*
zaJ$(PtIOk<5Gg^JK_t3nXz4?Z56Ik&di#e8k7pJPI$fQBgLsp!#Yn{$_DQBIG~>Q2
zy@d7zNrr8CY4;)kmhnLp%jUdjR6NqN?FP@a$W;>UdZ~%l=FKd|E#}pR+@7eCF4q#P
z1btj_LS^8zNUZhW<>G4QU)*##Cp9HE9d%1$OFam=M=Z63yN}5Nxj~kOc-N`<)Ri3v
zzpXLK!<vN9^(n7J-2}>t5oV`Jwg%JcW-bV`_6rGN3?@$x$WG}iWqdc-_0b}WQP--r
z8mbTkQFv%?i8uVQx_b9ML}LaKh{wDAuug8#UDQ&PrpQ~jclawA`!v~Cm9z7$Q1uWM
zeX!Yrx++|b0{E52PWhl4YpG57c;_N+Jnh1Ry@-xLvQUitN`znKv8Zl3Q0^0v%%%|$
zlkcI^r3>!5#3bGQ7oP`3!F7Ub%*n_K{tv*WeNFA_Q^HnGN3q$$f9fRgqDjf_{Uu@8
z+|q231MPD5qTBn)pOYE*y(*Y?!yl-HW=7ZfB8YZn-1B3a23ZJ;-+%i0BQ0X`cNa5%
z0JaVOy>v0)4B#yI={$52EfFeNvNQm%H4Tnx4UG_hp<#*AOu0m@ID~bzlD<aEw`>!r
z-HREt%)Q`%P`uF$k=}Y~Aj6GsVFm|mNl#;PU6JoJVSy&3DwLB8*7-MXcKh8RRdUkX
z+u}5A)+Jd1nGTe`IX9Z>@Jup3!=yO#r|@dd84qhnm&mBMMPu>1VFyb;Sm)cT!^z@*
zl6Re$91S6vog+@4aX*uB^lG=RmEUrsK=c`YmFO<|Zm?3`__I+Myh&?DpjZ~1plRR|
z@;8?SchSis;)9k{mz~JGoKUw=syf01?2oZJg0No8sTGwW(Kn}(*>NqpS^)$bDS|go
zTYLoHKJm&HRy}CTqRD1V!RQ|l!Y7MVknIM;if4xF(=v%stJW@kqG{kWj47CRt73$8
zTG3Tm5smG;AyCKV$rNHR<*7|cZ<$1I)pW?Krt9X?Qgk0V8k_E)DMC`@MLFo+EE)7^
z!I>BmI^KUIc=9@%TliC|r9>e2Ns#HAZ{?&RAtE|<p9)5|%`hxdJ>H_Sa8-!RGE6aM
z)aek_me)~u0tv{<vGeuVK1FPz_bSoS*Kg<Ns19vTYYrX<T6Nr<%k5KzTAn+;?hxs2
z5+F@!6CJqIh#=mwC3ZhR#TODq&A!PrkKbZ~bJka-I>h121F$7jXh_>Spi)fbVUN8w
z<WYuX#tJwv%-$&$Rjwc^IoOy*QpBJWO737yeg-2Xe9_D1Ml)TzdI}Be4Q!mU=tn*C
z&R1U;MA0l;uTDBfk|NS2UxYVmK_Ta2KSf#aZ!r|gjLZKu&U?uEdQ~niMcaacJDUEF
z?2wyAq(?cGAyF8WqbV`7D7n~d)q<z<c4PEmh2t8rBo|>zJ9elSJ`dLCoR1XkqxaZB
z_~`tAArgoqPt0U)Fs%i$aP(qJi?hx@BdF_il7(1TDG2Z=YzS!iWScx(^XDW-G(@T%
zW8&ajjG>OE>%kiAXOM-Qt}LAdS$JD)<bt0Zu-mIzJ%XI;Qf>Uga-=-4dGjU_<{QkB
za(pWhOvg?D24)U|9$)8$xJw_a@;X~1;x#^vhJDpH?tou7iaRS8l?QYKtEAzoyHT(v
z-kQzYJM4cB`5jgx6wiF8RObT*=M_eIZZuFej*Qq`5(XtFFdy?0ojf7S*;qwHDrx6K
zZ$nv6L{j9z<7MeNR=b6?Jb`z$gGU4Es1f;nZYE_k(6x9$kFB72Hv|<Ve#gLsO?p<l
zw(Q+uLm%+4Ni`EJ4<*V$r-m!?uME4@qg11hn;1r}7xGo%nKS^qqbLyZ2d{qpNB%xa
z3zd~8-o%nPY{v+!&tBYy&e5i1cS!uti@l|?*4u#TYXH`y4&1uC1jbub*A<LpMeF_T
zfzIAN^95Z&JjYl+%j;2XNN2ht{2Jj0GbL_srmt&Qp<$R}e~)>O-;XmyHJJ>R)|8A-
z3+YDEJq7v?`X}TuQvjHI`lkxw9~R%X)V>Matyqn%<48`E8AkX{3rj)XCI@&q#8cNC
zFc%yq8ZZ~I5~%k|w9tvVhT~$geOJ7)+(VXXB<T+Xg?G8J6ppDAh&^`D;;LrEN^yPN
zA=>znIY`Zjl&u?%-nM@}N@cY_xwae5w(MidvOrf9j+Uf#FSNOFYgO%TvVgZNTw9dj
z5rm5YJgOubq2l&%9tT-WTSJir0rC>K$=Cc8&0ogAin<)L`k(s5>K7h`zqg*#V~Yfl
zP9%4TqX&3eSb2kdi^P9#{uwdpT&MT}%e)NmO!*;k!oG3iwSbXcboBBhSO?rfXyl=~
zTEH&$pDe<8l@R>=MJ>zxo7g%r;@-3j5nny-{bfM@33YTBE=o5xdkMAt6G{>0+1Tyb
z<iOsLw$OEl{Lz(zEDx}`Zda`Y!ShRUy8FcYbOeg~>ZWKzbMi0!mg?O_^U+5r&(KE_
zR2T6p!V6X*|4-Bkw($Nu5*z3Kq=Y}$a48~nl!X_quEy;9w}#^0^<OL_U}k(w3t)mU
z3Yi~ZspcA~&})S>+p`343L|``Gx;69xub~LoS0?8ny=T2;Tri}N}UWZ*hxOF8`S%u
z6LC(Yz5kTB3y<$gWWe0(8vZFEi${YkQB#~YX3VSdmPKdTwoAF_Y~}P!#gH2lm})Iy
zreux(+=2DOH)c*Y9Cbc-jH#DRVdlHS=n<P5m$8Db8Ovbwc@zo7Ggt4nYxxytbiS}|
zTcV!Ng98QUCQp3g{rq_4k{E42j=w+tfp<R-cFXcevi*@ie{nGb3>EBM+1i#=GM;wb
zfT2_m5<B(+nwhV|asw+cO-NkQyao4C)Izy7U@JCW47-n5FKquAZZ0{J%;K|9rgOWW
zTA{lUHS^hCq)SA_@!x}7suhnlgRS8xn*bxEelia7oLcN;b-^RNMxQ|C4{Jy67o&!*
z2R~YEmYXomZu!7bbsW|IB%PJ#8#lq%hF~4O3{)jiWhLn}is-J4NQbzY-%MBtQ5YV{
z%7IIEJ4650d=^8;{sd049EK>)p9HFM4s17HD#hzNq=sIY|IP&JYR!K(<@9Co{t5O$
zKWj-rK2=W1Hagy>uj||+WHKS248^WtVL(4Se%U6b4m-a(szzl9CY6!6Y242t;ddib
z_;IS*t&!!ISQ`MQU&HS*MMBdQh7I_&+BrYiv1Emz*yQ06z^}zqfDPIstQZ}xU}M#V
zas^n(+0K+APc~m@R-(q;NE0nyHEo(9Y$$Ta#$Hv|&XM80FH&syk)ekpJ03R=qV48?
zD^k^1CX%NLrj`=5izBbHLOE8rEoTRP^HwlB=BCOwO)uyW;PU+|G>ymy$znRfl}3k(
zg|Va0UmUD%5ZO7MlB6a->PmK-cdM{1xGH{<7y8tXQOlYE!&ybQv`Zm-)pyVQvaLJo
z?=I4_p5Bjv>58-1AS_*-#v@?oF_F?Sg6TD}v!=88EK~*klxc>EXo`F>=anWWAIqB}
z=<L%mW66V0Vp>@OiX>5_w>L)GqGLx%jsGM-ov_TA9U^C`)xoto*qZ^fA#G@LVfPna
zw4uj*W?2uw=eCi1yf5vMN)>5H3?fmM>-g?LMN2xjUwQYW7x3XbQ9e?(jDqg!`$3wQ
zNa~~5ISXd$7A@_cyx1i0)HO{$F~Ld1^Uq#c6!o)si`+rlSY6?4L69u%3|8JPwP0nA
z@eO|od@#Salv&8!1%rzE4ks=y)4CTrK}4lXTQId_X+o4z2Sr`m1Hp53=Qd_V?W;}B
zWRRi6MecTZedtGjORex^PaQ7XlBW>in%f{osxByRhmG3si#(URJ-s=E@+s6^^WO_q
zxl*s`>yLnARg$OZ9>ZNwk7)S@ZeDGAZbllckSYs3CLW-pLL^``Z%(ueD+K8;KK4d$
zCDQ}i1jd7mN@jiuspAsBC32=MwJarc?wfHa;%Fe(wT4I;$541bT_>T9lU>zZE&2WK
zjg$GgLRq8%Ey|MgCMg<KBDOnZ9ehf51n*M4d{3K^?LgD%bJ9?flgGuN(`)KV@qrhE
z<3m&sytzkE4nNJTOEoB|pL9j&k)?-9-Xfe+!+d0y>2VJ!4qKbhk0mRV&obDABY1F_
z6Tog4<*w(*{K*^580RZ%s>yg^O}oW)H7Fi!W+v}p1~q=-W7*enAL}kdvev`gO!(IH
z2T1TP&1o&_m&W3cUExO&I^)qA(mXVJ--33orBwb=d8*~PJS3B7w=X$w!ez)3YRo6L
z!p?7Asyqxl`)GmxwYnfEh1yUyd<5Mnq~r;$YE;o3VU~nU<)s}j1VtwqNbe}#re?X!
zfUR&ushHuB2yc396Ws}^n>c<mx_o1}`lzaj;L!49e1Iawbx_93J>v9HF>E9G7Zt4g
zK4B=oi8W*ET5<{alUq=6h=OOGdMB@<6k%BR!3IIptKCl3LqkK#{pd8&cyDay+P}K9
zg7~0=Jbv@TP5vEr3<K?PvqJ-gzOWC$;4WHpWFRWbu9t6mv5<}7oHaZO121`knSlz3
zACT%ok;zfAcw_w--oW5ZNu}SA!s$g-zc8I;X6?hnd&2r}TOj&|HKPP2PNuCqLzi*Z
zGM3CeWoY6P4WJ`?AB0RKb^%D`{Y9Cw7nfF&{wPv8l$Kx?1BUaf8!sfVmT3BhgVg(S
z_EIiC!e1Xafja)O)-Y$0HaRlDvT>V}=GeGfnkQlvLOHl1{Mx(v>%5CgjofA5K8al%
z#65N);vpk1*yxzP!hM78q`xYyiw$FG73IA;+#2LpDw1Zkf4dwoMdWSg#`@z=6gPH&
z8&9dTcN9Zu8sH{T^@D+)glqKuA*yP+uGT-I!H>O;c718i5h~Rq_2ZU1$q|<PWjr#i
zB#KF*@@$bL50v8*q@1vC65LY*dnzJU1_8Hn!8$wXcAdm%3dNkX0BtI^)HHB)VJDz~
zgiQ6$Cac2a&m=704Y6zrbD{%TSi>kz6SCZr`Pd&pfv?wejC!$?++ip^rmrGJRuSkc
z@d-Sqh79ty%(ttEqzT-?J3Y7_{hNDZ-{o)0f~66C*VOBFKFl0qlYm%>IUC!l_n5MU
zmw+8w$y`p{rqq@{N=&LgVsppP9&sha$e$7PqP!oBws*JbBVoq|7E?0@``Nf-D)1`j
zTIP`<?Rmca`$FkCfoY-v>ib$2EHmS7t5%CAU+T;5%~NG>ESkud=_gF;wrl7IE9Oa^
zpSzx8?)i9hN{-;2+C(e<l~U6lpFj||VAAUhB}gg+4SoE9aoYaL;k>QhXOHaIm$kWm
zf=4JqZ#1iAYp$Q@`-HQc?m44(g^f2e96tJi+1FEA_Fk-lS}`;EmNVLR9}h^W0$`yz
zyW7k3*)#2{m5y#tWI8iEEJ@PXnzwHYX&JS^>?H_C`zSeKKkzRbdOg?e0d30!6`E82
z#DkzxyUH;aUyEelCDFJh-GstOYc2|&wumtLrV5ehKOr;?(hyUj9WF8h#Xl0rYuprb
z>4~sMtku8v($4=?#`^T&X^HI)^E`U%(JZ5)?TacEF*?}Awi@b1@i`v+T%IAk71_VA
zTMqd=Igz62^79qk*m2b=Q8F0c-w6cw!*Mdb|A0=cjT#^%?t6b%Y-TAO(}B`x28&;?
zOFmC9@^I-P{`M`*HabY4VCEOExS-MHwV&@7JtZVzF@!#B$bDY9M$`8(4aOe-mf8NN
zqp-!ToI3KPp;OatdmT~Cn%4-xP@si=QQ*@iMbrCh<B%iHZ`YQx*8{oi2EklU*ASyY
ztOuFBjc5FUI*x@WX(#OWU!{^*k12{ISsP49kS-QN2^vmqmvunhRjAxgQA}&ouzn{j
z`>1qep=;kbvS}M;!hCb8jPm;YBUtYlRsliYu3rUIzV|_ED{i+v8_*5pdQwsNzxqsK
zcAPs<RGmpYmME-AEP2b|f^>}_tJY@{;9BrnpWFE~PYo8$fD3!hC+kNUBH2KK?0if)
zKlcW`P%M5Cjt|~@aaA0br9ywHf^4i}FsXZ~QJb-WWLuSo`<-U<zm(J6Bs(#2{B8Lp
zs8NJb*)oQ<Q6fHFW2#oC;bz}2CS2yt2nJ!g-K}w>AZv))N;53pJJ^w>=8hi4MjU=r
zbL@BXara7SmrhC1jjH@1V+;F0`#sYw3&$#6g6_ndpD{D7%RrouBbYctP1A5&d?8)&
zYX;>lcnZCYAmIfYhdGGYX((y_t~2RpPgoLPObbkz5HsaBGxc$-#L_Am@1W7dn3e^L
zBI<0jd(*yZF_%>K;t=;A_D|aZE2$qvN6Sr}3{MQ@WM7iFM}v;8`#-0dhnB+zBHo2F
z>^R@Hci4X`u;Z>vx1+4X$H66iNdF1v*6PD;h#$#zhau)R$um}kRlwD+YEYDs<{o${
zghP_4N@0kV$hUMmy_m@r#xTOxV2t;ZmK!@8qeLSH4c%}cW4s6H+W_dmg#^7gX$=b=
zo4Xq4Q)otBP=O`})u4tt_EN5joU`r_mqQ~r>hU*8uWs~xTn%KRGA)A-vfk>O-G4;5
z%||63<>;9@TeXyQzrsex5&(=?a+~8Gb8jNgl!;m(5LqNNyU4&XJo8cD3Ky+qC#EYO
z2qWuf%p%j<sr=G_&cIRG>d~l};Rj|zUD?&8DP?MczbXFpJJHStRedLM=G$DM*Ij$R
z^iz$JM7+tfXxK#`6uieRm<n*e3@22D|5G~K<I3UJN>crb=U^&E&Qk%yetT%bM8O=v
z>R;-L>>+zxCJl;w-&lX1Jl}b-(yk)&*2#oJwZHJF^TgD~PGNUGw+u51rCdv6QMs>l
z4sO51Dh6IX_Q;Z&HN>=;-0TF~j%V^wT=9B)mkw{ve7C&cYiXW&5Z;#Vwr{<3x|B2Z
z8A{4=pvKc*!K{<ya_PcGszPi&b>XQLpT{f0#wXZu>$7jg4#8THB)K}o^ugI0`!&!t
zi3mItoP13Jfxd(kVZcpiOiL2m6{>TuiOBxpZZKtYw`TC8Md`b22;!OdMcDknmmtTr
zY(L1;foRO6C2J(o5NRT!qhc6^3eFTbR@8N+9jdXXN@#T3ux4R$XYvlIn??CES%BRj
zn{C+a&624Aew;>!mI-j1vH5Fo*%?}EIi11Lk)Ld=a7ar%qlCnV$={Ucq<HkpVV+?I
zq@?#bf>!%NHACS3e4mpoPLER0_(#ZZ92XJEBmzR$@A$z;Z<q>7-dkv7$*ws?(A2NG
zaP~TJ-l?nBM32be=2`NU>0o<QFeAs7A>CBR*Wk+fHS;rp)dQ6NX?q!~h{RAcSl0Ic
z13N&(zxLcp#3XT3fGaWjo3J(t9og4>24QH@r!Aj!#<`<f^1P>svA0+XitRo}k!Ow&
zyS`5)eUasQNqjjzrS7=R$RzCgZW&lK;M^yAXp7>OnAF-dB<#c+L@hD9%@?92-c8Ny
z&=A>znXs(1x<V?`NK)P~a@XgGVK!HVz1Koj6gEL8lzI}r<Rv18A|W?J4$zK0o?xTJ
zA=VUb6HPjH7d-_UD-<qWi}C_xAuzu!`?yV8F9&u;TuV+)+Wlrizt?7$sUy?2g~Eia
zQ`%i;#Ra_JhvMxR2j=nq#@F(3pHhhZo&n$uM^_&s&NdCevD=|Ak#4Ui6>dq(2nv(I
zt+SRilojWiK=H$!M4mPD_I=~a=egtNnY@{lgTq9oiKVA~{hWL5?SwSiv$mvfV?V)R
z0{2ktkg>8Lltwrx;5z$BUVK_FFqjbN;T%-jg?ge0jTNpH91pR#H~GGz1MmSuL@AkY
z;wsYNW<9#aY7)l*LzSe5@#d3B4U|S>yL0T5*DwL630zmhhU91&Qg!EWTWLlegww;d
z!bntlo?3rf56&)g%h3XKXBL!f^z@)cfIhTw;;t{??Z*BOx<U7=0^hzv&D;|--PF9-
z!*>fn(_~GHxC;er)_Bu9#zmtwBkS;xz_idp$efhz+xK#fn~&pij>KhaLq2d#&=|a2
zz^6va#8<nwFe>PrhHAT8>u+Bg8ZhKu;c@eU^*!DB;5Zxf$=`$r?F$~8kYFCjWo960
zI`@sahd5v~S_=zRR$~y8fa?*<L8wzpw0}3+|D}s}s?>{5l(_wb^KmF($?hX)DabL-
z54Uy49ZjHPIHE<H<9cdvoAHvKG<v8d`19n#qqSjpx+IAA`1}xzA1oe>$MQBD<(Q2s
zbow^sY3Hh8inzI#c6U1dcESia9Z4F2cV3$T%yjEdX+P!;xt8#BiWIVtbF-KQ(vj|R
z^td?w{dBHRfpP#>skYhP!?0e<+_|x)Q8Kxm2t&&GA|Xha60Yt(a6h7})&hm1O>aN0
zi^=_pdsqnw_@?!hK~~r)SUZ0Q<iqQ|P>w!Pa2mVI%7!H<aioW|1K?YO32r8=Z=?U%
zm+)wv{Yv;pB)6oieNoQI01<8V(^VP18N0{L3&0KO;xEyHw%D!5Oq;klY=(9+1q8>4
zyuM<C7KW#g1aORS*I($ePO-@Q6dgHjHBVrxEk_}DciawU4#RRH1{c==6vwJcSlu`A
zAkx2dZYBRpGao=dBlK_*9)PkjTBuL%bXgCdE=j^RLx8+DawfMIzVqW(!n|dMcLA&Y
zejh?Xh!nLO?}I1Hx3UzoU|ru08AMIK3I@uu%E3W7MQGx{s%7o%OYBcv-wr(JO48B~
z(^iCxfmvigR}|vDr8xw#2!~FG61%=-h5@=V()nRLjg!&=x)GG!;j_7XIFu=DBb!JK
zU(##d7!S;?3DbglwTw3T>TN0~KZ0c|6)zMdh57)SPHkp^0!$;Y+FiyVsCC~U@DJ)}
zT$4+-``&n6ZNLi&+r-dx+f*7yxU=lQJD#xi2-}#1TF;0*oFtn*AYgRZNtA-HZNrvp
zP|kaJ`o<XO5V$3)C%Pf9*lc-1I&oBCBdLU{J@U>Q7l4t^$7gaw2^CNK7CXA?P888Z
zEdi_4VbJ%&oDmeZz?25b=7AUEn&dWM>c5W9Bq_%Go|Xe=A<Dw6cP4TF4ipBC(A1eN
z4e*m3JG6Gn?hZW+aSeB?hRuAQR|9Pc390)cF8MSUtu!|xxH0hb@o;(NH7p-ZRZuGd
z!{R(wY`~UU0d|r}+TMH^c_Aj~UXF6Y(~ykqpcJiXO(3zrIgE#Ww>GiK>r8$Y$UqBt
zspz+2VMDEi2P~PJiN#vWwS>hArn!&cT8X>cfVpNwcF%FUN2VKD6E2YPN!?WNyjRk{
zs2_Zz)i{LoiIyPL0&S-w_0WkbqGAoaKbzm+FWKC@%$Mlw&D^l$3%NbgODtuZA4nhE
zg1FZ*qu=wM=AwJe8$V2oG2YnRFDAjrE6>fkH8@9Lg1limaAj}8HkVhJYb`qnVItA=
z%uu%Eryty1(1Q`D(|#U0Xi)PQ@NndbB70XL?BxNv7I(tX%Jw*Dr;X2{ox<|iG&9Ba
zW|Do)5{hV-tjdR&1p3MtqbA2YaC4i>M^K2|@}7orGo*tANtPP>H~5wi7(wRiXl3db
z42L$~F>7zaS`e%vb6cFOPvBz@R20CH#AUrfq<V=TqCy5?_4pK+v}pOs;2rnsBo1Fz
zw@{)dG3-6826=-rvyQL1nW6h&9%=S=y-qF2etk1u_Z3n+eXNlK=57-N2Xd%pU5G6-
zUwXnq!Fc$_CIhr>8&lgKD#zr9q~bXwIIiKTA;>w>p|dqc+V7~8OD7CVpO@+R&Oxq&
z$T(55fB6EyMwrNKKP$xp1SK>GZZ<s$x!eyO4;DU2PMjoWXe~?wddNm7ktAvG&5)EV
zRlZ~Gyyn4IF3NEmmasa?t-hxxw)3>~9-nK|vGbtu^#dad%~<bXeLF4)kAOLHNav;8
zu*zm|%lK}|Q3(B97YD_SqriRmiu8NplWM}g*3Kt~`-wx#aRj7fm+dJD?hF(|KY~kW
zK{~kiE>WUR1dLXYvBtx&rO4nx`f|DWNw8Ijy!;Dg!i0qhX_=1a^k-Cf7)QdxcauTG
zEg!7DF{H|1;<Xll3O3=z)fI!)ZFC4cm<y;@ja@c_B*S+jN_up!E;pZqBMWnhyxVHS
zn1YvtxzDA-)D3r6)RqS<i4Qj~Un1}2zNcl(U?*aLQmgl1+7E9?==O|JBDNAinqkwa
zWg!kzIUJP2b(+3fE#MnXF;KPL^`<1Q*wuc<<QVidCJX_%^v3ZWL8(D7?xmz<9UZg9
zyuIowz#C`15OjV|!@{dGdwi!%h9oZfZr60QR#xC(L6JZ3St8J%yN~lml~Xu^8A2vN
zJWbx{3sF~Q<~g!(!-Sbj{d=NZTyngQdN}+&LX?~T^%iL>^D(Bu9kk_H#E&F2fVt$7
zQefnbP5*`_7jwKY+&2-K2>40l9j=JMCxa+xrMS~$TxjT4!i|VV2<`bAd<a<3eh^M$
z&8jRCV#93nk%->Cy|J~@h(6AxEX492=Pn2wE!zKARsawb)p;!~Y~OyI3*Z({c{UVF
zIH1v^w-Q*Q4f5S<Zse$U_f5A%Tq2B*pKX}~4yifT(Q$mp2{7R4M!##nElM&#HJb!E
zu~z#D?lj=zjeC5>!1^5dLa7`Uv@Ly5cz?9{hxa5|@s==*v51j~UDpTWoa#FcR726Y
z4+^6^=3iKopJQliaZf4gdCt)6z#B$nA71Xu{v7&f;=`MgOl>n!QjgtR1UAkYaKi<L
zo7W_LJ+4Rx=Ee~1@)MnfpD^)*kg8HnLH7*I-FD$)b=qg%(~xC^LBi$CwI?j&j_a$<
znYQCDK40ft9O}CcrADzmFloDH2|!D9512gMr5ep#isR0kFx=tGRx2W@B|4%I5Mx}s
zv%E26L=_7t$aqRkRB{6s3ZJKEaX;`0uE1kzpg11D)KpzBFTGpw8BlIvCh*aBK{b*v
z=GU|nGhxkqFvRA~vKiZ}orD*`K|8dxGJZbII(1L>xp@s9Fs<a6Kd>9EQg^~OoFWmv
zYt#*LH}`;B&H{Ag9buqEe729wZl5bdDhn^~!;as42aL;nNn_f~Ia#v~OgBRDWYc5(
z`81Tv`i;$98#BIRnZ{M;9tueP+_e*~)ZTZWrQ#azfSa$%B)D?3vBmd^!%d!J%sXzv
z<T{BmHM{A#=uZ42{n!hA-1P?8k}23q<~qNhPwT+aI*cM*CD={50$j@r@2)vsgUSI^
z`<JZ$F>T&E>Zm_BX&ga4Z(qA11`b_r59TVlny@4yLeBh*lJ#$BEj2Az#}r!;DdmED
z$sq`I*_7_JNSJC6omreG&1s4NN`Pt#xz>VN5VT`AIb6Xp-iZ#H#l1o3q&)9oD8`z1
z2t~!~z0iG;c<FIhQ-EB590!_vT<aSGHcA^xTC2!jbGbatCzwREV$D$>ET?saNwi5D
zw1l;xEaJ!stSD>pMfWlWs`c)+#hdVw{L{Z6J|S|M2~()BLgDqKa{z{qO$QzJW6gh@
zf6B8W7>0|ErauYG#vyZT(&BMrY{~;Qd(-mM9rY_=jyJ{+G#+{FBJhS4-^<wGDSJ<I
z@c>p%qh2en8wS=-votdsQHX`=Tbrmxj?1)zeF3=2pOcE`*B}aZ!tzsFd(N|6ZBRL<
z7G~UUx1&xR7``dQLXEc6lkkSKe}9;acSZ=!7u9QJijT|<iXLFBhe4F%_);dDRiF)$
zJIC_i9m&u0bz%5TCR`u+d8QGC9rrqV6O~H9q*iK}wj+4DK&fuN4#w*<;zB}T2}5e<
zeW8Fm5`YPpa`m~#fXi_w+}(C;e{0l;`vDKS-tULwt(g`8)*4xAksX;)v<xg+K+o6l
zXGfM2ShA;uePWsrKxTmwS4PlzUL#TMA%qha_=0sd;YNFkrDTVpY$hy!G7OGEgHcOV
zZaZ|iM7>`fEwB^_dkC{&ua&DsD7)W?^N<rQ@AG7iNilrDa^Cl950=Qns5;SymwS)8
ze8|<jibTCf!kzXR9WXR$FEdfPz@*H`QuN*Sdw6ksOk-&M^!gr_bfataShNHYP^T?(
zVB3C>a9ShRmUuNs+G2U0pv45HKU{_dDF$jRZ^D!m91BTEm(LqyA4qPkHENW|g357(
zoYv&prRI0|1`cvHl-AC-naD`28P=rywe$I~G&{xJ$G6wl+^BG$kQujo*SUQ?VaY~v
zwn&hg7Bfv?l9E^u^c3ElCHfgm{(y~Sg6f2!49vjm4&qr*?z5<N;$&sLjc|NB{IZQX
zXJ$9L`>Hy$v&T$KyS$+t)wkl22@=KzCH(X-Uv`aLnJ8=o)(F&W5a=f<Qxi7VDL)Wq
zflvf+Y{<jW&n4gY2(y8A@twI$$8PJP5F3XcE_0vvKB-fBM;V!m3j8pv8**^MPM|A3
zG+l#Ksv&IbQq38f4lWMaNHtCX_e<}Dp;)_daJ850{sT+<5}<POVW$dg?tbPsIyacq
zf50fMICL)Ail@!h0iWz~RYVl;088dUC+je8^s<4<s5yD&X4(1K*)taVmwQY{!cdz^
zg7|KT<eda2@u)u~-6tF66TW0?mvyujI$``|h*bDB)Fp%{gJ`3~8K|QMT`{2amjwuF
z*1T5%b0cjYQGx5hTs4bPz=1FCm&fIE#~OwlW!OkCa!CD$QGf+sA1rST4kz>x#h)=F
zP)os*VLJp$x(8QE#cdu==&Mh{PzXt&_@;;2^GgU}4Eq0$Zv$KJ;mPeGmETT1Tul^u
z{v8psnQXqm7$lMf+t*S=<`kIjRGpOV*27>O>GYmT#YOa_(w%q`Zc@3^vTVaTbrY7w
z0?RC^8Eo|jPkiue)00~MnXn#)ZOQ7S#2t#j2foIhf;n#91%}7Ka{11b_$EEL@G$b#
zGfNqPYa{<_!=2R0PMBzUm`=VRB2gM@o7am#?|u)<s+pb-7!yW<oY(I8FYQDz2e1q-
zIXgTo*&11K;0x`dU0^D@Lc~GSAqkL5mkx=YaHCPut5{_yhRCPsm}&!C8}L-+k%kn-
z7uX{gxI6WYJxM5enk2=07?Uu1a0Tg!IZcBKvV<Tru=&;QctdODpe6KC8d1^JX&;tk
zn=5Sp6X|GRLIGMBa^bYrj)X}ym(WRbmeAbv%~B(=Sc4Ab?n#*Xx#jc7HJB|?xUxM@
zuxYo4a$dA}{PG&z+n_K&Tnl$<hUWVWZLwo1PD@CB9DGsc5Jf>;=p++v+tWXoxc4a6
zPVP9((tlw`Lf~<@x$}ej2uiP?BXAo?4_hAM+nTVPJrYUO)byH%8CI#r?Os_T)888}
zB>Og3xLGE&7rY;hPvxiFc~9;GqQ8f}Xf9HrF?S~1xT>r1bb0r>0F$jJD(h0-(;hsy
z9^0D<rilGuc>0<}g^azLqJnV!g{c;8LWq7P%r#wmg<Vn0c~28GnAY4K`_p4X@lh-{
zTp)v1NSN!uibXz@jByxE6@fP4*J#ROabXzi+{wcjQpNEdP*A!4-nL<1;gfYIR7+Yt
z#Ae>Z_psgAOKZec2pz(Y!!;c~p<G~a6P6^zY+#yh(q92{#~8*?bi0O7fWE^WFg6bL
z5260Ki@N)Q8Q-QbGE@V3hfZ3zIZBgi&Ie@+j^X(#ofR!e9qJBfBz>MMM9?GQ$_-kn
zxlhJqVD8`os`!=7cRvO^xa!CU8a*GI0mhAL$}9%q(7ac=sV+<{`Z7<#5C#;-so^fn
zr(vJ6$;XdOFWMYQTw!D@y4FbTbTjWbWzJR@*?S2rX|iohay@!TYdAPIW!xXGP+Dm&
zL%~6B>97T{^MYDn$Z0ID=?RG~3IB%UHrD<mIO02oxs--a*beE|NdgtZ+mt79dEZGG
zT9@tl*A~yvFoYb_1)9fqFON9ra6m0?q_R5Nr-l@Qrel*+d%1@8U>W5%EUJ5+)_}Z&
zo_$ATv^!Fu!rdVgQ_pN)KN~kAb@QQZ3rRvHMDr4+%M0&j`GT_f2I1o8;iId)oZm-!
zs8C0LOuRHd$G~PFT(oDj&~<YLLIUIIpLvyGhl0a4D4Un|zmX_G;;^zXB0_e;%K^E@
zu<V4Ww<p{oRt#r51FU8!TNu44o!inFg<%omJ<D(ib3$Oj$stX!PKbn-p%f@N<h>1V
z12|8@N-*;mZRd<F<085&fGK#`Gt7h*CC9*hj}&i%8{t>N69%an@qHvu!d(A{#G=B7
ztECP55VtUp2y8@T19pu;+Z+dj8#u!L?Y>rR*l|LEv;?Wq!xAz#bqzZb#to=7x`hyS
z>4P^^7JQuvP-2X0Zqtxh)W>|n6ibDJwCc`n3S|^61aUI*#vkC}*jD&nEevZ2IAot5
zuK><PN;lFpRGRxBQfvC2rhH*oi&92+-*<3%n`{^EJZi~E-eK9mrnHEZOpLxU;7Y%r
zb*}u#;Bo@y>NeRt+w~B^VFo<t2y7p|{E~g2FxC9~rgOeN$JYEqE9)B35&TXal0!Of
zx;3{iXic~gx$x=IPmhc{aEIcd$Lc;06T%Fd9=wG|OIHY53|W37!<8+F{DI+LXx3`8
z63R(9FE?-@a1R>04krdR*M3??O$N}Qz#E-_&<-Y;hjNo-@v_spbP|=9V2SGb&oSS~
zgfPNpX>{lGgmJAh2;Y(h7kJr?V)_P6STkWNI<M6fsmT7u=GN<F=an|G{4s8fYH7`%
zXj%hp+`xFfj<-fj9B`*@%LW8V{hU4;XgbwbS)>*U9l8s;n9~Sp!59IoMS>8Pf@xgz
znee4pV}+}BYYqaL6VW-(s%*$`zu>eei1cm#?v$S}T2yFn7+QijZVLgP<eh!Q<R6`#
zC={F557Q_aVilqb3>Zh&3Yp0FLKHqm3%UOr?UTSVR3Xg7GS;t&Gd5_o-|)TSFYn}I
z1*WRE`_wGgASO&r;|a@WB)1FjuD)*8D|S!*NC=$5*t%{J%NfC&fB}NB;gbn77~Tt=
zt7~6!QL1zjrXM*@yR_uWv6rbmB4vd0k|z@KCddiBU1`3$2~XlzmY~aBh=DGmNnUn5
z`1L;I5U}ZyEvs`pm@9Q`#z8@nClA`DlM;=-?2{<PTO%~O`&I(&_z2;JK)v4*9`ijd
z36O@qDf!Xqwt)-b&xB0~WJ_dGX#C)fHSRb!H`?q6AEbu|p~*4l8hR%vqil%~6Y2@e
z^~B0Wf74EM6DA|Pd`{_=d*~YOl=7mFxYQYVBl=f}kse%iz-%JUx-af~Cl6w$3!^LB
zZ#-)fmUw8=EGlnFzV+_AS5~Y5@!ORk0TPZnk{rdK@MLAY6EG;k*c~Plp<A5t_>yfR
zz;sw(i|n{MeI$%{cYcJ1g)jyubWYskWxG+wH4ZNX7BZx-=UV=G^x*(o7@P=cbBAWa
zgZ1(tXV~Hin3}yPj2U(S6XueN>x0)@J)d8~eY?|IOp6_YVUnqFIgyY0y!poJC`0j?
zK5#+ygk=cQ=N~q6)QJL9qcMF-7iEMGW<sJcrmr1rtiW@->yNA6*l{aiDj?2{!kA}h
zPBX*6P|XLHmC_G73yGj%^ReR|do#dV;Q;*h#dIxV8IV@!=UhwSO)`8X%q1V#Kh!<v
zHDkcsAZG=n9h~_dh%6+3v|;#aa)l#d$@=XDlSSRbOoXjsai!XaJEF)^C48t24M$`I
zy*8-aOu7+ihq{^7<vt1-uXu4xo`c~5S282s3LUd#CEOuN3LU#v)Dm}JIy2y!<yOFv
z#ecv=iz7ZnOJ~TqNlinw_=GLs#2w<@*B($K>qArXdKF+fIgm#2H2sF}X_+=q<8L!D
z#hosFI-^cK99B_3!JQ9Wx9wsSP_nLrHm>dqWpGch3Il5qp{m_ZA=I!h;Gw9*klnOV
z;tCH?o!1^=0rDFsQF){*V$A)WXc#Dhh!aD&uJo88ppeJbMnvwi8xz=9>kuuQ<(&5{
z?OS10RmjUu!jpWu!#LcyJKhW7a_gBCSh<MPWE{m4O`SmqPv8Eu%gHfrgfU^Mkh;OD
zqDeX$ok3~Jia?M=WzA3KdX+{|`*y?I&5!mLLXILO<%r}dC%Lja5g+GZ)A!`YBl&Se
zd`yZGm#YD9N_H*ViKl;lA0kVtWx1Iy?GxxMuCq<P+xL_(?_je)?%p_xEf}1DrDN#-
zU^CH{^k8|aR1MM>=_^?i=Em<C?(S-F!v-+@20a2q=UD|%R^UGPjWP~C-*~Wi_J>>Z
zM*7Cn7nCk9`ru~UnJ8pa+kw78vv*CH8^)XS3adPW2=WOk#(nzfFU!xPx@F+iZL7m9
z4TF?|ZlG{z*30I>F(E~I!t!o;fp+%=L5n}d;FY&{9t`af<`Z6eebSvoRcUe1JG6K+
zVXnQ1)EEU}Vb(85n79ZiPUFr+@I_Q`3s@-~U+NPu*@n3ywL7B~6Sz~KYKxb;!tutY
zr{Oql&kefqB@jUDxgzAdN*bCAGi`(i+B5hEB?KJtJ8UXf0RnTZF08KfdajnPh+wY%
zZazxuysR5HyLOUA5&t6{k}&bhZ4o>v_L~W>>0)eJCi@PaCSdM^?|vnhUEVjgcGHPN
zO&B%ICrBSQDO0wvFJ+6To3Ffyhz*zig+PSGfmGH^&mimNAaSTp7Wo$W7cp2QdMRIW
zAVC{0e>@aQzWfAprrqeG06zNa>w_>>+n--Tug&%#FuTMihVWs+nCRpV?$k(Ff}Y$m
zuWa{C6@p^Y)Ja@~TEd+$wVm3G{P>XxHb@;RHyV4*O4dU8B2jK>v=HWEp!YD;8*9hz
zE5T&fA%N(<zYrU-_RJ8&{{TH-jY!9c7JuMISDbBFJiz&2DulKd7=g0iAlN?BLVIZ@
z_j~1}XeACRj^a)jddjG-fo-5Y_;TTV#K$39kNGn}H~fT%3~iV^W0bQ}_UkAt;q3Qx
zqjl5vRwvui67KX4gsJ3ITcCqG(H!OJzP>DBNfo$X{cCPezXHpc=BIPXLW$c#LGh-V
zm-o;O6Q4&-op&^FJ82F=2vKagU^efRtXy=fUot%gl|B7EJP4!>%VDnOn+an&o*g=9
zM(sW@-h9|Oqu0y~EZ=Pm7cARvy;X+*=@7DTkL*qn3K4`RDeRegGJpgPgdDsuXM73v
zOc+VMeL4tFzmX6ALe|Tb_Bpb=RTQ)@rSByI@^C$jN#XGJ6?QA9`dCSIC$`C?g!9e)
z%1}UBQZ@$Z33GevjwoQO^F0h%Pb918!On9>NPuclOT}@cyc_sZLe5X=&%9giTUn9B
z;vUf4sJ=kS;tb-ZlJR&59i$bH3lt_@Xm1eqZ1<W&DjGT#;sza)Tf;^yGk-a3LyIH<
z77gVl9^2RJjuC}4mBX_WZjmN%ukH#KJz%IBuIrv0H*X2Uorfh+U8NBTZ)g}M_hN_K
z+F4|}VUK~FKkDm7ez))Lt`G@&zn@IiLC&z@46|!+GxUQEa-b-&X<O*ZO;~=$ypaUF
zaU%!ltkkgE5^G42gqN?MC3?gyMlDe)fp26e%ghrED?=wCgjG+-D^4`7#$+3X#W12v
z2bdcQipA>OH#4?r5FAT0F4k!8Y1tw>f;lCvSj*!H;~rY+__mmUkKC#ia+1!H_w>bP
zv~@usCx5wbJ%&?(>beq_Ab`q9ggKS(EU&a+yXv1*!gj}9G|F~ssKIr%+dPXIB|>r>
zihe`QWS<!DJ}z3to8Bi{K)Y|s-gvaf&2SOOhD4#@p|ah{yzE5fyITx7?wrEifw)Ek
zs?!mpe!|_pK56YF3GTLYH*Bc{>T1mx8@9t-7EUCb_sBV=hkwB2M+%GT<~N9zYD8>C
z|NX}CXN``-)HsxsWvzr+*=Z?G0Y;ty@3?ltsV79f6V}@;hj0?}Zq&Ym1)_#3;I{HN
zHf^Z4rOwPMB1~(k&*c+C3vp;F)GZC!HsF12VM`eIAw&`Mhhfawt}8J0ZL7(a$BwWp
z<Zd2>>&Iroi&`&-T&0GxhQ;~08R`zg{Qgi`KG35&^0PBg$1_9;I$p|rjIup)jdMko
zeFPN-D9zd3Uj2en?LBM{-|-;oi6XIt4ZWsipBbOB_zCT@$zU<ti*s@k=5nUty(444
z=CsmZOc#;0W4R?id%MUAHQjyvQEnbcg=mou+1E3Y%@JyJe%X7uC4bG~%u*}2qfah{
zaFG2Riz5A&cRX-*Nejg;-Ys_;o(v2Dc6&1)fgd9&bG(c)<9B(_N=(vThIqf}EJ4<A
z2_}aP(cfOB&<I8p8$>PLoDkB-0tX88`%q}Lo0TB<5%mXehry6(s{}<ID1CWhZ^&`X
z&<xJXrk2k8cV`69*#paH+blIM+32nyt#k@=zE0NoBTY~@C$dK)OWJWMbS`duv`p=Y
zL=bb9gx)n2!3^%>Zb!VsXUbRK9*kt&A#_-J{dqHQe(O;~YQg7nhZ}ng*HDg9u@1c4
z?ET;ggF%)=yfPT<(_+qKmTSy>BNZ)sW?6*}Lj+-6el)@moX~#VL}@l30c!=qW}%dl
zD@=XN3r&VCVMr6V&mu<GeNXF*W7wEOrw5A*Fl^`Df|B~62xn5GLfmrj8~ZD-r4X&q
zh7t*lo;JE`oe3_w%9k)_OPG5yqdvwgkq?$Th*ov`#R$>~;AxXHB7ilW2pm^AO2cvz
zF;yO{*FY6$%`);pJE?Fi;^<x&LBiEhiG*95r9$5+705glJUk1cHS0t5PcVpXb3w%|
zuHDHEuAYtX*h~YPYCG=$>=c-zW}@;sOT;<2d?Zc>Nq5J0F+DKYZlcu4w@50<d<n%t
z%0q7t%Yu5M^2WGuTtrT;%@TGdZDI05;)Jm+!lI`HblKV4L8TG4yE{N{k9-h3QlVj~
zgVe^Uo<2CabNa}%E}TxjWZk$A6c|?SVS60z-tfJtks#gO<E+1c5L7$b(ffu2e{MKA
zbO>lahcajrH#=ccyWKG;+DvnA+YE=R4U>)g&v(MyrdX{yxCA111N=IJid$B)zv3^-
zCm@}!Q3;SY69Qv;H9t)kWN1xFZU_)tN8FLuQv*}})NOl1gOICpU0Z+*E#|VU52j~Z
zhSTyzgf$NiKKHI+8_ZHOetAqBMg2Q!^9gv9-}DoCn@tk*4ebt)`vC)8IoZhNzR9^G
zz<@m#RxDHi3)9tfb+YhwqFg4xL2%tsmebNgqfHQjKpE3nYDymo>+8*pw4V>=8m4Ny
zxn*w=^~x1ClI=RD7I6eKTX^vpVdjM_z_r2D%KHkm=I`m|&FIgdTxlfSeQ6!Dq$5~E
z4Aa&xZQG*wAg+lCSTh6h65fX+mI<G8h~qk#Lu+%w7gpO7Pmf?u>2yuW0FKDOdR51;
zPe?%4&<Fd(6-2;XXchGzI2xa0Tt_&!?+)lWd5DQZX{|FtD~MBo<@~@;iSa|MyC+QU
zJ`y1M<RDE0p4Ip1I*zroAki#CEuu4hC9PR^*JN>hCP}oBy&1Ic>)ef)?E7NE@JUE@
zMXCdW%LrQ2PED`UNTHpz8+1tPiHHwYDV_VCc5{PQY;*N^OIUVa3+-#BgPQQ>&SO8#
z8l6PRUCc;_OJC&}l&MzwI(Q?kXglG`eUK|`r5b!;z>+vE&Ry*<BSFtztg5*W&CVFG
zR<~Q9mUq8LhoXu!mR<Ie`6@FIQ=1d(b+cROMk7h*b#z*Vb4G&?x%CkyY|Gkq+NQZ6
zJF-eTq4$k}<5(lOT9O>An+wN0UL0L2h+;mrGxWm-rKT=}Z1jdpK1pRPXMwZohi3Zf
z`ZfDaHE5QAC)YfO?v*2!hl9|5nUPJPdq1}b3^3;j5sF>jD<w)>oG|Vl3>%vcd70jl
z2Yksb>AKJhFG~^sqPg!g1qS3TgVG#i3%7(nm<A-de8f{QC1a{+)P!~bHOr-v5AD(s
z+FZo5>UlE-9ZQ4Gj4d2jL<TJ90?tN)7|>qzggFzTPaA<7K4j3wUa4XAhT>PAgh`U!
z86iUn<oEPIErrpbQM?MkRPP`I2)obF9m8~a3wM_f+`eQsV66;E7+aw&3P6D6Mvu^^
z&~Vz|2fkcWVQ-H5iXI<GAfIwZypfe>478Eb^Qj^GKW4%t_m<UxL}^qR1CN(=m3Mw-
zO_bjiDp|rMSiykx(b<_6Y-fQUiZJ{z71Q-S%qh%#!g4$s7_Xq;(8w_9w3D!Y=8gjJ
zEpcmE3A|{~wcv++wcLbZZR$2RrGkSL2AImZNaayXH)9Y-`pG%4|5|Vwfa;9f_@lM{
z-u#WQkZfuqwlCO-ARaJT&}*Rs>%e|b)BCYks2Z*_N_cabt7XR6J@*+u_=fqzok=Uc
zZkSEt_|TB7%=L13y@sB~ew^>=&fSbLt5B-HrzwB*b|U2RiISsTVOg#0JOEy#$-}aO
z{eL99kNXYSe2)hN_;6307p(T69~vf;z!#6gt8;pyt-5WDjmngsg#VxB&ZXCq<hsuD
z`W1I#fE-B7ct;G`uuMQR%Y;23G!R-Z3KC^Ywgvxt$G6tbb0bfcq-wwhdc@|sS&?}n
zGa~kDtzDqN=)EA?Qggt2Q_|+b;wkGMgJ2$~w>0wHqXQUzLsE$b!LT#}p4fLz^CU@p
z7+&<-(RMcOeWAi)r;m*XL6~|4T<E1Db8DMyDV8+q`=dIdMuVStjTi*tw5kd(wAQM`
z-ZMv+zOv<7n+gM+wgA=7-R@f=j54~tXF6fW*plGpbWm`0<d{H}%_F%uEc^7j8s9z5
z*oxA~0Z+#e6u4RSMi)-0E-ty)*RGfJiI7(0dr$3(r8LE0xw4<{D)+@~n?BWH8>IW<
zg21o#RpMUP4dWk{eO?iJNG(1c6jOR^P}MF&RjVh(b_vg<?sn<G)B==cC$C(wThL8F
zdqh`(wEIH>?@w`WL<;6dZw*k@KWj0zs00F~ia^H9wo_6L5M|pj{H=#}1TdYa%7-%^
zc7;I~vTj4uJhR0#s(QpVq!Y2yOR^{ma7O7S`Hsce5?~}=i{1)-%)IE}pR@bxY#b}<
zQ-MjUb{D=}(EtW`&@RZqW=bmVu^~fcpKBDP=YSV&5$*6V8g2_FfRD_(ZO4c~0jD8V
zCdUlj=oh%4ts)CxJIk(Y045a{DZG|l`36ki3%lQY=Z+5q271xjbpW~fw-cAG-BtvC
z{kq}o+8OP+-VW7&HQ{qh=T>5qhAWi!>dIU@v7q}V4%+Tf9j^u@++znGYux)UwDb$r
zS=Shqdpam!WAm1Svi+HCn9Od{rlqsE$4}uax+u)Vc1+!Jrd6%1`f$zws(rjKidE2c
z)prKXb_P5k_BBs+vtbK7A<rq#hM2TJVc6;<RCG|04~i#|0_O}nL~N3PseKu`8UfYo
zzPc55eTVDXEGb~Zj&i*vKd=H7qf(y&exFSoDCo*rDvq5;m?ufra&=tfZvR*5TQs$&
z6T67?04T<>9h3ouHU$iGWSC9#c`dGJwlz8W{Z&&#tau8WjdQ(*CxyAy(YNC4Y^D8g
zlK_cQhTXjwffF^Fu6<9J6H~y5mv-9~@cZwp(IpDP83~*caHBv=A28IP*B}(qb-e)=
zU|#{xq;A!m^YHdOdn*5<f{oN<D=8Mz;ABSdP$86jKcEDXF@BkB3lSRL+s`O=y%o4N
zN<+I#TOymuYYc>+!Vc+pr~#u(%m$@u!uwlA=cL<88rau0P?~ad@08aJQ!A;{5iSKE
zUse-L!_euocFNRVAI?5aV5I8ofgSp~n(UkFjPO6>3nu4-;=>080egm<e<Tt0YdAEJ
zZV9TY-gS$FKG|!uaqp|krG*9D=pYktCA4g7tK=G=C5NqoB$|>G#yiaSN<2Oi53h?S
zbyeMET6NYNCUiA)R!t-m^?sqgOFu}VmEQfe!hA=Y$Tu9C8^Z`3UJ4UcNNzP?{O#$&
z;bC35sz%sQki2ZNm9v0{v=i_h`*ghtC}nF?7b8M2?&nyNTF}P2FZUYDGDVw%5l{<e
z3mEls*CXtM`HDe^4$)zas~hFhJzzvFs@;Y9`ss|A*v`oEU(v4tczZS}oC|)Je+_A)
zj=lyp65F5gH=;&~v_90{P~7nzv~c!n=xtf#c)qOWsmVuJjp}XS?Z`Eo9AmlRI_PA7
z(RytANA`Q@MdhvD?rU|xWWStzqgK~89E5dVZgmY*^b`bcJe`Y85YcP_lhd9q6j8gb
zfN5dfT|TEEdcfP2!f%yvF`aG(rn6+yO6t_Q`G=!)(oyJP@nUq0fVU$>YJqQ(Fk%ll
zZ-=PLoJ7dbr&7k!OA*F`ew|vt+r3-1IS4`L`<9c5aIUii?-H0Avdd;p?`Bd~xa@oD
zcUG>raid}CAYEiIO$Gp@8Bwu!x8K&d;X+RX2dtr=+#@qj=8>To)K<X5;a)7f+{5|F
zP;5!DIR!eM=mFjvOP1Av0*#$5*y&9z=h(<q^k8%!q&nL499O(Bq)w2<Rk(%bt1IB`
zu)&1!w%{6`s*acijX^fyC0-RUbm5G~AiHgsQEiy^C{?eQ(00CI+Jib|C>r!V2*KAC
zuXm;RbkLKa(#uh=<i78~jhOOsET@HK06iT$M!Rk2h2CTHon6@>{QCgPiiY-P^CGzj
zqQXR{!DDfrXSCR%kW8uQFLh~|R6CE5(XWj3;84%SweG984T>Mmi(IaF25l}&ruy%8
z4Kgf|fif${Pg(csKwXO^ArEXhx(<dieFB5-!`6P|-U@;?7vtD?T%J-btQc{dPt~
zO5mJBmhvZZwHdFMLesj!3cK#FjB45f8$=Fow9^EJ%sQK6Y&pdjjIcRhjFx)prVdrO
z<)zC)hE)+oVZiA1E@!@4Hgf<tH>q@}LMJEhpHR-bI)11m$jTC+mA2yYymN|<e8A#R
zE!(z12n41^WZbb_i;MwpyOnsk9NU^+@IvtICPEGG`jF9+sutMDTe{-vq-4Cy8mL+H
z>fJC6x_p*yGu$5z%yqie45wU%n*#vvO&w(~1yQrB9f14xtfGFQG_z}gX@gf~E746a
zrWHaW{A{0L)8*8_d-<U4#qmu$KR*fLEe)l@Aq|#rV~dn#kkjBxTg?Txz;vU}GL}#u
zUiTox1E#MorNqn0D$aDsW!+q8;Mvq}Oud(aSPV?Wa+^Q2lgh&d^A+a$P~=I~fR|ZW
zz<Ejrue+<Rx1AzQS#%9q)g`)^19R99t;iPr<3T$T=c@eO`<9FDfhU<6FW<e49UOqA
z?lON01D0vXdq4SOjxL9123;Fxz^PGnraT4AsS@yZHk;NM8(2ICxeguMxX5U+_kfW~
zwzbrFp!<#;1@@@qy&`A>O?~fpOPl$239Ee!_~@)5n=YleaXIkzAe+N8v}=07)zd?)
z4tCf*$o3Vq{o#sDOV|U`#HyQ}(4z?%@LGByd};KM@^iy*S5I3V%B{uK1T8aW4>L8&
zeJv-l4V{_i{pFyhvQQuNwIh}_^;siq@90$qIGV&`utq3yyssq#(&EK9r3Sja$E72-
z)D*!;nCBC_lOM7@iuQuHomO2+jbw*|+fd9CPPa}|cb@?4d)IVDQr}i^y01mutu}vR
zS6Bc=@>s>8cTPT=wuNf7PDupM0SS2q47b6Vn2E002spze$xi-E*K`0Ea+;dwy|bir
z4}vx0@`G(ZjdH`=xj~Vziy;yAaiI`N+{&R{#>9yVQN5@WIwj5=u`^&;7S**VZMxeB
zaJe}xw4dlVid3N8M|u{-{={aA*y#|pIs)~@(W~rJlfr@f;)#%(lV`L7)K?AloCX2|
zJAIpuZm|JesC|;1vk=E#0dvw%g-iAk;dKP_d@p+Ec%-{?kJ0Hls{GMhmi+?7>`T)I
zz<tC*Qy(-4^f}$E5Dbj``^dhbCGINZ_3c+xy{{5o#z41u1iOV*j<q-frV+XQmi?n*
z-NIc@^ikKrF<Uq&kk{ewX{z@t4C|swJ-NjyM=Ll7yv>#AI82|Rp9M^NdMAD$d-zj;
zypAZ6<{D+e7#qoxBB3!gybE)%YGX?_Xcl45dysxf*pd$(4W<$B-i|4}Q_g?IM_Jh-
zprPowuV$$QV2teM4G~w}u!MHlOO=Qr3luOx$ZhEGA`$w&VI)_kY|6d4n7s<Tj~q&?
z`TlNoRA3mlHgxC>__}3ZYDmWn<`CQmo|hcH5h8{{Y0^N)F&%^3KJ|`RC14_HLrTVr
zyUPI=T4>u|t+BgT(l)e((zkQvur;BtHzY6h5|&E`L7_2Ct0Z9Z-Xjmu!1#oFXcK@=
z`qMipSE8LSFsm^6$h@MneTBV#>5L|^e7S}hG+z|7vRmu}#2PNZq`#a#qMU8sSJ$ip
zSw(ZM$U9cfC-ts#@xJ{q2=L?;y$4?*W(y4M!$q2VNrnFd(|vC)F8?U0#FhxteW=)i
z(_maJCKWvt)q-H;5R7gCHx*GkXFc7b0~nR_?$FQvkL!jfr9OJ+q(vRYhrC{;TaCRx
zlB;|6k%bGo9JWCnD&Med>*ZBL%~rtpiP_PC++8^d3|sF~nqeOXVMf4TUs7D}67O~Q
zVPHxDirys64e~w6ROCq^|7-zLP}+q}zXamcz8E7x0;7twwHm7$Rh6c_QhE|Divy+&
zc+ws~&xUrx`{vbw8<ppcfGNBg8iZzMcJ1N7cd15pN#UCb20YMdku8ka<h<ei*_Gk4
zLe2yK9E9<uY1h>nAi}iJ+SAg$i1*s!j_9p$4rq^pA#=kh(Doe9*XKP3oi<XB-O?N`
z;7LQ7rSb#n%<czSbV2b#rE$3iX(+28Tx8EJxt$s?{HkzXT>1yNCpg>uk$PPTm=g{2
zr%Q^s-a`;I(7<~YEQfy@)zV10xPW^_Y^ur7rFs$e(@Nzv29cw#xrV}Z-w^Mi=<Py=
z_R@m`--+Wm=8~J22V{5GO36q`)hiNK`$DzArMP$06`p#6UvHB)zs6uqH<i`(WwsY+
z)E5#ar-gz(Cur)GWVVLb&bMt<_~_f?DzP4`RcU|`8FM^`WgIU+BoF?N%0k@f7?hZc
zbH)2txwHg$TLjQ6dF7a7ez(Y{WcD*fDS{#aoRC*2c8ZF(xmM&R$S6XEGoB8alzsGT
zUq#0vU{dCE^0DQD4r9P2iLI}$zTKLPz)(q7b0)yz&7kD-NA|@Lw{2Cla({X?-6?g*
z;np1QVj$&v2!cw^6cv06!zlAWJYwWA?i4ii@O_wh8_H1&137$BbE<)RbU+72ENbdy
zVNBS}ga@vo7WsTBbaa5b`NK*SS?B^35o|kE`$x}qKL#PM;ddpnASjwAtve?)BsyYr
z&!YC4<yOrrU|8N`_>ieDbRX8k#z`g29q8Udkt_74>n>Hm+qIjHwsd@q8{S`ccItO;
z!r^|w>&{uiguDpJc^l5#{#p)JBP(^orBX4VJ&$%WH@wGB{=^7B+Z=+x4a-_q0W_a_
zUp=Qvxzz}H-HS^$9q(UWPIMgGCk<I+a<$&@Bc;|*H<5S%OQD!;2fAH8>T97m;_;V&
zdzCZO-4^B^dBXPAi)pWerx8OhuEVOh2Oa{&d7{Lckl7+G2(5eZrRt%l6k-i;!yGLx
z4lgPYXTSq_+~5w>Nd7|Wc4-{D-j%|7Y8Z=C>VZVF;(Y1S<0c)&kxMbk3Rsz4N6n&2
z5tKk_%?{&Ks|o>^iKBVP-mm-Wws4eWYSrlyIpFQaO<M|VdbZ-a(oRh~0)^4>95R`T
zy@kD=vX54H-Hl8QjMO^Z2NbmSez|~@S}yOW(7LzhgGJCKvrz6a_P&{;M4N<h33$74
zM6@-?FaxJ{ghmDw00lm(wIP6swfgAi+w*?Y)n=mljK@9jo&|rq@_x0JXdH5ny{#82
z8MN|>NlC!8cfD-q=a5qY-i{oJCmcDnh;En`gf8gbb7Qhsg};qsZJTeeJg58MZ|~j0
zx~UvJUhJL75KJd;*UfVFz{Iwzi-w?C+=I}-Cwr6Yu>AraRgXc`+%a}o-R(kit4mPU
z;vv9vo3jUjPO}eaSlv`C=`%yQy$8CF+veI7bRTUC!N<O&oTD#iSrZqSz`G+N<S@Kj
zbi_jjwdXoaqK5=9T`{|(9Cxp*3Qc=&V%Z<s{+eU1NyE^?R)$RI`GKM}Fud-N-pm2l
zK}i$K05jma$Jjp%Q9ukCS3DVpIzDB>C1f#gc<$HL6y!j`rQf%NdS7}(Gu>^m@~~Kt
z+HUCt<KR012pZ!>B(2|X7qO5Ep`N=WbYKETBf%LV$|d6)M!7!menpRu;z0cEu4u0}
zKewK)I#w~5xR{#Od*lfXJEFm3y`FkcaIs!S%`2)DW%pudIP2Z+_dD>mnngavMZT}z
z-WF#UGns>RQkkA@hSup)#~OAt_eZ@kdm~wqVPGvMDfVf&9BT%Jjy4>kX!Z7jIgAMW
z_1pBceJBeC*<CyD(G{yIg@d+Bm&WV}jFQYGi<`!!HZAzdXY&!!ZhAi`|6R4oambAO
zc494eDcxQ1RilZps*isp{%jU2_CV^&o7RY5?aK2U!HGVt_*fM)5#InSFooJSF#g$$
zOwL`=;9Oa;AuidB`|9m>43~s8HB$|Ekd*cO^Y-ss(sitBebLIoc9JrJ5-;Hx+-J8V
zTUg0^C(9ZHkJ*RSkOGC&|2Qo|fuyQXa28t;dupt_P-Xb`VL}$Cq~n=9ZPl)>$V_tM
zVc_k?G}o2T3Ciyr^9>PNy>sf~?&qWU$Np3e0lXT_3@=?yFuzCk-F<a(cx)d(KBiTK
z&|ETzUU*ChG9hKS24%THvf=GeIR-m07H+iDiqiSJ=xbJ0Tcds#e8(+C#7AjDI$r2C
zj@$O9a$>)Z>}oEGCH)+TFy!jQ@80r#WJ)zi<Xv`GjL04^4YQKW0t$R>3#uejrN1nA
zizOMJDEMWst`UXK_)!^=K|RQbNTjM|hICIaX<^#(!OL@9ESqP^QHM!p9fLxvoY<&M
z4oUkEAP8I4(ppvZ15+GfrvT$qt<>Tndc!ul|1*AMlNssY1Zi`JnaC67b%zzEdn{HL
zc1!z@s<&Fk4exc=ZHdgeta4FZ$Sm!@OMO4%*HDm-cD;_7ZKK5pj7WW&InwN{`sg+x
zoR}E1jn{JDa#6B4Lxm&4_b@^ix)WM6bmE=?Q^H*wOvlHm0aN4c4TG?Z_uA22X<;j8
z+X}Ii)KP=YPf>^-1tgNm!?Uv6ku3aQgQ8SjqPa?oFP>z{B~3sjvD~S#O4iIr{XWXf
zxCd<H0~~MYcO4*9;uP7P<%Z@Jel;Cd9Z$~X-ky-}=*e|r&_d(Ipzi@4Pbd1Uk)#uG
zF;9mT`dU5SE~4=?+0&LS_C>%F@HbIuzE-+6YD&+$#n<LN2#stVgt$_?TMC71Y-Mnc
z;{#su;j=c??Eysq2c{%#hFW}OvA!sUrm#bY58<>Ca3^1^Wk>0~Z30SNDOA-;ViA`>
zm6Wg<*Hc)#5V99Z3AX&wyWcnUo~dOgN#TdP4s*Jxaf@2hYuy)D`XFd^HI`WpxLGk=
z2S^AD18;Y2*vW*5caJ;}wD+P*?qyLW%3Dwc!RSfB(ysJhL-1VY191-T?Bcaw-Lb|&
zYwQv5UgVc6+mZT@r^WyuM3p!(sZeSGqq{`VqpKor++z=TTaHF$%H{%A{x@E(YJ7pw
z0vX-OmVH9I4Rn=OH;Ds(sXdq#Q&l?<>AISdTX+^2sVB3-=!ntotA|wfBym)N%Yo`;
zhO%qVte6y5p=lS~d*wZHoIPseCMhSy6l)v>@UffA8ByV~!wl_u5OK5dl_Dqu?V1Ka
z$)8?suaPNWxwP$3^t0}(7yL$h%zH<*HzXyJiHIJX%~Wp;!^z!Nv_#3v2b}#_vd|FD
z4VXy0jh1Z%MwFT+_LRl=JG<^mF38(W1Q=HC6)bJwz4efY<P^lZzOP1B%e5mpt@^qe
zRmA4N?k#9pX@!CGiqono6fvq4{IBj9hiX7`zy+tJ_}zpyc2&<YYVp~vyY)%|8zL9Z
zyQOH*4ZP17!XP+w)J@+v997xEsf2&L8a`(+j0xb~rHE+F@4a#8&7dL~KLx=XwYQML
z5jr`*G;}sa4kOBW4|PQCiWUrYZnSL&|9hVej(qR|$KpWU1JP5>9-^!_oXFSBh=1R!
zS2D$T>}<^2Ao5uP{~H#h4<x!7^?>OdynKA^=a~1^IPxPqEI6#T13aPSMP7A8Ss!e1
zSa0&cNRGynXit|ApCac$#(*sjU&<CNp2ZCp0^jCz?zbO1C}j}IhtfiCI0Z^}z&lqP
z%}Qv%Sj6^DR2Ca3T{e;nhA3USA8%Xi+8DZ9aQcQQ#OGwD&i1tGxZ!QbX+OhsjG7*Z
z<%IlQE=_p#zL+X3@7a5^7(V7<mg6#OtEB#C-d7Jg24{z6dn6I*pb9NjTLbNp=Y7jX
zoo#cxFqzO<;i0qKy0$qQE6;%Gf#US(o&zh|0XOY7a!OHyNdXU>cSnwERHi{Wr?g5-
z3>{++fk1p-wnOMY?f^EF^RD!GQO40xDH!KnjRB>+HN!xGagNrYolK~%GvK`zs_xfn
z5b!~pgM_v?B0d2Rog&rhICHW31@I)`pp~~b$J%aqQ3N0_g6exuD(#t}yK0cpOH?y0
z)I&NteDd9u2r!Ij-|&%FqGJ>=3Z~5l6dil-LHba~YWcI*umMFk`iaC4FAS^7nNmpZ
zBz<h6%xLpbp|evN>s!*~vIb0yC92`~f&YRtZReL<7y66ea3eD9O3Y${dQ~@K+9e9z
zFT-{pboOmC4*L-_48O0Y3eacDs1e(O<W9gw(~9n?_X|kw`=4HsiO6w8LUKnV++{>F
z%5Rq{g|m~Y?q>ytd{%LF7mZ}@OSe@!x{Oci^}OLgPTwAdgPi+|h~yd_!*Oz^3)q(B
zWq@&EvcmN0O5Ovm`(npbW-3u$1ZN#%dYPS4-DkL}tGxO_vm??y<$jWtC4DP{fo2PW
zyXxv3YoUX&eINSE6Tq@9Qtw&nq%T=1vbV0-<DkBCh`C(4m#D#@>pWJcmeQg~^}f2P
zb&ew#rQXjmh))$KYgfz@#|Y~gpQ~GvwbEOjH87&+ILKAd;j=n7D6%*rJhUuaID>XD
zUSn{Aq~@=qgfj$@-D_T13$=U(Mopwm7jcxRGl$*2jTr|gNm#`rV?STZz5)S_7HFq+
z=ZZYC)rWw|MqUx8YQdC*3g7xQ?*wjSwV=Z9-MYnrnIkg^6@I^a+Z+2d8u7zXbypK0
z1@>ySSwLU-xvYGXQO3M4guP>UWlgg`987E*6K7)EwylY6+nRWyiEZ09CZ5=~jh#1{
z=YH;U&i~7MCB3_=`d8?xUTbG3S=H+?n(2c<6+U^MY%$FGmL2QFgjPHB%K%AYwQ5A?
zB900?T<lD+a`u!)LQZze6RU;bp#|b@|4L_=UxLf~D1|Df0e)09@*0R(Wp|yafN@F=
z9K2<-?)GRbAB|fGh{BM3HZjDh0|Ylx3XbIz$=fmWTi?k7F8lz?dev^I61C8LO2HD~
z18dbtE6#&@VLbZ1WVIR@O^*)!WtU$IzCEy%z)aRMqA(2pE*n2>XRy+|$=HE`+UFf;
zz7Qq<cxMT_Ib0LQ6G!a<BEDa{BZTpPeK{74p~mA0j8^rxsfr=uHqAP<21ZfG-Fj1F
z3;BXy?{xK4@H5fxVWu6I%)eAmwcArcB~*e!_8UIVNE@n$r{jf%Gr0Czdz)5i<-T7N
zRtn9V)J>-cf}l<V2S-N~8bOT5kRIurV=D1mx8wUQYN^uL9e8)~3T+FLwb8pMvwW-Y
zBLyMFybY-6i~Ey&9<ky_sB99m&T-r-0%<camVbNLTwxZcqLPnKBS_N7ze&kigroO#
zuS5q7l1$pSY=PUw<p)sCt%*Q(!=bTMi03-*bYf4QvVJ8}f{p~5Ohwu~-0LI-Q*ETa
z&rb`SF*}+N9_Kp6Cji~1ux5t{WR&aN65J&9L@a9vBzm_wFz`g*VPVdoTwTj%*SB?h
zLBS-D3{yt}aiViB@h6*q=20ll$(uNOln4_CK8Vnk!z3?&kR%6RK?|?V#JX&p$|xjf
z;k0#9rbONv+r_m-+9%kPpNK*r*YYnJq+>aTJGfpUW#AV2h7ee=N)O%L_?5GN7L8sk
zG(z#AW43v#1L78yvd+K-n4Y!Ey<4Pr4H)4zx8r<p9Yw60f{#5yscx-J^x^&o($UXs
z<59v(UWy}=g5~A~S&4IbSopoULA1%D@e{k&5Hg0RkONX@kAt;z9MJU;3Ilg_VY}j6
z{P>xPYWK%Y&$(OBJ}o<Hx2-5JXmf7MU}KC>VAaITT2qYx3|RA3hIGnw8!K4Y6gFx|
z?K00Vx9Q>W;vDq6N|4v6T+i)NBR_56$X?gMGAaeUh4BgOzJ~7yGkj+S(|!oONNlrF
z1hS+0dF0$l*Wigb@lLTbJdoqLX$@P2+duHc!3q^$z#`wr1YVGLG8ITT<@qEw!=5*B
z(g^zd1*Ecm!wF#AwWJNQPQJq^K-<p_XBf7)Ta4>PxoMptyvZSE1&g|Oa&EP%n+HH7
z>?=D8h-zCLH9PD)!!FW<)Etn@&zi968)UcN!=|?=liBL+B>yAD_7!T1q&HM__eTYF
z$>a)Jt^kV$Fg7Mn=l<rl?jf&)`B8MNgmQqUBOG&1^hd!^dXwzmwrqpoLk=WkcE$uO
zBb|EiDZme3mt3LD5ym)xcO3q#sM<w$3IumiNYy&C?HH2dIMhs1?Porp7+`XajF%hX
zsZ7a&ECkO}L1XznR5EzFtv=~84X2Waq7ec;*GpS?Vyukz&wK;!v#c8g*X@v@@u;7*
zYypp`9x#>p_MI{_(rmQW0iV9tdm;{%1q~WA#P-T%c<+I5MYuzIrKZcG?GWypmzp~m
z4u5lfwmS}&h*aS5NgN36@a%~JHPnTt-TDZah4&f47`5Q$nc%@=rIgDB!gokn6;)i$
zloAN(og%cM1qV`Zscn`hik(F<4NGA1-T61<Z23vQ5`sGm5=~b<Y2Sx2Mq9|>FF%q^
z4Ut7AftrK{eEJrGV$P7{B^6TG>V(N&LKtrA&Q3bihE{ZP)t+F>nOcK@@}7AG=_=~I
z^mr(gBJ=mEBiZQp^rBVtG%AK+ZMCFF)2Fz%4Tg_jSXy5VY|%_;L6rgycA?e)R)2a`
zHzwEGe%I;?fPTl^{F9J?T!H7Wg3yI~H+(g-)o7d5S12eVuoDt@j>8!{Wm<0!($b-Z
zq8lFhJk)Ld@kqnL_<cDI5_CIH9TMp(lLLLXark}TRI(#Yw(!J{ND%0=j{Legku*yz
z823Q*OMo#|FQSSSd7tIQ4vRsOg4FW&cA?Gm)i~Q8E&52;QpWZHqy|dhgSSBIcQww(
z3cm-oSRDdMmyXrIZnQRB!PQ1L9N-J6dH6rhzZ&_E$j=KpFVrsL9?ZFMgI8mA>>L82
z^yX|k$_A_Z9kT;zp5tDxOjh7SUR^Y;Y^WWxVNrPNTLkc=UWJ1)C?v1J?KDOResv1b
z(yYF0H4kO82}Wt}({lm4&DiCx?ve0*+4@blur=exv2HM+mBuPOhnD+g-2N~~Mv(V1
z<cj+|gO1A>O4@YJ@|WSpymO3<6U%8lZIodjDg>8);GyrU@UF&G?t)8=;0lPFr#%g7
z2qc0#VvssCGSjs&?TxQr69z<-hCsEdSoS5|g$43*KUk#O;#4O;;CHBi>c<1TkEjsj
zV3O+H4Mi4~3Lu%6FSz@*agJX;9y=ZDrfU@C%k*^S0^_2HwgfhPMolbckK{o*=g~P<
zHW#_lgHA|o)l4@ivuOh5t~4*Y%Ec{#_%Zd+N~}V<hM{D|)h_<DCuo0LYFPV}L<bQJ
zjjunBUUBHhBp?$Xtl-`;l<nBBT<BMzej!jTfvY0G*S@Y+rl-TwQnujK7v|V}8=hd?
zwq*@)MmZu(h~a<us3#^piWmn#g>|<4nlwBSTGR?6Tisg3RJZgN;$Ksx>Udv&kOD%+
zhP_wGQRK$3YgS8voH0rcVd9^2xNlP{pnKQ3cX%s*1(UYcNErD#ZP~9`iI`($cM(^|
zCU?z3eC|b6)mCUfFpS8&>9@k`kBNDCf3P^)mgYXsA8IkKT&uo@g^W5AyTzIT4XmT{
zR+_k(p44tB!eTU6UfLklYK6jIodx%Vu43SFk^!1G<0>+sW+n#WnjelUQRfx1wZ$n?
zxqhFRjJNBSI?G8cuT2k;`hvC{>h$gMIsmNeU1xqkt?0?4+sY~lLojg|1;pPw4V7A_
z_nGvR#iRUyA@}@i$~PcM>WMF&l9emoS1ymKC^ReV9k0ko?Uk%?4xv9;)C6!8!>|Jj
z4P6?3>{hm!KfKJ<wS!*r%Tcw=5lvt=pQBjzpams0e$koxqPpNLyWr3KTtnPPastJn
zlT|6d_P}I}*E(F^Zz()#T%rD+*OJI=#{`#ACXxo#Jj*JcMn0aluy`9)o$)1$t}sSB
z?79qENEX5Yr`!d`el((u_Ts5}83WplRiZ61%P;&p_>oVNXYI&Yctd5zp~E2VFHNu&
zcC&*)2$3-7AeW^e2=v2sv4WVs0SZkh8%Bi=K4Z@5aliN<-+oU;;MJ&iQ^Mw+d9!y5
zNZp^Qv?zBtCga<w_@_?za;7Bv;gC2ij4PFNg5b|xam6-T_p39Dw@_;<s4&TA<6iO>
z(6uo^afBe-eHC*nnsMYNY*bEq)lUrCwu~LbuvYwjvE?QdxB=n5kR+~MxT;2cw~1)l
z1=<KklRPRmDk>X)idAjQ-a7uW`yDr9o|3oWm*_+i@R!OJsIiF_XOvQYt%gdvJB+7J
zJRP9Wb^Z=oa$LW-d<1{wTf7T@^TGTJq53M%NR}#jDe8pwCj%#K6BaSJir&ZJ!l3+C
zsraQ`t?sa|K{WA#x`>#ig@!c!O`6bdbhdNIXa>hOU?`$ZG<^`vAQ@(wrYj2rEW+I%
zf7<Jmzpufa4f+{z+r4|E!_!z%0KinJ{v*Q<vZjlilmH!F8Agug3`>o~S)j@Vu^K-5
z9JR1L1tR<$R9(hPb{uHgEnF0zkz}ChnOFC6kJ~!y5^iMH{jzhIqb2M{G-X|@O{=Ik
zy$9yx*~(4%uyQd313K|G+y^e(bey#%3a-rs(9H(SY^~-6aqI3;J<Lu^e4Mb_*yYny
zWKF;SAa)Y=E!>X4lG(qgB&AOJ7!8Gf4>!7d*Cmwv`#7<7fm-I2q9ibahdNi@nhjIQ
z4@T#u_3YfidzQ2Hqg_<ev*|?0UpyQ4;2S@h*fpDHTh`JnG?7&N8T5pY4!^my63{cH
zAxrgF>L7JLv%c<ic|A=MyzlI&b<q;|-b@la?a1-nbNjxI4c%R(#pt~abv<u%y<b%U
zPTBapULPm*4mI1<COdJwJ0E!4`LcZ36}S&DYbnbRk~iJ2fV0$mJ$G%Ru6#Pgp47wu
zf2vUUPL}yB--iHt@6~S`dYw)syd{aV?kfqUfH(I^3>O4%z{e+w6*@hDrsLDIp?OfW
zE~xQ5R@|72gc@26&s}*b(fsx^U{=rjwJC=0v?;FZbV!!K7sGbPl>qYl#XA!p!0SDr
z>y=Q#eOfrMxcepTICsK?yZ9zK<B~ya=%ni&*Lyu7WqJ4kwki7IrjO0$wmmg=@wVV-
zN3Zksa)qwzS?-9~o=m^F6sOzv{aLE2w)?w*VBCP6Fk~??&GXet*Xwn}Nisp#TN|D4
z^-9)IdCa$SO7mgIRN2R(@n=`qhxEHD+uQbME>mxDbqHr465=j~lcxigGn5)k4+lQi
z7!&~IEC8>Mr|9hjho=fKVV)&&%z%NWx6s6U|9q2rPf9K*<1pdQxs}Cuw!yIKxZKq_
zeRF(~elK$PqkiYU<bFy|&hy<}hU@JtAN$p%miID?@oIx*)A6{=zN+b@%kdgYaB4Fm
zj7^Y3ZxZmD^4n+hThUhsKs^@xEBSB01MDN~i{8e`+4Duln=lg4bAu;fwyE>YsiROY
z=B=P?(z&O`TO9;Kn&C<x3fI?N0|bH$YOmRe@_VS?vf(MH56;WunB^%b*C75lyvEHz
zp-=B*&16^m#0)6cP|`E-?}IEug3Fb~Aw69~f<+Ei(&hvro`aX5jX_?(HTFw)bd4!;
z3+(%Y{S@A%<LM1uRp5E|M9rVVSE9;S%oo4^!X|2R$WCAG54^Je)x~h1%S<AW=bkAD
z@rwKMI(CA}S?zcV4$~o6;t}cHZEHy2>mKF({a~@Wh?e0O{{QlV;eP!~EKc6`-kBf!
z<-uVLpEo`t#3Lrb<o9spYuxwljUhh3Ee8OYchIzJBtG)c@fFmU;Qev-;PC^M#RCD{
zdd~{*?~3t#J^Gm99(rHxt@8+&fGN~VnhqmD)C6ZXJ|j^2Q*o<*y4>}?+x5N^<NH#_
z$&b**p%AcKjQM`s!dG)~E8WGTEmrY>7u|RBCc4@lS!Dyba4SqeA+l9b5s)+HVo7<~
z#d<q9B+j_`qSaU`GR1I_wJDeN=-BPJPhbeRtnkEheD}F}kn>UW)c+Y2j-8*CQ|2VC
zDr#OvXktGO00!(X`NF!5102T*I-RQ~m)<|l2+4ZuxY}%N1lec@;PMn2)<T=h)IUuw
z@pWf$zr+B?Ecr%1^);Q}alGj1yuQ`x6<7PM#3sT%VR=ZKX5S*aqqZG})%pbRx!)6j
z;@7?7-h|q&ojBfiwLWhAdJ~0^5|%6z*8VtZ$JuQ3r|)Cq(Ct(6jJBqWMKP3{(8s4)
z?KT4C!S0)0L+xK17)yOa&2+a6=k{kj=Qq8MhLguNZ8|@Py0b1FwfHJKLw(wAano**
zGD<c~cDl3jvKG2ReA^yz2?W@zdLFL$t_~g#?y9afXnBX!UgiWt?l=JC0NnR>01IGc
z;Gi+H^<;W|kni%K(MQ)e<|9rVK5<?Ko>}jpzR|wA@7V8W+m2r0zY{bzhTi+Lc0RNZ
z{?(q<D4zBFe^d?1n)9Sq+3pZLIvVm-bq)D;J>qtizN>h@^CG-*0Nmd!Uz;K(uSjV#
zW?x=EK(9B?3y-!KSvx|-7qM;Yga4}Udfd2ls`N&ERahqH4{WYn$5zhuA+elvo~W^0
zS%0stYG3@}`*0PZ_x`x^{#>NzBfepT6Bf)lq4zov^L~&_;G1ydD--?_OT9wp^FRQ2
zO@3j2Im-fcakGyl4;MT9*t6YNo~*~j;jYN3t2IVE+qb)->%mE!zLRa$x=kBF8A$<8
zx~D~>6=Eb*LTlA|3!TePtx;+)*h2CROpgkw0mrIlKWg>o6Ux+NQNjBYHRmP)1%V&n
zgD%_utpd*R+T?-n)PhuYCow4UZjE>!a*UJ&d#N7FXo|IIM)QPN#@WDmZ4FVbG=|3{
zga>1dMD)jUc{?BCy(|j1m?D*R0*xcOTCemX@%N@lJ9Q$mvZ(TFOyS(xT@TY4Bo<U)
z72lz9RCl%o{i%KfNsX_><}5lL?Ao6EiC?%zL@;>YiYL=?9!sZpOqfTE`*_#*wqnXi
z?;tdrizjbm1oMnoNqE|YMX8=CgYP>Oell|sQ5-L+QUr*x6hiufA@(rMxMPdvCIqj0
z1phd+Os(SzQAaA?KrP)Lb+8@s|32>0pLC{Z;OR47!U!hW1yjLm!Md;sNsC^;T2eV(
zwrzES_ER|SqYSA-`2p!&1CL41QStLuw@!PZHv^c|z}>=&#!*QPUrw$DHmxOBK(r^=
zl#et;!DCsE5m>ND5#vZo+DpKhmP<oM6F5E;n*pcVyr>uqlZu2)>G)44yJq|nP&7$g
zjt&s5SjF!;dZbc$unnAf5_|~SXv3E(KR9`GGu;hXl?nAWk9AMrzoGorN>I8d8PC@g
zQ<=llpJJ?&WD|kPDD0jAugIPT^-ZaA9wkuSupg8-m2SmXsFdxp)rmpPE7fa!`+*CR
zugAJe<PvwO7FM5q7}v^n9DYBL(4SVRH(R=l5(UH@{IFu6XWo|>8Ns`|AVOm<P@SWp
z;W(Z0^s^JLFVBPedJl<b?;6HL5@W(rAl+`Kx;XWvl%`2I-SZwJ;c08>YgKZI!`h|p
zGsUxNPdh4hwoSW)sED(P*gj&q1$W1)2n%L~`_J>%!<ybP9j9nB0}Q{s$U@|*U?X0%
z#KDJ<JdomGb+-k@J+WzL8H6XS-!xLoW2H7aX6q9XgI-{8fgK)q1GJ-z<fRS-7Lr|Z
zJRMgBCaozpOy^d%{`&><ZD9u%w{PI>NBH)-mx+YAkVTxd_H_8QGiQDXMp^N5I&~z@
zE7YUF>{nBQ!P$c)(+l5^h|I8C@f*0%1harQ0{}g+Nba0uQ_30GKF?DZE2+HgdIVL*
z^Ne?N_ZpkXf~4>;<@J<wu)&L4IgHDp7G#t1u}qbZeWo5idGF@JpYc&&;YYilAXYN5
zm)F=P#xxj~I5?F~ndl`5IW2$jlt=I#$>k?;JZ}?r&RR?PviHugM=m1%@jYCgZwnRd
zV)Ow`!zO+~$hh}d^4)}Ue)hi%tts96sxc~Gt;}9c@#JXvQ*1c3@P=ab$(-X9rj?LD
zg&Ck-_zpJo27oSo2Z|Tkz#k08{%s?sz|;v>y2Yn%3pUXD?fku(dV@Ig%}p#^DQkaK
zN3NlQ;-P5SH?|tQBJ&Xg(zYxeVNK`RgV49wbKFR>8_*{oF+Aq`>&4mJTa;w$JswTW
z-s=nZ=*|1<QkUT-81SR{W{-!Ql*%7DP+Ns!?hROOsF#InI8fW^#GBvUf!nR3ziq+s
z4<N58FTgWzBdDW&G2L?Gv{kM6Xt;kAhPLn)1rNEdrLFCU28YfIS+OF($o+1lqj?Gq
zhj^%ZQfIg3c3v<v_O;$Qb8Oyww%HMg!rAxbw&KnQp)C{PvugJqVObr%Wxhs~&x>~J
zz;Z_cORE;d7u8h?)QYo&XQHEf>@$CM^-gu)e4)TBjpRGyhDwB0izjf=)wP)3S@B7A
z?_NyzI4DYS72r#oKW86rA6H!!B;9AeK?KE`UO!{Lp)K^!*!?{&G^Y=nbsm}m5uZh7
zA;%>5OuSUv7n%}U6Ei8lKF%#cr>#VGW_~DPYnVoH^m|H(HcxD$aCWL~4sZ62NKPo`
z_Z1m;R>(T5V&Pu398-**LbAJaQ3}MBj}cwi?6?rQzA{yJmI?k(U>0&T<t&r+9ZCcy
zy6H^Y)?L7d4-ym_?03vJ6Ku|@?#RBwCea^=r{Hvxb;I<&aV3JsF$d0%B}Dr!yPFwI
zbv1V`xew!X!&BJ0tdo`*xWCGA#l-02m@J>bb6w;XOf?*P)!pE`@|?B{S2#D3^_w5)
z7??a=_l?Ih_f=PTrn*bKL0qRLhL(#^p1XT|kVHcTT`BG{rhMbvb71ondMWM;$c7R=
zAC*@1#htM|+Lzp4nsVv#TVG};;yXo}7XO$Z<e=5aRi><;whv8mo^mhV&&J%)-|&(j
zFkj>Io#^j=-L1*(7%<;uu(e_`?8I^x!AALly&lQ!v&`G6xJX1te8GTL6z3{?q`!BC
za=)+-+0!>FME}J?&9m+rf9$#b`f<svz`R|<{k1h^P4`1<w^)wlV_RIP6Bx<!tH~Z3
zGp?#r1Ybpa6XkpTb@Nd04=axk$+ji)JqG0v=edn?Zl6cq)XFEDJ+yxKdw%@m>gyl<
zVufIma#S%JHC0WyujYF^K0^<V7=T30ciY6P3xTdeAY7=joz|>wLbtZ(I-83V1>d_9
z4@dNSj6#yduEQ>$o%eU+3sp(nYAppI2UIws2)pm&c&9@sE07@M1fx&9NF+=_uaA3!
z^w+|SviQ{e)@ptakKyQMY`a73*#whfymqX*Usg8k8;@M_syun777X7W_qzFfdpGzt
z7i@XXA2>)i(kBVNI=x<!^3v9b0o>MC<G`Zm0V1C5{yA(@__mufIA%jQ;tU&ma49it
zvD`~LSKzP)CY|ekgL1!Gs@SssYp9U2YrUJi)3tp;+jsp$n2#W$M(^=J?(qd;m2ac_
z#3xJX`uCGpr@t>(Z0E8UkA|RZ86o@?0_SPxpmX!CEkK`+zoOl9?X1bWC~*4w5{7Y!
zVyPELnFutaQ9mRK7uk;TwoF{m;~5>>M9KWUHL_Kw-YQ49H(me(LDyr%#{1TT?_=8J
zmoI#G<esTRu+*uzcX11jAiP@#7kpP$J^)*v8C2<d6nU;AVFhl`_q$}9@d#xqXXuv)
zz}v<9H67sK;{6IiY+`Th9jp;Ung!a}#>mmh!C3#Jbc~IWt$_tJ10x|l;pc#fgN2px
zUoQPWTrEPKe|b#*^5_Yfm>3CJ=;{A923FSp&@3#3jEoF_$Di~6px8g<SQ-9<{p2&U
zv3+npzM%iU{%8Ep>*qWJ2mN2nCyt4QnULi}?`LCYXD4L+u=T%fFn#b?*qNaJV*k<o
zIsUKA|6l)!{fqzb@<Y#GU7s=x^bE{@jeQzq`&iUx`y78RgzZD<lm09C5rj|vUk5)N
z_>29R|KLIY7x%B+KRP&=nf}%Nsqeo|eD=(&9}D^?kpDVQPY?ay@n4(&1mrUipY1ad
zpFR7B{r}?r+y6IF|H^&n`t<R?$@}b?n3xC|SvdaopF#LfTK;POu=F=2^q;x@$Lzn1
z&o%x_{O??TGXFM~|41=>ivNrJk4Z+t&oF#SeTIkrucH52_;3H|)aUq9l#S_QH?aPP
z{&ezR-lr{&5B@*A5B~q?V*1!EA3^=Z{%x%P*-h*p3;u8WOu*j+eJ1z+Ht4^8|C{;$
zHTkLZf49F`|A^|}jDO7i@BYKFe|N&axDV{V#y&5Ne?t5J_Mcn+Z>Iid{_jrs-?l%;
z|F7Ktp+5ri_wxH69sj}qcUOJn=6^17mXCyd-VV6A3F(yG?TiWOL~U)HgpC~y9n9^V
zY#shX%j#Qy%*dKs8#p^E>Dx%tC>onOe@t2Fn>rFQL;sZ#6ts0C)FPx|U}GkvVPyZj
zrsx?6IartobqMJsob;{C4FzmWt&Bem(g`>k8rwMiouvD;^+|lnK+_57+ld>So0>U&
zihm&0|Disk^4E;0`G*RIj~zs)_zy$h+T6;Wkm5hy5q`MmNGMGSO{e5!Y_0kcO+sGY
ze{X4@V*g#Ow6Tq;lNkdc1H;E9NcSI#k&u~@<3ssB<mZ0<XGqA*#76(&--mf8<IkvA
z={p$<8yngh8Gjo1=O*ix?$m0SKqk=0qOF>I<}uo6z4GyGlKjyv8RO`vCO_Se4|qLH
zyr}|)!|LSlPZyC|94bgimE{Z3m}}S=(dwEUbByy1H$jFqma$tnsc@D@roN@5wt)b!
zHQ@BcB~++&tfb%J05FhTDcG&*%PZlSt*-+r)Y`zcA+kVZ;6TKn7@!asR683Cpf|o8
z%Ut%rsOi+!XB|M$g|vmTtVu4uUy(cw)wGnjFpSk8AaJCVP)L<bJ=Mq!l}u^<mNrHK
zJXHLP>pin0GpoH|{5hrhvJx^fH*_$><mBYt!#7=seqkkb1ub}frx)G7=LXle2vPtc
zh-?i^&d*x)BYa+H<FFz!)H;A^TEE`jfdT)B%=Mn}soSckiKWPiiMim+YulV$7qZ^_
z#&^g~*c*fiVd@O&sCU*qiT5t|y2d)!rW*iTT}na52%i?Cy>qr&L!FZoOCsu%+(8%F
zFW;qc86BW<SZUaFSZN?7F~5?~8n8D$#ZouUUjP)N*U+V@iT;I^o^{~wCT2TU$Cl<M
z&kx``XI6(M(029?EO&Q+caV2S@V;S4`Wj{zV9b=%cbyPlk#|d0mY0~8@|=kH0to&z
z|7d@Ir0&a@lP<AVUyQl_^@+h7y>o|<#zH?|%jYiruU}ufYRs^fG=S=*am#&4uh~$^
zsczok?>0zYF@WFezhP<uO#NZsp()KxO%OI;V|W3`k69f+hilRq>F=e;{8w_oN+1BS
zfG*PIHg)VY5DGv6HXgWYNLRp<r{OE#pIg4mcO|9wB;oNtgZo*L@~;rYeq9OdphiIN
zgad<9M;mW&)-=}K?=j?n-%4w@eD&||7!orBQ-d?=QxmKEUBe(#TAUFY9=}p4)LSvN
zUPHA3DDn+VPY0sRD%GxyuS2TfQ;fBY09*Ni-kHVaiMgexi`-WfrVmJ#`H{$3<e3ch
z**Neht*^8Bq0#k5h;R2`rL3&4#`A&XoA-U!zDbg=B{Rz%JxfT9HUu)Ne#)kZizxGx
zawC$cpO9F(z6%<uV;V#Cj14NbUkLTl)6u=8h<aWWr9#LF^P-(EI$!IPtm#hKRdH<;
z!T#G<IE}B}nvz$hRjS)4ZRoB%O}RtRNs<RVQ)?J|I8t;2yvj0)IGI=%6)ifGw(BwZ
z+#Xuj&>P<j-?dxBtWhjF1I=#-kOc5a0wRRi%V3~)D@d&xY}SXI?N+_0nD-ubJiN$a
z(0S8XAnXlVzboEXw*y85dCA{p5%7;?PF4U~@S0;`Y-|Kj(SZrqtb)`4TDP~_F}VsX
zI;N^-yDibKFDF8~&9GZ!!og^&+4N~q-=yS0D>~^+yAmd@M}nuZ_WW#R`i^WGoG7+6
zoBc7mkhmwo3vRfCej(_6oN9{LVSv`15I)@$`(^+*K`8@fG+|~VkwITL)TKr4T;efn
zV;)Ew!|&LR;Y(S%!zoda+`?t9)Fu(bivokgv%Bz`9EM0Q5&0s%?B3|w3D*{Zj<KlO
zcfP&)bFR#_6q(iC?H=`4ue`c5vhAe+D&5wUkXNqoiV4$n1QN%Q4<}y-y?>s}v5{n#
zJw%j_isKOo!*7VqZ`_IQkec>{3Q1En_)m<@aI1;&vz}$=g`x@xAVj?yICclaPxn0?
zSSOwH>UuBUEK$bovj?Ix%|I&0-^o2rpmPujPTY#0Vpj@8m~|xN-TSx9OLjoUseX3o
zF60xmof@Sd*iK(ZtD`Y4<{x+X9l@GX-0->HKAnJo6D~$psGJ+mYYPwux*e8QQDyfl
zbuDgf<iB<W(N!H?_#;8iCQnjLK|>ZqZYR!yEsw07-lu<@w)&F}yfzUz?*(k>(5vTs
zyM%ZQb7n8)<r?$%c|hUBpl*~gA^_I%xbF`8>mLu_=?Pm7P@Spi`Qa=pBH!(lW%DAW
zEEdbm!40tTi5zm<tR^rjE*2%o+ZckGJFr^{mpKfVmxgTNBAMGeHq6b@Z~T`7Y_3&u
zF-nJJ{k?YsdJ17%@?}dH<fTaqaM>_Yj!AtPb-nUydRINn?WopJrmKF5-+h%r9y0>3
z2Wt$?G1wK|*g85rCq^SvN=efxAl{;_2GMYD3Q4zoi!0={d`KcwX~)Z!)w>JLLxmu=
zk~{zdQxfnEwAOD1*@kQw*87XWUl~VBL#?BoP#gQ2H*G9KBK>||8+z=WILEp)f>R>?
zVSUd_drg{HJ!c4e0wTS1Xb=IRSq_sOa$?>TNvR7^Lfjo1I6r^AJu9)|-v08`uKp`-
zHy;C-9W>W|uQf%mK{P-8)}W;j!>p9Dgi3*by<Z&Afh~)rXeMCLv@}6hxR5~B-cZRw
zFk;^)lVvLB`;+w4>XNh^Dx_lGk0Sc`RH3B@ZLBA>_>o{P(=+LhNShWdvXS#ZGi=yo
z%EO<H-+9Ob6an})=RBm))a<uzRANOi{hJUJ5HIIASTOa<o-ZDQn1QuS-wy;D^nRR{
zzA~pfT!E>`hAu{)P^vV9_*Z)Ebg}Hc5g`&cXp0=Y&dr?Mjlr}Lyjors^kVPaARXXW
zu#@-+H#{s>5?gn{K`k3}@t)Ym<jyE$Z>FLWwG~Q=@9{|Pefc=6WKEhGE>yWKInNkQ
zkBl2HBeR@Y9tYMsm(Q95wy?618GjBJ{K7C~2?0LY%%mt^p}|6rTuxtRV3OuJIU)Ye
zTl1EGqlw}mw#VuReq^4)rNCsKQJ;;^0e&-gJ6cA7{af@^hr<se+~1>)mrD{TIVOR(
z+M^Ex5vA0Dc1Z4e0Dm*wB|gsB!;3FpE;L=r6bgPWXuXKnP}aIn@KXdNLpf1tnY4C#
zixZ&P!)ff7!5R?uw$<r4#;@Dys9ie+k3)DCxW51R8797)4Z9w{s(-$t6|*4>+*!QN
zNs4$>lc3xQE-b;BdjntGu4qi1R7DuL-)sH4(@v1cw7z&gKzVplgPzjx2GP!3>@Fna
z+!PX&gj69D&x)!JwR$1E!aj^hvuUgswz6$y3d`$(Ij~*1(Q{PeX=Ww7KoEXfIdtks
z$VQ*jqIQOj$;L4kmt_!_Eb%gN0`GK~oatFK$!)!QDi|5dQN)v+XVFsUi=J-FV<;Qt
z|Bf+@INvaWQtkn3`Mc`+u7IV_?KLnceHLj%Y3+!MDZVad?5Q6E>{+rRR<Z@w+F-3D
zwp|+D(Cd7hrpHak5{7&YaXE`M^ciE6Sa6MAZ0tUTIXNE837KzfGE!y+9{*`qhXTQ;
zQ~cLF8_6UwXO*cMEvdJ9*Oht({M25)^6|URCP;G$3l_5+O_CKO=<2l#{7>#s6v61_
zEXRzpx#A&l@n{eeLfQfGcaor8R`UCLorW{ErQ&9)$pNmWJ3k{_YLqED<q?7#%$VSB
zJldzOIBL80MQMw0mQ;m$WT4XpnxgS)s}(8MeXFScR3HbezHf}IgZV73GQim1n1~XO
zy(rr0Vvt<!CabuNk=W1`-==yXJap;xwk)4eD}}C<MGYNkoPlgGpcZ@RG4N5%=4yfR
zN|V|RC~I&h54gdpLhK>4F|3kifxeY-^B6t2XpG18moZ;>7ehJyf~TglO#i4)lUU@p
z78aRqH_$Q}@|zgkP;oAw6v)Bcwt^&;UylWYCB2rKns#Wr*f+@>!TRuZ&9-zep4NIT
z;d07J;%+I<ueh!o?y<65$^mWkB=B-xfL4IC2Q2(|Dz?Ve%kL_@5fA+@2AFKvJhxjc
zW$M?5DB<_J2?|PQ3gPSBTnK5liND5wt2Q~~ekB@rjG5S@hnz7)t+kJrdgomHO3SA`
zI4##yx?@L%xkSs7n>$wfsQX8bUnvAClZz<Iv4373YEg{HCqz2caITX2m9UVav|gdw
z#aix4fRsLIjOkvlT1;oe|5DwjWe^^Hks|C{1<mm>tkrYUeKH6m;g<~<Fubej2ptXF
zc%SXde(Q1kim5{WR3<A2RJ)6}?~!p_k*pfEv&<Ec^%{+`{lh;bm&Lr23ci@@jhV0b
zl#8Kre%u7DmV4P{`B$Y1Y#hZo^J<>lHc{BWS~c~du0_R=OzM3p^n^*gHZf%j0~gHd
zl<cie)KU$(C0FRDSYNUY)>wcht6}Ap1vtVrrrcR9S`qBIhBwLv`GDj$ZkSFNjJ6Pn
z|3G{k8LXRB$^>S^=zISGPWZAJ`?Z+pUHNA$@0DpbQO9g8Qg`wy3igyNx-$RHh}*k~
ziXr-<ZKAh!9}JcM98X+N%<&b4KdE(n#^snehM#D-^U)7W;7U|7ml|DDMR{AZk`<}w
zgk|{ZSqG`7(Ex!nEx%pHB-Qg@4h*OKL~+kxL(^N>GOojy_e!)q)5VkpA^aO$kaVNV
zET$c7k2|3@u<hnnAO9B0ue3=o)l~NLuDb1hNg7=4Aqne9WsrB5v86n|%z$th=&c-h
zhob4)%%EMqnHl+B8LIeg3wc!}*p&kdIV=Br_mmrap}jZXtFvC<?X)9th;2&HDZV&f
ze1WA@{(hUHhn3yP+Y#!y*M;t0a?##i;IgS?YN<^;0&_#eatNt#$|l@7gZ=Vuxavv~
zfd$-{H@M_N-=68_x5pfRX-Pj$DD)MtPJ;D#&IU}=bu_Fp#ksyWV_~ICHhNQeP6!sB
zr;DiEeXvxNO``<zvhGM_`8y9-($I~iNwSd9epp1z>6H1tc9A`AbyJ#;a#}ON!>prl
zY>L#&w?5ZL*}by;c5_ygg9(H5xTl&A(<eVAA5+qsP&~{2^l%Ci$ppXw?J?!`ebvf=
zWmAcP83C)&H<Ql{z^_}qSH#ZWy)njY&U#uJ6oys_RQvPG=pM8y8eEdkJF9u%x_~O3
z_8A+lR-}T2<}Ai@vJ%E36kuh#sqA`pW!pSz9pt=cysX@pxNYc0N}?+RJnQLtExnbC
z>iBby?vPnQVrU9yFWfY?xmD7w+>hYmix~fO%Qw(AYIK}E7Zjih`Pfps&Mb)xSYIA3
z)BVXT9Wlp!TcUdSrDDMm867U4nA;2%w{fY}Fas&C{<-;W9Rs@LpT*eeV;}GCUG&ZV
zjj`<VZ1Sn22HVzFoRJ>m<s`z)HeguDV=j>@^tO~r3O_aLVE0a1O}jh<s3~V&d9oE~
z+8s6QMUyD~4Fv5azC97)p>4a7rw0InlH+656tpXS{99o!4M0UQ7Ar|%*p>f;189R0
zLH6RfklYJaK_RXUXDF*seo##KsE&X;lJb)C{3v$>2Cv#j>pPFMH$`aD(#Ge$!==|<
z#_X++ildy0=<Dz&LeT|A$l<95cN4elx)Q1bi;yKGa%9GyC{9N2wLYO04bt%nN$4q<
z7DF~0yehA^RC|0Nqs)J{19giO=jEQbMCrj$%__+TR#dyqb#s-ME?yK4VydpM6k0dZ
zrYrWvykw2f`v%%5@CV_JzK4N?e>hCwVtS>yw`f7;3q$d!{bleUv}iGzEPmb-b54kD
z0@C9}{7a}CLZf<_c>`CA2G-d17Z(Wac0=A6`?#I{nx@b^s>Ic@v#(5K{0D}S5GCjO
z7q&+XxQiKiQ^lLN^q59Wsu6N;nmI$ydZ#g*EDBo|NhY$d_c?;MGfsSl9o0LgphS-j
zVLwjYboC)21;#M=?h=do;`q131G4_?_LQbNbjQPw_@X#eDvvwm4O&{88pQoXi>Y3^
zJ}z0}7{7Pf2jYgG;ST51K0Z64UA~H0EdOQF{B|BSvQ~OP``i9C@H@x2KUF$P+to)=
zmt=muuV{^I6*y)}v5`UFE2h@3+qlKI-@r@D?P%n34i=veTqqY2L+5`-DCD4LB=cpk
z-?oPDFu<tHnCF#{l))3>rZJtE{rDlD-`T~Pz#`a0%;kL4`jGzZx|ac>@^;CX!P2b+
zeGtco=qj&`$8MRVK~QEU%Q${e16A8OsChEe(C=ntk_}ox_d<ICqcHVJAzyGPx<m)P
zD7I<{o#P=VQB7m5quWt_iVLs%#CUr<I@E(-it5CP#YMfpMaMkRc|G=Lges87mkYIh
z2He~6e0s(sz@zzPQQ(*=U8i7V{fK*d?e9L3UOv3Mz{YK#2?}JJX@`oS2ZnHX`lE|)
z-D>aJGsHIa6APIFS;Q-WI$QX!EFcNO&UsLC{Z4sPrA$2l3i7yL@qwO84nO*sY?O~R
z^Ju!&WMV1o5Fw>lmFvHeZ=AEhUtApz5oR_KJ!KxE+4sq&xA+oWAp5*#8#_vTA1-h7
z$f0x50oiVD5jL_Mq383$i~wa$t|f@pbO;7&Qv?E8w&0iFeeHQHJKx&&{2CCi3c?N)
zuZ0vaQ0eO<n?w!(N)ZN*PR>F-NCsnV;tBG(sEcNryTqnZwz2QEXt3`N8IW=xPbp#e
z+ViKiSn?jOpr-(uaL>}<<FRd6;{>4d$A~jEE-8d05O@I`gyymfdL*=l)j$j3taai_
zc;D~C^nvVQ`O@;`UEZ(Ao(bTxsdJqb+w9By$F3y9v0(nU5FXTf749WToIXZe9yhw9
z9TG*@&@dD(Patr6I)OQilF#Y0X>q9lmN0?0dCW2O2S3KOV*{{dZi(fA6ItW{L8kzu
zKiW+-EPXl9NP@22yRcjd(9IgRtp`<aj|}0?Q{7_`lN(VRP$D^|(W}IqMYU+-Co@s$
z9cCmW5o7C{iTv#i?h<3f9&tggjUFPYm}D!{FI~RF-@!B`>{SxmGR?fATVdb>)@d^2
ziBG3-v@|6mnRqsjr!-<QuxP-btsG~2yn?(T6E+Z}RttOv{dT7Qv<CJ1kKON95=SiW
zSPN@_fqWA^M5&#ox4BULP84^!F;bBu?A(%;_~q@Yx2@~@Hp4YS6S3*a#}c+F9zRM3
zz3HLyI>5M*=#?}`=dB^$NJ!C<A}RCtF3io~lkF6{^SZIR!U2>gk@MJ!h`AgChkqoz
zpXVNe`uGRTOqFzwpx3VcAfFfLi%dmBf6knJXZY6msMRix#k`(9JJu8@Q8M}iJ>Gsz
zh`f008-N_2{%-TTT1(dXx=j}{-f_x)QjHf@)IDTlEe*=uAmU}udU_e&B=IbLY`OsD
z#kh6`madw43t;LlOTyi+1MTj5$`wFp)h*aBXnr)Cn-WP|d@SY2QW5y}6NUa^lk}U4
zHTSps)~x-~<#KSK37J_<D`TC*rs2^J(Nnyws9FA-fdxR`<{zRqDpj;#2E55pM7>K!
zlyqTG6QqVBdIysP??71??LbbY88GkPLq|i8^p|XcPzp)}LTz;LBkf;z`a}wBt$iX{
zJ;(@x>z@B;C$ch?jh9jeA}bVxa<t`Sv+M<__zJ57Um5S{6LOWa@V|AiM|$HrGL_KX
zGq>CqdCJ=s;La8bVG(jPRlf)oz>LL)^|g&F1Z`Sc7+>h@nS5<A!}rMirDpj{RkM*9
z5H>}cPbKrmFbL*YI$dxYvfRHMWh5L6?I#Bnd3qPuu6EkW=n963O9VnBFXj(@RhgBO
z3)9KkSCIQ(Xz}y4%o|*nsV&&?k282v;_+8p82V0v9(#t<r)?Sa*)JZJ8^Dvnm-Z$>
z0Wbw-?14s{snqFGi5Gqv)Yo2)!1<n9!c)9AtK?;>#3(xys7)4qDcyD7Anwz#&PB~V
zuSjp)=5_UT^vZsSpm}82`d;27md6z+b{|z4>S$%%K#ks3aEyvmZ!v4^-Q%!U7M0F4
z*swE;%(^2leI*snpbIleDd#Y-6kj@*2;-d_d+d%rmt0Vu4i%sqKHk~Czw3*&&pY;v
zeG-T~9tpQg=S?fY>G__F8+)U+H%Cd=f~dApj_V}n04ITLs2iFg#?%wL!Gtg9JkKOt
zj{NH@r^Os@w=+|hs<|h_^s`;+{`Nf;RiJD6<8aPnWGUnl{&rCV`>eLr)yWD!F|1G)
zRyw@469+fCX#iis;%d0_cH8}+Yea*~>IQZD=FvQ?uWYR@q!#%q`H_SjAMLeb1x(Fi
zM8Gb56Qm$UJ+&q_QHZalil3w6wy>3w3X8YfanyWtbNiiEVy%aYkvaN4CBDi|Z4F;j
zJLZ7z=}2563=ea)e0pYU8c<m%BIs~guz$LI@1K>&>m(Hl(zboIPNx2D2cz#si+@f>
ztAr-x6vx$LxdI~+4h&u33J%<M#NAw2`=0X#Vg0^Ls6dc>6)G^KyDAdIpplbzkKM24
zD@KT4%4PAHx8ho_9g^ka#TU5|u9lt5HGDno##QGWG)3xv?M5_H??q9CgKIQlR3<~F
zQwxO&6qyCpq3zONPZ^X1!&tscYP_yoF2f};QoaycCbja@UHh>XIW0g^LCuq;sUH1{
zXuZ8yi@G7J-~!p$C#GjzVN2gJs|fbK&hZcaqjb3d-sPRHsTo>)(ZlNH@5X|j2_g1d
z|IP7gY?0FMP$3Xu$-e^Lkz@IWCzQS^;fZRYsaO4pYyjEvZVwo0L$;A@6=2dHTTch;
za^Cl7c+Mp=5BTFeLwf7H6#=s4DGG7-F1v-WX0at61dg4X5GNxMGm&YSh1xR;^Fs~U
z*HsxWE1-}}Zp$|u9m{pm%0i2b4!eJ0zvpt4b__Q=U0oeoC~X<!(y4}q^Ec+9?>MV0
zZD$v;Rc~E$u*tDu6n9MO9+{oOW;kIJ&Mk@YR+ftDJt+CEHwm}pY0XIHWcT{RT|to@
z;7NRAaHcX0!?cz)M6*|)uSgfAC%<m|^)UhG*^D!mp$qJ?4&0Za@e*Kl;zh;t_!O~(
zvU;&tz|XZ%Vb1Owi!g8l-)Vnux$7c*?+mz<r3&YvcaFM|whOmNYAJfp{^ERpX*pB}
z<nZ0%+fOO;4PF(2wN5%tr2D4{6O6<Xh;UA6k-90Kogjh%m!zwprGELt;u<R}<M9%G
ztlkx<aEKiy*;~}Z^h{pLgw;cP_XGh8ql>j6BSph)%gUA0-J?~FoAs?<R+Jw<bgzaQ
zW~YL{!WJgA+lR~Y=?(tWR{xoEP}J1`V^eG4$bT)E>qPf#%nTpb8h0A<4EdhrfHPc`
zyWpUh0bZo))*sI;QW0vjVr@rrkDyEA2|ZiAq7rjDU<j9en;R3+Dw|-lo5_$9oGzHc
z3Hckzo!D7$je42q3}m?<d`nZV+8H9J(<au~vnVB2xwW%L287&Pu2|Kg>tcD^Hl{N<
zT$3g<AL4<XbI{JG9^kHT?JJX41Zow4y|i0`xyCUbqqJ}it>V_L$?6nztI0wNU!Ghs
zeWJY=<IlCu+OME$rIq+4A+aGWF?-d+?26J7O$Zv}S&nR6b^3SC12p;NMX~&K+iXf7
zXE1f2Ny4*$ch>4w8ghPm=f^3)bs&&&ngu^i$oGi1Gb84aDft(kWhm+J0s}Q6A70C$
zZ0M)NuUFckg!9%>_^nv|yKfJvS&%C8rUu+g<IOTDH^r8VzRXB)fm@>Wg-;a8G&>Z$
zOWxZ0LaWl!iMU@_`OnE8dq@kXvp?3{Ll)I(?vTwqZs=edD8daqC`6y+#RNgwpkHXW
z9{4r4-C`Z0bcJ~{5~R(Zd<h=x4A`6I6L1(K)kT(73E0Y=%;m<c{h_#&>+{vyhiIo@
zedVKs1=4zj`RxAe0*Y%UsgVzSW#shTNU6*l@6Qo~3V7q{W>Rt<)MD+dri)Iiws8}$
zCpysj?!iYouAnv<hD<)1tCx@8aSoUBm*r!=S(`}^STW+yeH&euLj3L9!7&W|f|Zd(
zJQheO97BSGu8?&O7>Tgq%~Kr782%QVUs%i9#seYf*>DtcU3%DoQjfu;NPlHYSNm8Y
z!_z>e*G~xZYf(`upzY_05}uK-!`MvF=62bT83jwGEIjVmj|BFl{3skbPTYO}Ry>La
z1N90su(v!Cb>@t+ph_GVmqbC?b#-Vf%vQURxmCRBTw$S;tXXw1;4<m3egqQ+#7^As
zq$Jcz6SzXEllDWgmj?6cF8%#)z&#h0L;2I-oiSvPG8>9k7^t}?U)pqk7RHJ@tukFq
z!3rPj8i#ysBZ_rC{!;#ZINZMfJ$s&U922V_FTJg7e4~*W{4GqbNn1)VnD5nZ;0X_>
zMb`7}00LPVK(yX?=7Dx{DumX%c!07KZB2kv-4}$etiDNgRiGr?k!;Wc#LU1>$|XVc
z3s=$MM@0Vp7-fHfICUh(S)Esx9;UuWV(wk+7prJcC;p;lQFGboZv`&aulk->s1^6Y
ziqZ|jM(WDKCiRF<XvShsxO>m$+~K{&pzcxd@LNU-N^Z{_shzc2eUnMKImQwKlp9{}
z(yk}sB@Y3BNRbT*EZ9Tpu@+51r3E8{(2#-H7u9e#Hf0U3RYpBz#_*_8!x;<$63|rB
zU9}2++h4bLhABA@CRWXS5s&KO46`Xh!-Ifdb=PZfN&daAwmfO9)hHC!MlGFLHGC!z
zLF2cSb``V&Gx`>772B&2AgyzWp7w=QXhLBto1RuFe@apDLHl!%6T#p=K7{C}aIf*l
z@S~%xltPIvH2BQ;u_p!Uz0h4m!NT@7Weg9Nug9&y;zSJPh_lIa6Mu15Uq${SJ6A8;
zqvxfU=Fg&iRzZ2JekL;q(?5l-LKuUHsL1cWTJ%g?EmJ!2V^y|YH9M+mp|T16_YL8*
zVKgs27uHZB){gCh2XmaC8KPGeUpa4WDk2f5o2ctERkX1MaiUsqNr<v(tCW`$zht=F
zO|?*T;o|`dxp=UOMg`b9!~RuHo<Qm1yZXy7X^Z<7{`<Az0;(oVRJXjI>|JP^=exPi
z^J1?4?{sG6F@Okm-1Cn@0|#wZM>Wkw`VW-aCR#Q;e|)L)dLkoLTq{WJ)?I;(#^Fe$
z-wNZXtc^*aeZ@tj6xFXrhG3GcLzI<;i!ytjK-IBhSZPLVG$6RQ{LMX$q0K{GVX&6M
zQtvKs%*%*8>TXJ?3%`OP=OynTI!8O!^{vuSll1G&eJgJ>+YYUw=2FZnzBWgu^bZVz
zVTRUnMbzSG!3+sA>lF#`5GuAJZ5JXSlVO^IG{ED`Ko=&yL%5#iR(8;C4$mgn=xs#5
zqZ2z~eE~Iab?j6CVGFuM!c|5I2<In@&2{K|6>96X8=fi@+a~Yg8VMK|DX7oxmph-{
z_Th1jG=0!Z8MfpY60IoKhTX(&%Lm#4s10btq?a}Xp){^rr1L6#R8i4$B|$<Ki!<2C
zn@95%hN=4P6A#}G1)?{89IYJ_nR4t6zub|%eqK?a5PuxSIP~;rZti2nu)QWd+=={K
zRM9m2T8yg(>vB~d1CGpR&yEDJ9i<_9NALC4yxJd{SWyK_Tu1C~WSEA+wI95=_;tpw
zmyMDc%zl?M@vcoOvTBxUF}igd_ba?5CVYH7<NQ|x%1U>rb}Lc!i*+WdW4ni}oPj5F
zPK<nbQzS_c*lc4#lE@2|84HLwc!7>)eUQXHMS)I^Kdf8*ETYbsyj)w`?LwSs^2yUJ
z#EOLtQN;^GJ!`3Yj?DlsG=kbaX=-MPu|Gs+5nva^c-^Jz`)c3IFz+tdi87jxBqOgN
zPDTgb3Niw=ah!Fmcv4%~=0JJ?`VTYc%@YcH-L}Hip|!4pT*V63#?UlY68h&vI_b2*
z0<XUg79$SpTehOI*m>*6GP$D9gi7{6+|$)grX_b1f6usbJXr*66R)y=g+>C`-MV(3
zu>q;=25c&ssXv(sGf7-cG<&8DG+-OizlAV*35D?h3{IV`D~>f>Idj%6I4qP7%0-Ho
zvzwe;<?`)4ennL~W3NVK^Z^YFHe#(Wn6x38$xz#F4x>rydM<sOBF`lhHO?9C*Jwlb
zfHwHOGK*=p^b5zV&3T>MaiF-?jpvFakFl`|)41-m<brPT5@y67gMJt4z4&G$-up?%
zJN2!EaScu7o#Xjh9G$N3Nn1Z^Q@>>1nk58yKoVtcOV1^TzZ@XcIfqeAgte@6kOHM{
z1g#j9?VyO~JII2&hNN?(&JT*=if)nW)V@S`z-kk*K(Bt8JS-6nIGIwd9xA2!L!hAR
z1t$)K$HM%zaST49a3?JiOYCH_AH9lLKBIn=|Bj3SHX4<-mdqD~%o~~Z&rv<*Q_}(2
z2_NY;^4>Cy$(2*O+b!dtZ@+S^Y0qAppUJ|EjoFW49F~?0gT3TeT|p^0B*B~c(F^C>
zChV~kQr`3>3@1D6qBfbhkCAT{dN<z!07Xdt9c_Tr%iE!a(6WHN2tHG<Xu-Bh1(>jP
zIORFykHY3SQ_BIt>=%#<#&*UC)$Flcze2_lb9<%<U$?rIGl70p@t1&Qge;Obom`v<
zzn}_&a9ay)+U)Mw%|nU~n}a1vg@yWVG>hrF*5-|5h{R)V(rYOG&Zt*dzk7P)hZHaw
z$5}7J4G#vDDOtElomQ2_=Pac!QfJ`vSwTx9pexIjygQF-B6;PMWT6e$cp43{0b0s!
z51B@<;2$f+Bg(4P|0qa7E0;N;REr-sQ^)m^%`*hOgOVF=Uz?A2**p<Aqf+Si{{cln
zy1!zIQL`D2p*U@r?1>jDu%yhe6WxwEnWf&5-Dl;atmx9-YnG8bs9fxPRqScR)RJKw
z1D&@5Tl_q36u$GDji6~js~w#$iI6YD+NnPYvvy3bj8^EH@tDm#NVt^$C6VHJp)!BA
z+yMw){#Lz@(mnDD)I)TYt3e{2HmOA-u^utwJ?uL{$|{*Dg~N7&4R`~M06~1e7z@9j
zNcCrAi(6yyJ{j|A9dT}HeFVad%W9{v=%86ZJ&*fkwxz*u81yozy6*N<sUQxiuQB;j
zBQBLA*BOeMK~#TVBydru=!F7x?&dt@9LSNHg3c;1Fpy#mNael}uLtsDODR=)Q%?cK
zLO=HUv#`5Gi6Fsp-V3_p5tpX)3u<pXDloj4_Kxgl?hk={f||bU%V9HHM-50YdHi#g
z!0|sbtLg~OVj2jbau3BW*xAq5O4Jz`!F<0rRgP?*ucPoLH&+tGp3p_-k57rXt5kn?
zs8liMo_;hSemZ7FXGsf4A5H_aq4}1Y%xN&_Y9YW0ftvgRyY(9Z<xi;^pmW{%I(QmF
z=6)W_vd>mFCr(372?SPe6K7p13PMbbuo-qhUsQOZP1cdHUODPQ$H8OI79MogRg(v%
z#lGOH>kLOf1}V&4OQ7ITeZa+jySo13RcVfVp@LND5ZjghQy%i<5k7%>F?EWmTJiN@
zbvj$bF&S%Fr=N7S%$|ANM~m#j#4ew*IXTxyS%_qtIkSp@^zB0m%;224(pYTPa*L!p
zTT{<E!EO{-KBIdg0y14#^WN>2q$KprVuI{8Mz;4mt#jrSo;V+1^v(2q4E%MdOnftw
zL)WX+>|4!Jmdln4ANy3sAzI~jVvyA<nfol7RhK3#($rDfAKwBvVRs9R5hr5>u0Sv@
zgCm15FjP-;n}oPhuEHevX##`}gTB4SVeX=DZy_9@>|gDeSzkDyTSF)Vy7(HFwp%u-
zy%CHgm*n$-!qAAMGk1UsuZ&N7XQ_^sd9~n6v8r;}n^5j~M=HYL0bgBO;wT@_nXQjh
zd<`EOJkN*nSg5tVm36!zGn~sJ0J2)Tq^UPGrl;3&^!9p^vPoLlrp&L|0^4SK0$)rT
zIgrC)L?0_)p72omh*t3x69Hix4p#O9qf7yaDo@0yS_69<_hxmTO0ojx%<I!p`lNEE
zE6L+n^K*PsMf~?ygHExLlCEtFV(kd}F5dX@npwFT8mFq1rq&+P{Js_PhQ{YyyNl8<
z0Ay^op-D^4g7wVh0(H+t3>=8yGqZUhtgXF2gvc=zeLM)1$ejD{YiI^O4Ib6+R5tOz
z>2Hw0I`Yc(i2NT#YcSj0;hm+rLNo!}GPkI|UQf#f6f)E5#izCmq0f{<_n|^)myun)
zo-yb#v$xYA54@0%y6olVqDbepjs%p>M;e_Z-yW6wlIem{K4HAu-|~E?6!z$+J^fCa
z)P2hrf6+F^{Ei;PdQCJeij7>PPY%XvIb>|BDBFk{@aQECykbA)ApLwGm=iSAgzx*+
zzuDkNTD-8D2Ty~yvz_q?XO54l#QFoyw!1;_)g1%_HJ8p6js%7O@@zN%&r4|*vEm4`
zYne8D)~r+7%NYvF(rW@?qT~s4@wKN4(vPuAS+B8{<7Ee^m2aN_bio6Q>|+{;OOH@D
zrx$1q*$TNKgYJY6EaM_|1gd4)>*&r9nCxCHj2(X(t^^M0Ub}1CW93>LqoA=z40VK3
zSdYIu=jyy%i#?X+27qPt3Nju)T@IUTT5?9Um^JgrQ2gpsOp+TRC263FjU#x_BQhEZ
zJWZ5e6Ra0FXN=JPguZfD_hL=h42wu%)u{FzzU;NQMEUl^nQGSr{>d<4b_9d+vri>3
zeUI+&UYR=Sr{!-`h}vwJG|1%|TVDWp7xIPo=3JEs31B!kf{=$z&G4`?Pz{Og-o=5+
zmg6f+ckM<6RU<~)xFQkt6{=3K{+MsqeEy)SOjf=aSYh8UfoV2SP}$A8idXBN6IK+f
zECM$r{(9Y<cvam#V4;G<2@T7bT3)%A8v8IXDNw7%p_P_`#l1>nsnQQ}R?+l(->T6)
z#@u^g6vldH$23p-<HaJQxW$whKEUTHDW)mmP+{{xt%$m1C|jRrplM&Uh#5>v#L+j8
z@`+1D>hEdW3#eI>UVqde-x|aWOhQ`{cmIjhMG)T{p3xk+KixE3f0kfhVXnZx)&N2c
ze;x$IOAJAhR7gq-*?B&On%0lY2dcQhjWP0CqI~EkobCp~Hli7facHt=6x#*Zx6Rd@
zgiNVZQsRoH6IDCm%C3@V0zc!_G-y)FlHAma31Hg^zsV&{2KFDfv76hM@BFeGUgxC=
z5(A{7?nr3%5}>`1B)Gxy_?+;XtpS&z{O~a1bydB487_+$z$%nObVL+QqxjmHUL5W!
z4Op*}gZBo8%DIKJ^|Ud;)+v+dp&!PDEW?lY+IZWE)6xs^LqzQRR$8NK%$BcSpm>uw
zujv5z=xAC3zo5N9N|XS0K{*crzz;9qep3^A79s2P9-eGzd!7U`iTtWH$GI-hJW@&#
zu}=;~B=cv4(82^oFXcnj^O;g3)+``JU8&&nGBT#uDl}H@&iE~$zAO2;WESlqb{o7m
zEF9x+(?3sSmfSq@IhwD?>Rv>2_@X@u=s<rpU3EMgy;C4~1L@l)JwK4u#gFHI#lksD
z+#IDroE~$L6n&IT6AFCoEAsEYZC4hY>doe*!QYYg0P)@6S`9w4r&~va#^Q}aZjjE1
zbzE*)RHV3`=B#ABsc)c^X79quoIqrF#}ui>Gkf07TIq1sgc1j{7iE8t(sU`&sEHX@
zb5~!E@dNSPF#Sm~bLYEOj{vQm1W}RKd3fHP%t#I$I`#QYv*Inb&F<47^6;FWQFQ!l
zofj}8uAJFjZ35P6!wy!f4E&h?#^?H26E;mBVwPN==Uz3{7QU%(_--AVC_)(7fE;pf
zU?kt%u|}(4^M+%&flOCEO_JR4TXx-OCt)<oyaun~yP3&i;fLMcv?DBOZS6vsW)pwD
zsGzucSp=S90oUT|pR;s^XT0{UFr?2nH7zQC69K0az&vCSaJ>-ihq@(O)u~Iw9mp-I
z&|AAUQ=~KX>_r3zgdrn^*Iac04xT_PuJI3gDRr1JQJi8pE(P2S4}m9qBAhM`z!f1T
z6%l*R{-Wq|G3hR~w$fhprr*g8+J=>DCRF)OH_5N9{SMdZtf<<4o<DJlAfG}U``W3>
zW!96Eyor{6r4n7ns2e>RM@qYq$kXGwh&3xnj16vR=c|wrX537q_-7~U9^K@3#%ivm
zZgFGP(z5|9p2^v|o=$$seNTo*PovUbaA;?5%mLcF;=iKNzqW9wi6aj@lH7+(0MJ3I
zQbWt!JS2W3>Gju9-y+YgS(SE`XE;{O$Knggn4EM4!hZ~NjlW7V?8~27fkU;5)XwBR
zcB(-tp;{0D_YPg38G#q*A~@(T>p#qF&)AgdYu^?;S=RlmC@al``oSxJ)>@!i?qz@_
z*8xp3mLujj_F6QkJDHFl<$!7`Vas0|jPhO3_bG)!k)(+*aSRZm6;AUSc4#kM&=Z_&
z`PG;epv;13<}fFQ$5CYGxRvTwt$fg9oab<3?1vAALM%U(wU@JDu&3K1zQS;49?6M)
zklnmG_te;4s8&s1!$=Kfo8MJoL*xg9M&A%!Oa{wuq5N^n+_|-s3v^!&3-()O<`Zzh
z2KjQ0k@6l=5PoysA9y&52{Fv)W`-8B*IhzYeww_|lekkTPu#a|y)dc;9TYbO8gR%y
zUhLC9^p()zA!4}E!N<%Vj=|~hn2M#1d6YfntD1U5J9m3hQDBr#iXu~~8;u-4<<lZD
z7FZW2<JraIGQlOKjDy`@teeDC2$Wr_(UWFpN@O0|{nZ6w(MguSF3VyhpZEe<;t$f7
znv>DXk)PIj=KFL|8Ms6&J&q^uUah)@HC9%b;}c4XM<xWH6FUW*wokk09PKa6=O=ck
zm+KDRzO+u5Oart`>^fc7z>uJ!em;A(69Az-$}OgzxC(}CP@@o`lm#pZs<$xB<1_n2
z-9F1|{6j{vnN8rpf}5b;&>U(Ml^w4boZNuf*-{>5roFkR9vSu{Z>mnZg=i~=gcJR2
zHtZeiW2rx5#J9eHv0U^Wda4?W32RUvZZxDCb)Bia%MZjMdG>>pA_A+6-PBQF+0X{;
zfq+M5qDVc1_xlVsJ|mnWCGd%M>=cs9(37#aJ5mD2M=iQLbkWXrAvC{waMs==6oj|Z
z6F&oO)6ogMka~W=@;njdE4WO;Y`7Iz@F^6?c5Q_b(Hq@u?qh)MitmF$EJc24l^!*e
z2<?@+bs1z1;6v%crhFYS+l<+Gq(tXNEB#Bp*Q9Qc35+UrrT{}T4nUYZ9DYdXRR-N7
z)@e=;(mA2XN{xyqn0zNdTX^aR#@#`XdrtO+R)EV>jmccx#Zv`ZScKP6@2fixhCkTR
zc8n?FDGv^oW0P0E^N0}?NNr~96Dr<bssUk&XdPs_o9HErBROZ?wA$f@;mMA3mp9O-
ze(mxr3f;Bo*J0wz<-SocV;M+GdXFUmXhd7o{2W{b*Anz0i-<Zc2qNXj)kD#51Oou_
zZPBU?esGv~{g%XVaK*8_Kn_}RHH9nshOu%&P!hE`Hxb=Y4Xhi>LE3qb7YGtwx}kRg
z<F|1z(fsm~+s5Bd<d!VwVGZ+o7XkyeS$d1*G)ScL)yfcM+}}8f?LD;J(rZGPO=GT~
zAvo)NAGLu|t8LTy@8%+JKjY*lH7yD>6`e7vQN2My0ADFc3CS(@im1hO3f_>Z4c0#D
zhZ3y!?QHOD9erRN_=Q0TOmZ1ic|U&a9zTjDBfIXSM~o9hhW6H!A~pW{fYVzMA_eQn
zLRWSz7Jg98k|L>oanDmy>&yMBg3gbzFjVo#1*_BTX4rEv-r*+JXx+T`tDrE-1_&sh
zk6n^EqRqKcv!z;Ed6-cqfIJ}D@I_b%&f}+Op1dEL)klIQhH}$^c)JLH5%Pm0np&$7
zIpWzI$n{Fi`oN94tT@w>Rd=NN{&aK0134J8Q7sy#&(j6-xn9r=5$|xQDiuSRG|KMw
zrtZ0!#TlCNGlf`}s{yKDf)2GLQGP9+925p_VM?@(427vMO!O0cQyVF50O74SN8mVi
zh=-gkgo1%8<kOD7>z$TxQV^FAQMa_chUb%+knPln3Y7n&xO_+_DM?ip(2I67!yKT5
zqo`qo1?Xz;Vjk2M=g7|HUh<=JsZAX53nno3)HS@o8iIs$sywT9P_&0$8;3`b*oqAr
zNqCxXB<(B=MhE2xRenM2^;WuvEzh@IrZB1l(_-0O{OMbv9EZKna`mHYzYhql5q`SW
z)Ef+h%cUeZ$<OF%ntXS{W_1kZ$lRit--35=Jn5M9A~r4mpmpXdMw5WHC0k-HvSA`_
zaiWdr{FhHvKzs|cOH%eoeG>ETNWyPu>7Uwe$%wYbn*3xMUl!p|s$(;EN2>u?cU`s_
zL%cZR>~z)wTnrlSatC(aMi5<lx5ni5z`a=tTVx~4%IFf|vvnjs*!N^&4%WmrKiKN>
z60?ed)sP-1W4vKu<=xz`1J1e}cg3CkTZ~g;gNJz~XMQ|itrCUAmSqJDK&C1e_#K+2
z-L%R-8^;!ya>~rMFxfljl<#E7*rF=snDEoTPfkVN{P7zIt@~BMPN5m$vXow~I@LwK
z8qok9#%l50-iLNLI0GYV%l2e*Es)@FTsBV7$-@^C%SaBCHwNcbf>2eRczwqTxqUx{
zfvpU0ut+ygQ`Xv>`HVViNkR#}e?M9{ZZ*WfvzFm~(L`vT+Fo(KSu?!7TBp6XU7hub
zNDcpRzZunxTj&1a;3X*4tn})5{F;kzwotmImktFf&9w&%(~nQL6~9wg#P^l}f<sX?
z4JW9g^?k4z)8>aVhy1eV*%mmLAX*n!BvzO3B<sYeY8X*L7hl#eLdn`jk3r&rQ;|}e
zgpj6&)TuE${FqCqw~&-a{yRAxbpo=2rl8<yM6n<33spSGU-?f2S=hU?8R;$i2M2=j
z@KAd!o$YS5_Om2b_l)sb<)$hMgjdG9$R&2JT9jlvZ!S_kx};M-h4Va?q9R26+~x0K
zLg-Lq0ZQ<O0g_zvbeG1UNuU}yqg0|M&Y}bD9x4;ilO8om5gD*qN006GV@;o@s-4sU
z#aV$9ycm3HM$||`Vx3qdWwcT7v4<nax<kVA!fZhAIaOUVejpeDk?WFrE=K3;qnk6I
z3(_i@?u#f>xsnI9R%!z)i!Ma>b(MPib<DIP-bPA81jF=iA)v5new{HeBQt{YS=O%<
zbU&6-pk{2__;aHZv?-j}oprc3NNs(&JC%UE2O_bOq`PAEl4smA&~Y8iAA;z%i|IqW
z4J60|>pI?2c9I?|6#Ii7%_YF3QY4(Ob$<+D173J$?&%(j%y6O4D^7(;U^B>ugPuW9
zkxwpW4wHuXn3psHq`Rz^CW2&1ibPYi5~tJDuPhvkDkQiJqS#^L<*g6=VixNHyvG(&
zGi)9$KDt^Bgnd(8JdOQMo51L`VXkx<_B?Q9)#;z9!WkST-)WWagStaG><{QP<V`Ia
z+O8rD1>B_yj+zc}tTbAVvhj!ZizpHr+HXqk^&-P05WdikO|dhwD3C<_MeHvA6y<2e
z5Kt>tFIr7dg3?G5d)@cjSWUK+B68C%Oo(sJBc^xe`Zqd5Y@0j#PEd`G=^?;0Iz634
zWJpc{<A))UjQvV;J*o4U+*Q_^CrQ1-0NtQ<1uoN`Dq9lOxCe`4S?m6U(qnQ2O6zoO
z)~d{{Ifv<S5RNS1H{p(EGwIB#dUi8-8j7gWPUXvtNtbfJrXB5?$MXqE>KBz6(-4d^
z&Iwp;<u-3$$UePw^x`$nt&7cIDSnPY{G(l#Xf!uIz$FXnhAE^({=@YFEkFNrmVJpT
zGI9{6Jr;_tQM{;R(yyM6c!@j_nANP^#x3&;{7n{mT@}Rl@u(7}W=zNK&Ef&n>`-LE
z2<k^HLeq}N6`eg#1d{3aXuMeZp}L}bz4s*u%k_lZ`Sw)@tWQjO(qw{%Em+(43SzjP
z4qHH%*p1k)(>d?26K_+KgmJForws~reIDu<mk~U@jgh7{nLZR~$%EW?{Tj-6Sy&7T
zG92VhUB{he5i+4hY}c_U`FI-=IAeCVyx_PY(%wKAd9ZmrGyT*-k#`aoYiJ&Rfy6dN
z``EjxJf6bQ>eM%LlOh7b-T@g}3c?_sw_ljeBiRC!w1nfgyO_!f2Oa5TO+;TPS96DJ
zzwXgZX0n7QhxK-+$rdgl0%q$M{DntN<OwxA?>5w#$+R;^ZQ%_;`5eH1IN`tNKysy|
zLn;H?N#8_8y(4?$H+K57#HE+%-kr_CTny0OOs$?Cq`T5yJM>mi*~_>em)aj7=!Av0
zupJd{qRp&P8f3%Vz2LD*I*r|I)ZK5`^SdWdNFRHLq7NMLc3F#<WeVof1e=CW5+0ws
z;XYJuCMv`vlLF<B<r?>z#Q)>RznNF7Z6NjwiEZ{YlrPVd=nM76bUoOu`)!_#pM~)L
z)|C3vcxDklD&ns;OWXVof-FjUvr474)X`TW3)iGLm|KRN@UWn69L=P5Zx;rD7aiI0
z{XUU6fU08nUwn#<rXwgaJ<f$GcQ%)p&nv*W)0L}+S?|(pcftHELe5ra;p>EQ+IQal
znIH*`7180cWG&lZ1NXeeximk7Ea`Y)E?LE=nGv<V$XN*TL4rkqq^Cx&h~PQH%F9zR
z1lpwjwA_Ce!V{DWeODu5rG*fMXa!?G%rEi_J$@Et<&5}?e8>`EHoyYG)QTQL-m-ET
zA_`rEF*$Y22~N()iwlg9Km|#>GW~4IT%1@R*CN_K4$d_z6ksRWW|*+L`r+bi!ts4Q
zmq_yy@Urr?6ju(;uJs2Yc>s&LPou@e#{~g<oI}hy@;rbbv`JPUt~RvAHydrfq5z-G
zY}Y=uX;~1~e?_@=2B#k+g*IjwjQT}Uj7WXNl*Sn-9NPD;yb)if+?Vzd*xH|g+~{{=
zepZg?T9JG+3<-`bFT$c)U7>j(v6kuPvthy*w1pI|R<oA(?{e1Zyp-sQA)Npk7Hep;
zZg3k+M6YNZ^_;Vu(86LPl!5V4i-&jkV%+Um!CGk#<e`aK^fKC1F7uctET8RmXI~1#
zt}^k*FPghxRBtttxj^j#A!$zmm(jO3nN2(FBvR+^MRR%G<ek9Y@;sRBDC6O*rS}G=
zdYC^$*Y7NY&NpGotQ;|4BOE5ExNDXj?eVY=!<IJ)9ow4twpax~JiU6>`d&P?1IuUp
zviH{~azxZi(9*_?DQvuVe=tH9jo@|F>W5rJW^!-+$v-euaY2~IZ!g=3DWw>qv&Q$r
zn=rIQZ;?V!=Iyz=4w5byo2^}keNdhZD-xSY{FG`w20sx_oXPvox+E;iZbS<tF3SB^
zd=8A)ytzkR#gA<51n5hvi1fpVply<*j%RTbaVxWZeD^!{q0+^`Vc^^oxGgsBE5i0l
zu6ia$o@JjabHzCQl_EO69-}ZLmU6`;9BC$oHW%hTE>Y?Xa)&%DmL>Hs*W*^;(U=F;
z@bzV^&Mz6$v5GO9p-q0!j;<6fyY+@DOspWY^EO&UsN*Uzi+_Yr&VhRbI8UoBIRxM&
z^w%Tjty;Jb)`Yhpq_gl(K@2*}R>!rg$vGt<5`dstEv_a%+QpG^ziCord*L`Ql3I6d
zu!tZ|PEC-cGDDKqY3zUhs6rGT?6%Clqn8^6jwxxC13MRK;h!3n5Q<I|k8j8#qwfZ$
z27As@hZK|qB3SUZHGo`k7U07h$u=R4-_eCd7bXFA>0=EEZ_(|&k*LZ-kkw4Ft%&c@
zaAJ<mtxNuZ_jQ5*W3QufM*8DV==Oe|4drKb-8k>zFXjU#Kq%EwsVt2*f4^UP+ur}F
z&RKsWLUgG42CMpX9YA_7_1RhbXJ)Z8VZB77+nf18u5tDtgdr<rv#gPo4onf+Gpwl}
z1?%Yq9d=ME7zzn3c*(i$7>cGH+h(};5KXtS7Y9lclI#?=jm2v-3G?#YPbgmfL0ros
zu=nnGKv1}y>6AsJS;hspPY(9*E!n=!ggPeYIxv+jn@a><d^JI1nzZ2O7&P>c;nOWa
z|2S+km4>VW+StdLJ+8_1R8s@Wy~0KpV%`hYkg&-;{d(dib!C>BG<ysEX$eLNIL4W_
z@uS$$V7LpB+q=a$U*4FGCZA)YDdf@R17K|5Wy5MD`-8{J19yd!e$d>mN^e<Pql)b{
zCKosL6_s>dyg7gjuYG8hncSdR86T0YcS7SsT@^$=lLkr7?M1;NwEEVe=y}+Hod-pI
z%^4mGt;8NSG~0@1qZ@Foo*}qDJZJKaMzm}eqTMemeVcB<-cdJ2M>KIZFtTPl1|G)T
z-eFsFPX@DrqweuW`#7jTwlSW!3^L=g9`$LKl_SHAci?Qh7DmlJI7<DQjQheuwJ#&W
z1dK?m7IaXawK<?z*8M-{UTzw<O=UDm8pWR4x}4CsyIkRjt41@JW!*1Qnu>*k9OQb@
zh_(}{5CYFF`;DUfbemWGl&#{(3iHsi=}(u;*A9~{)BK}_0={|H=RQaAiBu<y&d@vd
z%Jm9+@=!aDjhSEn4+B8_zq+@WjPD&DE3@|5hkh%z1a;v%=_9xh_V>+*cH@_OLZ?c}
z4inx^yjmw%0vl;p6?OJ{F9`OOv;nxgy8ny}9ap;~QdF0GlB_qHm9>~jS?O7$Yttyx
zOFrI`+ZkIk>UXY}<sc`>K`id|07c<Yd}xDOLe?AwaP&wlgzyulNHgTq013_n>;AKx
z{e~kbEKzmT6Gl6i#Fm0vw4S<bm{i6*7BCH9s5-ZdYCu^RmCL$-=)b$mR+~Ju{n00;
zbgl}kUvwN!^cuhqJFkv$lZ*{($g>zwe^b=pv$Md79)N-N(AP{^8c1THh6U=tq#Z;S
z*3qsa=4ii#m=?6Vmf4FbT5K~5)Jd-^*7wD)bJl4BF+h6_f@V=Ecp!h(PBzK!uDef&
zdT8@L&AFyQD{9KC(-rCi4P7cvn)I9Lb<whQH<)VT7i8~<Km2jOv8r~$oXuwECyu{$
z<%}P8U`n4&&S7gcpaK@qcLkt%=ei!&u*G?E>?fi<6^M@Q)ZmVcupr}1=-Ndk00l7D
z_He#&^jAu_Vw%9vtXFliDOs<;Q820u^eZwGc;y=)091{x*EAWk_lBCyr=fvf5~(a5
zB%K^`x1D6?lPSe6aj8R1d?21Dd)sM%sGn&|;3SsLO>Ao}=m=P^piUvQ1|4XIXgguB
zP~{jxwa2r17<Z--^JXo)6Kl4IKE6YTs5b$#vELY%^l<G@x(Guv9`!8K8eS&EpHZlU
zLEIdpa<U+!x3y*cd^@SJ1h#9DLg9xsgG^N%2xl9nZpujKu9U3nLeQ4r7pmydVZyd1
zJX}7Re-w)2X#EhJHBknHZNO<Di=dwJAOlu&0lD%~YXUw3cPBZu5amS!3asKrp7SX3
zsL&9=*8YR{l@s9saX1sFLwR|Es{b$TN@;LNP>VCWDSMxZ2!HnvsMdCa!*51{Ehpf2
zd*3hn??WiBG_SwO9Sx<($Nue3bs7w6l)y253j4~xnXI9@Ocxuu_+1|ADz%18Wv+X?
z)a2o&k8i9A&WN<Z%bh!EZn3eon<f2JAULce!Oq=yC%7<Bsl4p~+o8N)+U)svQYey&
zVSxg`!x|V;Wd8~^n>U_lh&ZCeCWwmIw0z}O4`xQ^R&w)yZ9;wI%x5UY27m_h^OYSt
zd~7f6h$zY?$XmrGP%hIsT!>Yj6H2q;c^qgh5LRwe&#ATB|DiOT*j829mMylA@6^3!
zo*mJO$+p-kgqfgB*urNydgRP=Gu@C90@sw`ISAGv7MI0d=T@-&m7wAEw>pb~iY-Mv
z43Ei)R{wXgfhkon7S(<?T`ujlFYB1aJsBhVj=VKa@w*MQdRn9c;sz);BW9j421WG9
zM<pQZw9cr5#!Q+bbMm9Zrz0w<p`K76pNvQ9oaJu&4gE~(2$e1Zp3^QIGSJIXQ{e;9
z8ypu8Ic{E?7F`6B1Jvfiy#D;jD0#S%W(v?~0bPRoaYvZx|9OI~nmn0BPaW<^;LC4C
zlT2w|?>dF@4si1cIE-wl5{OXjXrT_e2XmRwJbF8T?F?u<S&bcBU|i-d8S=>UI|n0_
zY5&9ic0fqvcg3S=eNauG$}NeaO>5*vbYLJZzSf{dEPJLZvPW&I5D1D`u^o;F7DYp|
zn0Yw~L4ZoWYMXd7$PIp#4HeW6<~I^I`FQ<Yup&F_=SBwGkj#5h`vMbtyj#kqBqx$W
zS0dw!^<83d<<;a3BMo2iSJ9X8Cm4mpkTnWkky)rEse&eS)cTIyGp*J2-xB>ac#SJh
z;nQ&-it+edlsG8)glz8z6{G;z{hzE6+8N6eu4&U@Az)XetV3+YzH@cTs~m|9*C8%n
zkZi+T6vFm7Yw69UfM^rwVk+w4-a0B@tOTg3_y2WczU3KIP*xh_M}#0?=NGqKuL4cw
z`cf9*x@Akn!0h-xXUVBrY<o<S>|r1YP8d===&R!S9ma<_V>o!V(~Kk?$tm{U^iHc0
zbPbqM@Fng8m7<WM%DobhXl-MdP}eF?^s}ks;9mxla&>{V0dQ8hq+WAxx;hi+l~3KZ
z&dX=EoFPjNObKHfRr0?iFW)pBE1XWG9}?AF;%*(eIwvvK&pNtSUS!Y5@f3k>S0id{
zyLYh>l`Y^Tl~-CS_sT^FOZfRR+j{s_NulY%-1eFiMmkSe_Q)`<`R_6dpPeEOaV|(x
zw^uN9*HwjlIj^=U4^|8cYOYG_J^^xDxodn&q`U^vd#gLpR2M!rv#&3uLX-%u0nf@&
zD}sxoj_xjUlkFSqMq8o~bMZu#D_D79qftF<!H*&>d4F%r|Ar67-r24rS%3xW#75|H
zjE*aOx#5Sv91%VEX_3)eUJ*g2FJg%WGbLzmEre7?|6G>*4c?Uwu9Zy^V&-wiC+rc1
z*0^TBgJYnOaZoaA5BmPa&#T9dnE8654j=;Jdr*h!Y_@&1N6DIcWsy%Bh-HcER3WdQ
zxm(A`vG-#A&wEXHtL-|3V>-k(xK1(Zp|+H+9p{Tv=mQ6&s<zZVnU^+J8I!I|D*%q!
zvyf!^r*R}hcPahVb0^crAm5aLkgKOM9mICf+@atCbZSk`x`jHs$D%x}yaDzCQE*dy
zWV`(}q^lYfvdjXO3o3boGms97Ty;c|=zwm2Zad20xLum0_ilJU+L}4Zo-k;bH}UE0
zK=1^l<`9M;@_oVSt2jW;Fwa2eCLK%E%s_$Tqs@$rVdw<iVvz;LD!;MkFb{a0_?5l1
zN7%cyDC1k6TwzDL2w=lHa^>t7Z*KuoD0WozM?8D-3!!jgW+yoF>YcBu><?~^QiRxy
zAkq&>Y!*)SW$JKpX0@yP99H{99V0f5+K`Js>sMN*BQ_>HJaQRqp1_|RZxzsH&8)`}
z{RxU|y!_o%By|k1)9LXsMsf1%qg^mPe`1G!EfPeT=JM2_Q`sG}z>~EjS+9YC;&Rz%
z-^kX+Iq^cgaZ22@v+@n<*S1`6*kr6x7nVtbUl^C0;@(SESm-N}?(W@u+>5%4%vkvR
z(wvh}IQCKFy#k<g()#2e!$#_=mJJ>IMe=j#fD-0<z|UW;wcV0g`UbH#eLS2olc}oM
zWBMkpzz89{14j`g!bx-c^Q0f6kyCq;(K?C0|GwyTUU6UUhpX}qgZAU<LO2q|VTfA0
zRj>rURI`WD-W|*yfHEaI+CcTZQysD~_mMBGz9C{xU?z|(Odt&B0viG#*coR6{)}=^
zpMLf02PgT<Q4V*ghM#7eC`Q>^l>hS~$BM!(3tnAmM?%)5WQW~RUV$t8WG`E%v>X)f
z+{T#&E>+QFWk6UfB$9x{M>W{uZ#pnDWCU&PDNuxCi!G7@3e^9%?RNt%Ah({YYzh)Y
zsM}PC_LDM^--DkA_C0~|-l>W5hT~u}s}&19!nmwQq0tk~{+tq8gVkUooSZqBF@*^x
zg6R*2wWQ_4{1Qaz>MhjY`gQU!hlbYb=2$sA16b8QR`;z^rF*_B?2_x^1~<Tu-f7+X
z1~G~T7#DbR+MYc{7Jj)Px)aat`{R6>ht3~8CdanP@j|ph`oOxysK#*D^6)tagbHk(
z{#berGp}Y~adQ3g5uGVKnY=FF6-p-6wYDhhWcwxI*0dC(s%pmOk_~Sj8@2!cf8lH!
zHPRN_-)vgD@x%{N0x@Lh1csfmp>gKOm0JC!iq<^^%y*ZKk!chB2Y~fhFODoY1LbQP
zrR(J==ECYiH(PoNVWqdUhkv+vIH6H)fh}ZD9$-pJ?*=CGLn{CbQ%#Xgnyw&`3z<;m
z7f!r^VM^Qcd_!|%hF;=e(;r#w2T&d3k9VcoIYQ#ZhLg4*urau-8k+L0{W~{KInh88
ze{s~bTcsjYq+Mc%#t+=ndqGDn<j)-{Sulu7uLd5~3#!XNK@8K#q&|F2-5eLfrKE3P
zCLKM{6&B!~uC$7q+g!yvUF*;y$z?CvG;(%z14mpE@%Aj#Xj18wxN}4qpWr@_dU_ux
zEdWrVF8x4Gl7Wh)j_4Y*X%x7*!DexY5_5jQzc~&DfCGeB`}hlMJy(*QFQdx0bwCPh
zaqX5{<JWsUvOS_Hx?toGK~-6G$@&zk>w6|_9!k6xf9;bSBwcU$i$~CF50I&F*R0vp
zAraxCU!}H{Mq%!rnl<-+(TVG?{Q&6O4Z&+3(hSx<T3TJ5AIFaqg8Bvq;C^>Ny`{XT
z*&McU#I$v2wRYMqq#5axDtB?ICS*Px=Fy3moU-q7J027=NeL95sl1%@$A+qv&lKL<
zgvuoa;cHvx%<C3CMp1*(x8&aQ7fJ~gc8{PXTSAmCvzHA1>YCY3;NK+;!NY2{26gwy
z17mXnR^*pzCH5ih@oSX@R2UYP-&9^ToCERFMYZ7VALA4!AUI$I&?Yc|?c9C0!AD!l
zyUk?)$<Zs`J4f)TFw5kY7%$gz?j`Q6PW;86!M*hQp0-4$t<2iRa#dzJmblm4_qIOI
zM%%#Z$GjX!Hnz*|S}eAr+XIkdI&)5G8_{fR#X+yL-UBveMSlkzgRtaxg&-iW$9Ac-
z-zN$m&k+hnpe{$R7yc`rhcK96IY$48e4_uUf&qx@Tea;g_&MD2Epv=XhIEytdoXm>
zOMw>;f-D58gS7Zr)m1B_yy*tPMn+$J@!s9r#YwHr9sWh=^ukifoCq@EnU6HLwN$AT
z-OXb<D9+>}6*m*pUW#_8n{G0m{1pcB<lw}2U_CBICG#h6Dto4Oeae{_V(4DbHB}Qh
z#(1fYx^@mDi9P+aebBsJ;UnD`Q)FiluK%XNhEbznATS-Zv%>gf{+=B!ua8?*1XnMA
zeZbTfYLSZ@!JXh5oBPaliUxJNKymjNN`Tn8;%%EHuI?!Pw<%ivU}mRQe;lxofLu$+
zd+%emMc4xj&88<+N+n$A@;j+*tEnG{9y4>jqR9D7TbPD;moVrf#7>c;%Ap^;OFQMX
zVl9Up*`8Zj>qn_FQS}$z>Q6^Em~Kg@iQ3xC`4eUb9~gvFZiJNq=v71y-$req!!m-%
z<Ckmo(J39$?<Fr?v;C$9BG#ib2!+yqNB-mh{PK94A;9w?Q(Q?PB9}cT=*y@<{5ztO
zIUn;o@kGYge70n#<5^f}vIp~HV9Rj{LZX#PphBHD3y&LOm2*?&kI0}JV)dw~U-d0a
z<>Md5zU}7JsKEQLiin(%=C|gm`b#!xc+BA#r@GVh5m(zk;Jh*QtU@G+Xaik0-sPl-
z{otrdL2a3bMsI3ok%6ibV_PQIJN-7XEz$XCk%9Bvlu9YyNc)bp`o+!l0{XV+VVi|(
zPpPmezm`DEE(zxuV9=erN(%p9<Wd#r58vsuEEihS{kRQ|3Ywf%4YYB52UeRa1rx@=
z-8mSnjB7Nyef}QYL-Nk|wpEj--nU39E61TKLT6ZpSbAB+yJt^cA47DNo@iWL-+m=<
z{vs=}Y2@P2<k9ApY<(0)jXQ<Vg!1XBHfP9P1OyYF*X70B9JCfwmPg$ryuK<OjA{{6
zI{Pngpk|%|&?~h6-As7J9C4?bRrN2HB^z5P#lRg1+`v(Z^Qht~$EB8n{qu>PX0w2v
zP=|}|-Q$(^`Odze|Le<My6MCEE!?7u^&-qCbm<5*;@y}l#^}JDa4hW-R}apX1BCm3
zFe88_RkIs!*EsDhmU{qc$Y!#1#<A?A_Mq!xL06tsi=5Zf(o0sAq3dp|trAX&K+K5m
zT&#vY$1kapOP#!1ugc%e*h1N+YbE0<uD%+5nA9pu;Sn^?#DGTDva@PVrVCOr#D#at
zB|<ey;$}NRkyGWI@{^*~%<KIN?Q)H;(qpfqT#9mtCL>)$o`T2j1g+yp>k{k`k7GrI
zotN|8Mf;bqWu^^5KxJMX_`&n($~4Li>b{uyfc_NUB<7H<h#nsgkdCxcBa0#>>0}iE
z)lT`jXR#>~<2Iv&)E`&Kul?hDH?pYR!B5SK3S1I_=!g*3rB){g5Xt^*ON7-s-Im;O
zJ71&Hmqoq(#{W5PcElz|6xix+X9B0h+Il3Vo{J;B{9hnA(@^i%KoAIBl#BObPcX6$
z`G2LH-hZCch?FiU3{NSaXBJF`1BlRPP9yo000b}#H5x<=nr{x=#~U7W_a9aoT3K~7
zE{E$|R1w7j9ThO>`rC|~D(A*cya%UULQamF?kDAyxFikw70_WfKakbzmBUL1Qp~KH
zR5Y*v;;o|9nn5kPYs`yvvj@2avm-V!tS$$_gQs&!Qw)+?`lA2UVIqZD;OFcu-`qC5
zc><@>mQ&a|L0o^YLpw$67OK}%yV({&^2+%_G_F+GmX_b8qy%M&JE^)JNxc|Kz8j)q
ziK~f`$fuM->uq@CqS-XE;!o3OC6oCgP^aAn*rr`QrHDwjg*iS#^66dJ*J4iH`rA}f
zsO$#`V0wSG_03>+LOo0z_+!0xvfL0(rI?ga6vQBToLc63`QH7QQ!zpsU#0@UIGkI0
zE5hlv)Xn$Rz{MqmDpn-~MUrEf5LmQad~&cKHe+J%;J#c#-&po?Ghh<gT>Wrch^kyU
zXfJ<ZIO}KIt*&0ozbVl$Ldh3UF=b`=d~!%am40oY5l;2JZHV*HQiQb{I~`UjG|!Io
zo=mMPURl?*REvN;Bp@<|C4RZRPG*DXPHU8}f8fmaL%|*Y-Xkc_V|9NO-EPzycd!$k
zN`8b^7Zx)4qX9!3UHfX#VASQA4(Q3^XF5_WzP+%;CDb&@G+&rNXiieuCn~B7Y=r1O
z3&F{bF3a&ShcxkEyCv4ikvn!(h$AIi<M+v$8#4|XW@5(nI+a(&b?t?4g9xY3T5Csn
zW;PN{m!?`Qu5X7og8s=xo9aJ{vadO6M-eW!8USMEc+tjjNU%59@O<FBLIIRBr9+b@
z8vTCYnsTfPTP`ZENy}?0o!eZMbc;L~OaF4j0`9r*hc-T~B*U6tuZT=n>+tAF&D^P6
z^agS6OV8C6<p35hCBJ2YM5JF<rIakVig6RrC8i^4c(TESj$sG(XWXn5Khdc49vSQo
zJ3*NGR7`P2Z)o}J3tFpFY6qcqOfPk6590Ey6^zbHXHOz3sxOBsx?V|?!o(><wDnQV
z-VlIh*fLGOF1@!Z?m#PGqQxABReeVsy<o(sEVXDOob7QWE3r^qut`OLa>pGr!}xJG
zaTM=s$KMRrsi>ns(9w`tuxNO-JMB{8b%RKL>LV~L{?|SIXNB7USR~>!oee`?>wA4G
zlvu_8v!1hpxQWHBt=U5{-u&-E1VJdPQr%wyc?A;dqQH|n<5wAn!E7LbUAkz=xo*E<
zl_T<#V&{^7;dGTHcbS|%l`_|DjV=Px=?L^5?1qI!pPk;zFKtXAS3~r-YPX)uOITjk
z#q~HzDdb0qX^5Gwm8>RSSE76KB<??O8`y3}h}g((7aui!qN?~nn%jJh6zf<g2(+zp
z-nh|*v(Kce%IV7m%*4F2;;XG|St0IbMo8758?}?6^*t9GHH<>T*2}IBI3!G6JLwEJ
z6%RUGp^iR|0m5I-b#PNGI-XE2*}uLmV@Xuk%F_wvrr!bEXuuJ(;=EpvOjlaqbztpX
zF_UZ{E3(pcJap9t>O0WxnUL%N;CdjUO#h{m0jW|r1bF-X7|VY>j}?unrcm3Lc)WL+
zd4+`Dpnl;-1x&LZa=(lenBU#E$JSWAD>Ogs$Cq9jjevVsn~i{@D<-{&==H6nttfm2
zyw8-I)h|gUg`$Os)<djO%G<urQ%6~xX;(7KH5hJ{!_!<&pgrDlv9r+H1Fj#6qvJ<7
zChlV7?Ofsx_!G|}q9<muOU=X=h9{1^L`#aD>4SYaR=|>mE<c%?C9El&13#r9sy)f?
zd?b#4sj3BFJpb2Rlm5RBcfXS|q+0V-J52WNbNBm8>1xW~gJ{efWb9rin$`T=3`oWe
z2cy&(27;DE-vRW&<Z1+kb9|DciDn}q(1=j1x7HH|nVGKICNQ6cSUL!_Z!1UO>Zmj5
z%D85Mu|uVSFCR(6QTyl3sDlYa=Sd+VflT2HY^ZpS95t#+T*uOw6)y4`p|NugYS(Nt
zhxW1)U7DG8ZFCV+qWm0M_~3-=4;ff=c#zOD9?nczX@KL7ElF@s{vq@UN*OS>qNQQr
zxI_y+Xy4z%a6Kf-j+4KA>;W37`gw%>`8XxjYYAo(4+2-LwXa=E`XX7?a-zG&Qr`9}
z$YwO8v^`{hGAAY6TsTP~g=^il8=_rStplOWA3|a@a~4TlycKq?u>oKR)w=}I>B%Uz
z*P-Bn3vMZY$7rT%lK|Wwk+R~+I8Jjo0sOD}^_B%IMgbQ{GAZ(=U)EBK1Nv=i;~Sp!
zk1(r0zoUwAiKLV{F2<pTMR`nh#*Of?E44+n92q!HFY#UVU7<=7iuoB1-ijeXx?rD+
z6#-DNqoFN}LH(3QTU}z5k>Wb19a;Vy3^jgYbSGyHKVtG>l}rd&$o%D1Di<4HWKL}H
z?f`B6kX9UA!bqKb5QVnkBC*)5tTIBgs0`>DW&Bk_epY>{2zWZq_;72BMONXySIAV+
zhRLdvOy>At`EVHoHfUKqvW^lHyc<D5n>Khy<H?ja4Y}_jL(o<$iPIuBmnEcsgFZGo
zsX>sH^2ZA|8w5);u-SXVFy{ws47F8wHN^XSD52e;8LK_Qza4&Tlq<v7biT1!fOPtf
z)dEWD<}#yhJkar;6Oy9$uw_;25I+xYJiYIPnUQ`eeG~Mkt{R8cct^UTD-CI6jW+Df
zgT1QL#pcCdhb<<or6j-LbmT(yr-?3RK)BP+JrkQmjNo8?=~%0&*%mdY-9EXH5JKLY
za+TfFKnfLo<JmXo$G}{IY{~iU`2?Ccxyel>K#Kke=O^2Y7Rqz6I{1e*WL!wv$V3tx
z**o!mb#zq8{7xyVkPkDJ+;|~t#t_4R-VNb{ZF2a8<lU@6kFlEbAI@@UX3DmvGlML|
zm>x^gmS@xVR#(`d{%m<gYgrHuPPw;#m?0~#o+lt1GxJX#{q-<&KPzhmO!ThrGoAO2
z{5_azpXpKrU0btLOd{pLYsDuL&0Xo()<TnQftboZkCJ`DCPwqY19gy06*$~P4pj1{
zitR*#ai~VtsdH5$(64@i92<=@5T|ZG#ErE?Hx78smbeFYtj*l!3y@x3;2Ev(fMd)Q
zrhiFv*X0?CfO&XS(prMwc<^~Wwx}vqt!+iN(5$yIHr`QWd5Bh1Ijx*g+3J5mRhw?p
z&iumHw1aW&kTZjqvH5{pzuBwT>benc)Ja_|K^^O7eU~co2Np%1x<qhh@96Zbj5N=R
zKr_L#QPDLjDb}<QKD#1*r;bEmiA&3=uEfsAltTF5o1@@nAM^s0u<2NMsZoU+!ICbK
zTxv>M>3Ygq7`qMHX=|a5Dld2;ZhI%&j^rjV1)5M;hPU=ogAhBB9A<zN#A`#4DAb9Z
zZoe@x;1&|+mI>K39=bT)<+gMhRUg0pE{gpDRh}&lrR5F6iZEmxO>+S0w;UBYlW|r(
z!A?s}Qm`~V|MWVA0F{RpE7`l7C4Nn2!pVx;YK&Wa2I~BM`MKnO0o2ywk%G7AfsGnJ
z)3ZaoZ5Pd=(Y~XxT(<t<s@E>GfB}3K-2<(4n;l}1cv0UAWm&o$v2nYDx?4^hCWTw3
z1C>*P&#~cQG#&J2#8^Q9ubi~^Fk)BD5e?Zle+kZeZM`Vg>A3t}@h(`8uom3}uB=(d
z&TAk<+J!G2%gOeY5vo3G*Z*V>WRwB;WsmP-j9!ykwc1;bgGhZGI`T?3O|0`NOD-s4
zHY_E94`QCb9Mh;Aj@WlHsJlcMx~o|BB2T=+%PN<p0}6VS?^|inp)+Ei_0QgE_`KjV
zjabG(RGxXS(Nnf9bSrTMD0xA^cpG~Y4-Knx_iNPs=3tvb;N78)gikafarO~xf&IvL
z^eBK|j%`GTgw%4qX|VM$wMR5N)%hFY0GZ)A;fN{agCgUx?kx{B54UECR}!BsxOda1
z2P3TY%ZjLR*MTdX#<@wD4rj}Q^{*-wj$>5I!$Po`J2cjM1tHh!c@6zZbQj%`>q138
zrRHKnhksSn=L*69k4(#?2tv$TC#Nk1r}n9R=wrIO9zq_AU4Fq^l~7AfSZLDHJem0r
z7Y|&)n^~~{g@XUoSd<vb`@p{d?pU@W82yx{DBL<-0zE((b5E94qR^bp0JfU-RKH-T
z9_o+<?f-H=kKz4OG!M`Q%m|>(*4W}fmfs<UX%5iNB!oOjPEps^Qm_gQ^d#8(!e+bm
z?t^DBHXrAH;q=I>WsovJ`1Duc*$6eY5h#7=BRJXrKms76;Q6KuS0B4YnCP!HQw&<K
z6CrR2W?*9dpxL7dk#$8XJa>^aQzLR&tIMghda=nSpPjzrN%he684`=ero{C!4;hM^
z0BU$kg8QKk;urrHBPzz?fYjMZyF`ASh(eVXjiA=R<d>Mg3-~n^9j)1ppqwQd^Lucm
zhcUnWbds6&CW5NcvndS61P!!MiV;vaBCt4-C%@8_(t;@rMb8hZ&8TOU0o5I@y5=ZC
zt4PfRX1L}`%qn`c=mYe?iL#I(NgIDQU4++#u!~}=kOp3Tnb>V}RvjsVo;L9a8ARNe
zkvXWLn<-JZKORi@{_guk(<CO=gJ^-crTbcfP44`_8aLt&DC#ZqFhsburBc4f$Mx6;
zclWt3_^c9{%2Q;>>sMQ5dD_Rlg)(=jf;?^1*?&j^fsdG)q|V7Rc+gn*vfUjCN^4SK
zPrOoHVI2hVfd|CL@aQj`?-JbpXoR~R_Br`3+~-f0PcQab;Rm~Ktz9V8d(@&^r~JX*
zy<iPA;`Rfq(e+)s@|s<3w94!BSzo#_$kLCv&+ln1S6kTJI)zZaCpi4X%`K5cFnO4P
zYjHmPgC0jR4*GyM^=CM}BLRh_!b-Wn&7<J|1nQZEG`g2F6wk!jOG32SKU*46j9Xme
z;Ovw93Vsyqx-41ir%4<Hrb;s?BX+1^+Bq@{i5d{5mDKz$GR^-?AX}mC!X`XYBI^$Y
zeJbc?0KafbY2s$d`~?=cg+_jaANOw^=V(jT6!<{8uy-vuRPc{sTj44>NMUdl_J9hF
z^wT%#kdXr&cd(bUqXSNrP8SP9AYM?Kl%JFEP{<eXYf}i@);Pqs<7J=ll3GF}lf%2;
zuuuAU?DwxcAWbC5hl#hBs^^5iZNi-9Z&l%tm5+~CpCuKz605Hwe&zitfj;k~LFXjN
z7`$X)TOXTwB7P#K-~rt7Ft)K~swtnK+yQ@P3&FikH^rWQw57gFk)oyXm8{+W%0~4U
z_0SA?GSlD!0yL2LmP|m8&K1nPEfmCXaX~Dgel0ZYhg24}uT%PWYQl~hL6669YlZ*v
z=>vx?Dojp4b(xKoQ}B~%iSiZ0<-e%6!$3!klIJO>uv49GSRc!w;oAsx1$^>HDLLqc
zTG_Fzv=*2Xop<X=@oQE=8{7{dU&v>^tcnDkmAXLV=aHfzz?M|D#s{flo%}2nGti6e
z4BJ8j6YVx5A??}sFtBYacH;jf#`wUm><w+qy*b_RGnPr=Fgf2Jh}UW_``~O~A_O?Y
z!*$VS<~8x$kU_z7MQ_dFMK-ld&i@qj3g0e?6XdcIt{BF>w?`&*KNuOtK?*Jw^Jvh;
z-=BCcfXjC`L`EJ4{a1nCyf(l^WWV_xJ)5C~I=M=e4zVfS0}SrkZQv`s4y`Ao?_~8F
zFP7<s2q81`>8D)C@d}aEw`f#5R(D_%du1N?w<~3-hq<kiFyqoh!g8$hqHBX{6}*VV
z1Y56ct!w(c-zzanSB>FhZrrQ&kphqPk55D`$ePm#NX_zUm))u3@zZIErjMAnvw<j8
zU29dJV>B(t;<(5c?D*weJk2C$Vpey`qmgEI93we(03Q70CigMcCK&b)vt7zYMV^mN
z(B&2@7Rnz$a4dY+fDJ|l+wHCo*?p@tk%+0YbmPV};HWbC9X_!MCh)w+Rm)Y%hREk}
zksZ$Fo2kYlLfZQ1#RtT!+=SbLxxXBre_gK9(%%!}6rLAB_Bs4c^Xh7Gu~JR`G_BnR
ze-~#t_i4n=Aa#FMR9i-rt76A$Il#@#w0RqJOEr-f*gh1!$)?d27il^#Oxyr-@G?Gv
z)&I(fVrs20ntf&T@%<D=f7Dzs%Up2)K|sF0;rw^as$Qy0dr!}c_HOZi2PXzDnRbo<
zWfzYR(J|lva{c*@QaYQr8B`_k6SZ)?WQA(CX&gc;KBgyzL&&NQiKJd;Dc|%pXF5zR
zevrf@;24I`d?hroi_;Zc+`yC!!8yA}9=8gWyl6`Y9fa*1{y@HCFSur*PBpKyE;4&&
z*u)Dfb4ll++WuALLC~C2{Ib2T4<rbi{IxT6RZ^?K7DN@snrTc7$o%QnB5DN3c-QJ;
z<1r=hW>_s{g>FIbVKU9m%tw}j1eGY}&oLVnphKBkA)u8L8efh{&y`quJKmnOLK)Bh
ze}DV(>c&GydB~(9_HKZ~+mPgVo8aY?4Lk7BbN<PceRdHOQ2<tBMeW!T`OK8LeqAh<
zmVU7|mAoj}V~|~xke&}L94VS_S_V9t@#=4qEyo9has~+Lm*Rdxe>9TJ%9U`q8{`sf
z8bn1#S^)XI+Oa-PLdUS`cT>~dFhYxKk6X^_ZU#xZoNWE*fN_)*qp!#E$(8<zpR>b#
zOj*$vo<R<6E<tx)PA#HSOz3n%sS8q-qsf|F^by^Ssoh(F$R*{07HO7!Y}JJ1VWDty
z(YbIPd|e;y33~%20=Oi6`O@2@mPzkB*XF-cy4!!e7JIt)u`_n;1L*Dq*nzrzbnn4y
zW}2`pp&jy(4u}UDCFAty9g!8h%<B9ctt2YmD-YP>Gtz)iH%mWV(AVS?AS=kS+Mm_P
zFllR6^QKrx{%&@!=;}IlZb%D!<xNPYy2swkbF8bWBs^vW2#Q$j&k$XY3B3CI`-9W{
za8*+{8apHu6&5+qKC|C7Kjh%R^tKynLm2&Rl43}Pl<O2LvCS3?$7adi{E;HBkE4+_
zSW{kf!dcw@Ctjm4kWGCDR@t7khfmQ(cTwCZtcSH0nm(Z@A`>>1ipQ|BZKJ5b#(Q>t
zLNvj-a#CkqdtrCh?hAslfX{Y8y$6%~pd(Ou{bgq3x>jFJ_1>zOR7RWe4$p)A-pXFr
zXhJxr+h4Y)FBk{DRNgNL$-nwwh!(44jj6;R2K3f7M?D?cTbaDbI$LV!;u?}Oo`v!c
z=(EfO;Fw<WPx^{Jt5x+@%`dxQ^QQL35&bFG9Y_49fCO8qp&;+-$WeKjU<8|Lyi(y$
zh>xogsRzN3ZV<K3oD@%ERqVOhkFOQCD?w>9%Gf`7ZtJUk9_yW6VQ)#}iiXM3PgADU
zOlL^~=+B-wJBNjZ?d;rDc5}}L7&K%>DHUI^3}EWyi6n%iSfl}lNx#2q*6a5n#FlOS
zQ)FUzrloNb^1WyALT%2-YD_Q?*J%9SH%_afhH93`UFqC&h_#$N5ou?_p5jt4Zpwu^
z$~?ZumWFOTv>%LOnLG8JIe9$~0swu53Bq4`jLsv8<NfH0B<CU@RX9%?p?yfls~h0#
zH=;*KTZ-PK`s9Ql9lHyDh$C(tjt#ZSSu8&bE)kXc+o$mSwn<Ip`^_xc!rS7rG3v09
z{1tN(c*L5EuInM|85CE({CrUe(;e8w^fV4eK%=;3hKitJ`Nxmfuy`?&c)MdKi^YlH
zK)mFqWgwfjR<c@Ai1YIV;rO4`=_1UPY*$>rr;*QDl@f3!p7Ms-0lh=Uv0ElaapK!$
zcuTxCzBzM%NUfaOT(XJzNTotpqV0}i5W>EZi6l4QR!s}B))-7nPnrQIgy(^k@s$Py
z4HepKh4!WzoPg@Hu%_?fc{7^ovVqLde6LD%K@Knmo?%|dZ?A5Wy;T&N;Y4fal5&vg
z*o-jbDCw=q=e)Nx*ijz7Mt%wYG#y>=MDAY#M`Omc=8{@m#9?W1jxjM-Q|(hbei8#r
zyG0RO&`D|w0#+Nq0DKrf<d$BajVT$sv;_Mi262(Fx8N<It-UHk+f{W@n@YP!=12!Y
zB&_n0m+AlBC^tr&r=2pz2a-3V@qgL9^|I6wY5$cj0*@rEu(4ocu(`a0>xSCGwr>0*
z!R#2|zo%r!y(6+E;U_)c+*Yc|U&B?m)dL~J0-HY1(MriT%`qoqNShp^E<_NXMg|%<
z3Kbu2-JA+kGMFN!YKpfrkOp%YgMAJG+Y;2&vH?qViafr?>kDJJQ7+_P`I?B&d;Ez@
zIkT32%NdO_!X@nIW7hO6(1~uH&oZWwc%sjy8cvg_^fXMzWr}d<M1?eAZQ$NbsbeGN
zJ3wLoM}^23s;SZx8-{Q8&r^O~D99Xu6VQJ>e<x~uk{59?XHwE&GBa6NT23nm;FGC<
z&Q<3Xc0NoISX%@*bOf=;I3L`KQ6t!nR6<@$L%m|_*&;ODB`Ev52@0Yh-mic{<(-Xa
zie7}!bFS$wR!TM>5EA(n>@dnw$7yu;Lz6|~CtbbEz3uZa-RPCZ4JLY$9z>q^1>l_>
zzRMYwk|mWzR;_Eh<nx<OXY~b0Js&q*TJSCQ;Vs#HDD<>JmRAJJ3rk0N#%zPF!Gs{a
zj1`k$!0~K*sxGrAR;!5L!b>3S+gNHuLNwHWth|(N!h^dEW7pkjpaJSFi6tL8ZBwHj
z{LTQ$lDSTt^W%xo&q*LRMH_jTeDhi%k4NTn?c^ut>NL^~vmQ3zs6P@VGK1~7k&)Am
zTMpHSsLOiLp%G`qi~!}JvkPwn#T$&n@Y<L<6cayWe&CsVy8*8sS@9e$P+L=C?LHMX
zi*{8n&V6ejgfR=>W>io5wVY_NTcJ$*n{LW^PBbRGpHnWx9=rE2Q!OUQmTp4%Mpiv{
zV5SF&$Z5WpQ9GA9U(M*jKKv5F-O1y!06sMt7S+j1q+{fY?(i8+Hknib6lv~XM0V+X
zXNm{0ez{n*mvmc=pOftG;x6<V30PUvHrUoXhMhqW0erLX1DL$7s1W^Kjg!rfq}CR&
zBX*iQ;*>%#F|K@7=ihV9Lu6z(<2|;Lb6x*F{8?&ehPt1Ofu)w#s^%-4xJJwH7fBJ=
zF?yp2+*Vx;^}L`ewds`|Kk5p30{CU1#_=c5Y9ww)l!Rm8pqAmPCe^w=qDT`)Xe)>k
z$g%onU$OVgEOi{yvC8Itv3e!(%tCM`gC!$xTmrc2#X-R0xwAeq;HmG?Yv6e}Sc2kx
zkx*E3O;<zXSRvWA&!;8S)9D^AHv`1qrQNOS7Rxlk-qZGaKd=qjb#eju+v+SJdZ<k2
zyWNr0Xj<MsFc?uBm*6Ec=$AwZgLs;xrbkk7T7b{*bY5qG+{t}L&`De*g}S-}z;J(J
zy1RiK_{A`TQ&m{BH7GXEqiizeS_%J%!1HFIQm)fE4BZF(fh@ED?5>J#F(O&v6bCel
zxv{56q_76_*d<z{EzGcz(A)Os31BD<c%SU)%9mKS8{o1ieApUOGAuC5TL`WHqv-X<
zskEO~Ktu1{@=~VeqV{QL76nn}!@!RXQ7i#3MBjS5ml&3qY@{?-*@RhItnOd)KtekN
zjX+>s2oa4`m3*^#VY?h^xr4!QMVn6!uZ*oBCm}3iqHG7FsX;Dc4g@>%9v<sA?m`y_
zDw9^T+Yd`G@gj=?-V4`2x$z^u68o#TN2z9ncuo36t9SEByhK=m5>q{zX|w#P2qM!(
zJ@SplaI(;B+plP<G_BgQn{tfLHN9(A4o$Ho)H5>22c&J0^4OuY_`N2%9{NjKqp^bA
z(&ZNLi$3Fsh6L$a`_wdq<U^Z5?=p^7@A_0EBr5HJS;m1JU6CR$&_<=L93WJHtKHd8
z8A4M`X8A;<CU|P#B}T;OiPqE=Z;5i;qpTc%Fc@Sw7@k<li{TVT@gKvBn#XnH1K_BW
zp4lbU5OPQWmn5@szlOZfYgPiBYDh;2p{@J)u&ap031>GqEbuYPXAnt0h1g6a5suHy
zqHPb(<$JP>>3mL6S8$1bh0R<hyCXtB4I@e^ucSycRzG25chQhrVf?c1P94`y<)zMm
z1&JNP#I{yGCJWG0jur=}*OhCr9=`G?a&(J1G$VRy#AABh5n97tFNd-hf!tnQe=&sx
zIACBzQ=Vm8Om;rPei7xLB3bYjhnFUNZ*AuvP0F5OIvO4QsvH{rX~>Vj?wq)!0xH3i
z^-4=Uo}#lZZxuqGebE?EuURRDrkCq7;=qJbY7a!=63<Ghx#?yL24;2L!pKO4Y74Fb
zubSo4fscAoVlJ=M8^c&n=h(pCA`-pc?K;mnQn+H;!<KbM6qDv2Lmt_)_t3kRVFYiV
z*BZk!n{h=6Qu@zG*Fm=Me-Ko2R7(fi+=MCc;7~LzrwT9-N!1rsSOw2~k`KGIj8>@!
zqoJLQP^yb!N=-M3sm)*vqBAXb(E;cP_&6fMlrHX$Drn~L=+C|zW3}(G;?b|2ZZRI>
z4Oq=#2wY%s)Q{<^Le@q&x86xWJ*q7>XIPEsW&W&>K1`pS`E+Y?uWm}~ZDFs`xRtSA
z$u%Pz4<uYz0{%POBTKbANH?0(m&<=tJVA>^d@7Vf<;FmiF*VLLtdJVM$&Ny3gE0^a
z>OK7kX#3e~3cC80h}%<nJr0`cI{_U6Biw0ky@REr`mb-VNm;&)Jr%J*BN|7_q}OzA
z#9ViX3EF9pxcpid_Zw8BJ&P;+s~|0R#TCN+eNK%{7UKCb(|Wo9JK}BNRU~H+J}P$a
z;Jq%)WW^}TY3#8Z9KdQsN~q$!K5p%Vn#kHUnt~xN0M5H7P`>}#4)>e_4U;%m9gA%r
zBBb(tR@2HAt-`0DBk{y24QF#jZzqKCB_o)x@&IoG=VfPb3uLjb$E$_wvD$`_x`q#g
zc;qaIzC#=b$t0}8>ubI@NQWT#6yV+4!CZrISW?vlUB@_9$n>_V2|fcV_-&+gx$t@+
zA1po}EK@Zv7!q6<NhHn(@`s>5QO&_ThE5r{Y)npKfOIkmZ|3^_xVB#a86U<#Modhv
z9J)R>=gjbKa4aomwU%WzkDJ{FtD}fXG(3(S^x(cNgnN_A_nh4WVC-22BF263Jea+u
zS?-{qeZ}d`?)hS|VfDx_FUZsX!p_ak6A1NjJO`Hi=(NWj*dwNtw?U9bCS{x)<HY!$
z{Fsr7@4Ty1<B5tkYi}gVR$^KW&AQJy`fWjbV=T2WTvBR6bP~j$iVx)b*6_6-$$N#t
zqYVC$^DSnT&$Ps$VVX{@fTncCICfd9ZLPOs@$qR?Q_dzQ7MOISgkM}@mB9*~JX1{+
zx(i7i|7+$v(Sou`ZUMDDlN^P54cS+4ZKTnlecZ*k3rs+A+p(y$Cp}IpzqBi@XFHNS
z#H$d@+QH|RLYoPd`=6Oo@M_XxXQi)tLH2XlFK%!TL*Zle)y?;$3jm)-dsW>hShU=f
znHb;94_%Q?sf>y$y*qO^XBWHpBCz%G1x&CXk|LOHU!NlJwFU>3aIEEjAjk|=X?qRj
zq!*$m^x7uPs~dK2NL1=ywRCtK36x$`X$U06hen1-+#uocZDgktGeY^()6A21Kj3BY
z0R$_Ji5`$arz|B4h9FcU6p;YsIit&-lbF=24H@xU_$GsMz-kEeKY$I{u(I3N)IS)b
z^53A&TiW<z5kS>)x6xW}O3eoCq3O9TX+}tg%_uj+JW#I+owP9TmgG=-GYit!zTgL3
z!J3Hg1|{n5Zo@*H-+dte9J}4@Cs?{5>}3IUFeUdFw&mh@&OmM-I}2jIwsk=rVxgQL
z65dsU#YDnVt+Nxd>B_szm?!P!f=Y;biDjz@vkERGD^GaX$u5sgTLcUEk(A%<tdq@%
z-|aV@KAV8^eqJrv1k|yA*u-N%dY?*d1yg}>n813tqesUYY<^oR5y39P(4KFCRtsQ`
zW^H8y;u>31X6(wLqa`cpcH991G`aB=D?`mT`#6_&mxP?Vrg$)eCg~N@+KZWdlJc8i
z2ZzCjuEH5(5QLOV9<A-emK`TOq4V75B<Mmz@XUPt10h=I{e4p5cwnPciblY<-Becj
zLNzYBAlO(FoS!-vj{*OJK^k9-{JVRu>*ZohAbN8gq~^>2Oli=mOAn8%SSrQDtr0&s
zc|vYqlv3^V%5uE@{y05$=L96Ail@^ayfHCa%%vKwX!RC>m=)4GRj6{dW(Ar82+hQI
z9MPCl|L<!_jZqeSlFC{`gYyUqn7rPU9J|``n@fE0yfcb&R0cFp{V!ANOA)8-4G4~2
z3~|}iOdliiV#olE|13{DAxbK4i}(UA-JpxIdZgGx!9+4sn7hUO5D-oa%FYj-Ueu9p
zXP2Q~%zZ5jdeYll)tq8p*IRNee)vCwMjW={@egUU+4?9cWk}jj>{)lKEadBUOUg$Q
zB#ZEF%ab}7Y&wNIoz86GV(hx*!@NgmxQ!QxGosMcQtdWHbHT2@YMBJdYktf-$G;$L
z)G8;0_5Rf?sXD|A9J<Wu(xa|q-cU2fhHXSX@89&L`^INI<1n|4X<B;>^KhTO@$aiF
zj^pqWri(@Aflhd*^mF_V<>2#iG=*}pXz#%OJHlG}+DXhnsFgL>EYN0{Ra~J4lh1p!
zab@{Qu4!2r9z@Gqg)0#pt)N3a^Q?*&t8r2f^_&Zl81J<8EJNjCRjg_8j6Dz!{IJ(>
zwNu)V$HgU^*HRDYXANXNeO7AanAY-2+V#j17o+R>T7<A|;37(dryQybF<WKj)sD&=
zlH77P1?#7b;9O2iZ`*uqAKmm&7v+tldcdo?XbMu6(N2$hPce<q^9D?BAJ{JI%FDL=
zNx$ASfpGnM$UFckz?SslE^YZb@aS@xQ^9-NO))j1^SkL0khG!dwH#NpJf#P_lEGr{
zmg`KdF`1}@064;^nE#%jM7n0di0gDi{!F|tbKlmm4`6E}UIS#K*f`gk8}^Op)rIUU
z?{YrV8W41*%E!v4-g&zWgnG)V7+&bSC_s)%r9(|)K!<$}weXxnWz!nNT8LyNGEsP7
zm+5>^Gvh$nSI&YfqRv?j6}9O6*ag^6L?CTF)%hi;o0LFSv_N^vOPm9X-9U^5aDK(W
zWAP}H!*Kw^`Qga;=vyJ>8e{Bvwj#QN603Z5;1rpDiV9fpDds)DNiz^)A<{tKaUngK
zETH((TnAGRX8NQ{{ASM|jaP{&0c5!NkaFi=xFDwWVvT1VP6;@g2h73wi+t$DyqBiM
z**LM8#VU-tEiS1x^$GflKZ)}&2h8GRva(U_ZLsFL3$1093nm+~ztN}y=E<}H=D3Kf
zw0U@a^Ef(M)qW9~bfEIwGJ@FEup;hKzb7D1V~2`Q4&Rb_2al#v?!?S@uuu8Hi&ZlI
zf03ZMQCknqJwCtuJL^XYN_go)b3WctRmy3Z7wQLHa-H_f*r{y8dr)|uMCTd^2?FSK
zUC$ReiqvFB2{=hN6sz#((4-tZRr~b`%Dep&3=}H&wBP2nogWeQdK?tDRSk<bo%$Oq
z+3bEd6=(1f^CZ5a0>^IDiqGQBreEU`kz#0D)hL+SFhm87NpcAH#C`R$hFXy<teaJT
zTCDJkkbg;VGG9{7l%{E1=NuX6RMHj=L06zgFb(fkW>l?QgZ4TYyC|e{zz#T@E4>Vf
zzw?{h!Uv_qdIm0O#w`y+*#iq7wC3UCMSzSh;^?LlLm)lZ_U6~1`qdtVCm}=Pd#b9<
zOsJH2R+ZCCCl2nKTcB*RzdpFZ#|2l|oWNpBd}}XNri5_q5{3Npy&=Wq&+@BIxbHz|
z+yPLQMQ&m-9)n=s>*rTM%$u%PiRGQzGMD4c<!5v{jGCVFY+~4OFWuVCZK9<5?U9O!
z>u>4y5#jzyEjGb0`gr%#a{>)r1B0$S%z=Xyee0Q$o5Ous$9sU-Xf+~Z%L>&)qc#J+
zk%B)ihsAuVk7d%=D}c74oA0Fsw4;n&d4Gi4$X0fxxg0I$x28n*D$C}Ee1ARYZD8<#
z^SDxnTkyZ0m5j;FcCe~VhDZB?*ltkM3X77hfZK!Wt79l;b$h?2-c7e?U#K61Rrw#F
zX<s|^e8~(3jEwaR-Zm+^O$AiO)g<7i38<>)m&E(sAEi>Bt8`=ySquJRseX+m>*>Hg
z?gOizynKV68-AEcy!8<mZTA1Y3mN49uP*_gnOoN94{8cV-4H*%sau*RjA<A4-eN>t
zDq!;#D+I*iRdWUjalhyRj3<KtQ7k3*=8e2Ln}<+Vx{|I%3)CA7N(NVGm6NH_x)BS8
zodT)2hD}s-##7RnhnlhQobCl*^|*Dk<1TH|crIIySs2hC1G4`?l2$D_COThY9mjqd
zQKR?8_KFw&c!QX`4BgixQig2kh7Q2c(^S#O5_Icd2aGx%tbyg1Xcc<U`-tQvC#*6U
z59Z7o!%lGR^0b0XgJg#2<m>H<E7AT!n6!xo3#;~qof^vQIB<2njtv7YCc*@M&a)xO
zuwwO<(R5z+LKMk!dIDV)y8&boxvOoG#J8N(Seq#Tg43c+3IvwGX-ixr82LIB&x$ZO
zjo7G1rqI`271-uVd#pY_c6Co4Ec`aH(cmWWaGr>P;f|;=G6b5m1tNO`U4=BUm=e#9
z>pt~7yIVpU!Q>AW{kQD<TQ~jn?i}W_K>^LC-=v=e=WzM1S(N|`=F1X4jjQrWsKU8n
zvDY&K2eNa^=6z8uF$)eCA-pO6-X56RlyiFXQ2f34xVklb*&@>@7!SjqlCRfUCXRi#
zcgdCUfF0}(czrlTz+Sd2aPu{lmtmSC+n7v2yOJ|M@iHgbm)s6WK_vCouHU)Ygwbvx
z8k|FKZTUP80ZuQMlsv{G>+7%eQT7-?JHX?YjA!lx*}^j6tV=12%6^yr-H;n_Gi)R`
zLRTFUh16#3r~%j!Xbjgws7-!-C?DH2K0P8Nlo11Ny5^{+Rcpc%qPiGBKn|85K@k=c
zv~lP8{DIw_>^H8dw&f~_=nTn3P9|SwX`nxfw~$2#6VpxdS7%1Y#R5ueg(bs)C1?*k
z#)6XhDd)#Cn7d1$D_4Bd`WQqw;sM{XUAgCP&>JpO>kNb&&!!Hn1)Wu|8lq|kxUe@b
z%Jv$3h&^ks$D3K+6QaV{LPL1CHR&}<Er(j_`qR)7bpEDt5hl7xq1tebB?6?)j?@0Q
zydJ8vwY^^Mb+Q^j_PkVe?ZOL7MmL6y5817UGtO)rk<ra~@1;aXL($E9t3rQ|Bndio
zk-qF7I9iLHz%7%IcT?m8zrR{P&XZN=7cKE<_HaOS##0yX6pL4t+g!VxJNRv{>8Ls~
zSR>m6a?rrg1bO`&3(zduD=Asp%DY=SBlgYKqid1g70*8Aaey%RnVQ6zOtb#;csLm!
z#d3wUdEMRTZB8ecJd5`75ubA4t;|th^`6lv4s8;(7jNtuVjh{Wue2miZ=&xr{woeK
zTl09aD1s3L;0Q^BPl~~Ygg^LT#IrN;8GD58=jmqyD^|0-`c9{IOp|<11K#_S<lUT>
z-Vm@Tcp{CNqz#AaWT~&wPhD=*jL%K`xj-b#915mHBbGbq>4+|Hd}4MpJPV`t?)#)9
zqTvttjQ4IutTWK2LE>d&Deb<@#nKbIx<puSS0bj`=p<YogrH<0&1+?w;&Y(|U2X|3
zS81-3b(>`~yqu2lK3kFvBobY_QUAwE8Z--9*-v<0&)XzjL@{FgpFw*`h!qm;;jX<U
z7iNTz6oT8cdd~9#LjTUI`UcTSEHJOYb?hzZRRPH<Hr7gJ&s>T=S_%_RvG~te9k@D@
z_^scj4oZPd+AXd@T3WLPlJg}A`mv9PB%l-47m*+2CfZV9=*(QM=oo+Isr7gg!CH$X
zdr`>YK`kPx1PX~tauBqegnkR->N$x<@Y%eK5w}rfDwATJIic27+*#G<=K2`VEDVXe
z5^5e@7-@}RV;Ay9^bn}=#0VfsjtoqKHIYGMy>=K2f+tRZ&{>E`{cZGVH;lac4Ll|a
zB&rrcGO>-o1nQXt!S_H%7zXjsYyQEL7f8O7Rly?io>0p<&jNp;FA!{Uveqg%E%t>2
zsL?|atx>D)l7popdZbs->%!a{6?BFEBC>Cv3r+aQ=j908!z=;|Z2VUd5C%+88v5w8
z9Jm^7>SiSEy?}88dPvm!S018}-lCZmKw0d?Ay6|0m!}GrXX_%S;K_5OHGIiL=eUDR
zImhe(WXKT;Z+{~@H;=22&Tv+FZ*KIk_b*@uVGoNShxd>S6ErbdUk4eC=Lwih1L3T|
z2}beu6eK*v^xzk#jl(>9coDdZ%xOQ<fMZDce)l3)VOzh)bRYAi<`}0bIQP_JnJ=+r
z7v};fE`>bKhXhO1<}7leiY~;ApAnPChg<_Cm$}M9kB&>-p(;emxAx_)c);l+2!xzm
z<v0DE;-BIinau)=9Kf5uRRiQm>N^;Uk^4h~i@I5-r_5m@I&rh{QM_6(vgH|!oWvv-
zeq0&X31=3Pt;zgfg3w~s^?zgeulXWxiL>ELMNT)K{lVVs5~^97$7=5DZVmXZ1L%?n
z2c9iaAk=efpj-MG#td92fgVim5+y%fTtg>yo!Mm4R`^fQ@$lvnCymk6!s4|Ba*3Yt
z2}vX5;m4+4GXeu4chz+XVAs4P`wh(BFRkb~p^Z~(X8B!rUhr`Qhf_`Y^KC|nr=p@j
zPoG(OpU^@_Nmwts+Zv$C>lfe<K7G$rkw@#TBSXtbUp~FBw<066OmTg=VHh<RR0G6s
zJI4btq;V6!R%_!a&cM%9_)c-&mgw&iFgu`6nM)7hph<{8Sy$w?F?fr-LGFs(UHDqc
z*^JzNe}C1>3u>NAcd_{Q_-#<>_A<?*AsD7+`BDLCY5_AYV5;;(W6S*FJRsC2=FM0{
zQ4MptB(Tt<K_D+B)eRdl-ZbYw`NiB?o_?k!D&!^lm>4X@Fhy@ApU@z?h;d9w?Y}|r
zh8voaq2v!IDVO|7nd9k6o@FYtpdoNbvite_PCb7xPWZh&H0%F2O~<Z#l&*0bf{>T9
z7hi^U1Ls`obTl4Qtm}|KbQ}s72xX`QuWw5WQ$pAHS~c48#ojgf?-AEw{c<_TC|!FM
zoFOtTN7oI$bi>Z}4+@J@Gt0@f#~p+`MtaJ>EJek#&l(YP7u=xr5S8NY;l*k^@7h*C
z$$8h9My?yJzji<;^`j7+v<Q8d(3cte97Is@v?#0PgipcfT*k64MINLFB={?*iW&DF
z4q9180M9_Ocv(rMYHGRTpoA!{xD#jveOpn>$OXD2OGfr`_Bt3swqc!kq$s_WJC0OG
zr-pr0&rEATm&x^MFMd>VHp{xCZ%Ojq+H2Zz1d=rUv9lbvsChguCf7y)5cFn2lpv2T
zbD>xIit&MoqTTzH6Q_DU)TFT@OZJ5M02nexL!EoR_SSstHHFhI&v9oc+AdS1-^vMx
zQq8vjX0;u?g|55-jza4F+d$0Qa4&4?;v44cUjMH?dOQBEzUOgSg<E1ISHE8YvlMC%
z_wzT9zRGeywV(qZSJ~LCAAqAJY2QK#HsY#ZD*licJEqf3%S8|@pb99<CZrkeHd<b_
z;(ceoIJdYTBRXgRl|q+t=a#J`(EF$w&DB{{gmiX7knAvi0!A=}!~G492<)g8$$i{H
z={ely|4F><rO3E-4X^Zbkh_-Yn7{?|9J^`?qJZt*tb0s@D@#)g#%|4~xLIY*8}P@w
zLTu;o=8omD%oq>Gj0t3%gw9%Y&*b#H5Z}o4sad1I$$iN_1x0Dn;j~(sH86qRng-(=
z=G*B!;W6#EF^b&HiD<seeyv^ru<*t}X&e3G(cU_Hs;V=QKNr~kuSs*~=FJOajxI}e
z38XIqx{IEsAEq*bi#00W8=s>Th}E<4-q7l!O`v!0%0~@QjPgdRnK~b#4Xlg-d3}LL
zAY6Q<puhJ%yRTaYCa`X`YO4Lu^*gzbGa$6Ozj<T?;c`o>J$=!(!dOISSa;y<mG&A6
zXL^E)qOf0Rg)SxHqs0Y%@u5ryKK}M|Cex7_Hyi2S$)6R8tZY~i@#(AvocsF}8-?Qq
zdE*^3y98`WG5f7FW2QJq68vjR!krI|ee22JicjmJ%u{F!_g*Yzv~Xe1D^|K~ISL%x
z_;6+PHoLwkwF;qX*IsR;3wa1`IuX7)`p<!g3!w38k&yNDr+l#hvDbs=1dZ`JA)Ob|
zwcCt-*ef6Z72KkcIocTlK)5&y#9LwLSx@-1gWDCuu#jp@eQ)!TVA)e*O2G*Y4=5o3
zrEy7J>TE<3C~G4<hJL|E!W!$v;Q<)gU9-KOTvijXXT+M!v?d_(9JQ-oYU?0jgdzB7
z#R)6!sfTXx%;v%fRdx!@<(=`7Al06*0ey^;*|Qt!i&CZswV)xYXIzT4LGN*2QDy<L
z2r4o$xWImVm!`IXESirwHI<(!b{F4h2n-7-Hc#d?$OPtrQ|IE(T{Qn-E524-Mnr3W
zAaTauKy^0A<(HPe9vLmOdA|`WZTU`l7tdVAsRf8^%p~Q19GRykW%kegs%Jx`;>px&
zQJwm^q^8GdHKl_BE4&SOY2b~a-I3X+cFj3dF4GBq$$aX{ov><QL!h%dP0#5QK=r2i
z2(sJ~KDIW>_Pq%adQmuy8z*3{jd1_+l1$J7t_hXK@eF(ys(CHVCjeGfdeMBlh-RGU
zxE^oAYIAYI80u$4lcA4l<onJq@Qc5)T%z4&sz2h4%icRLHpq?R4B1;;AE*$R9#*T?
zX?qR4F1@IDJbvoU<3DJEIn)NU77Ee^TmE93>}4YucEMAe4I+7lm3lf1-jXDinPx!R
zz*+`2z|x)i(cX!O!b*t<Irue|5Lroz>4_)dO(B-{!ihvYO~dFSy5?3x5s^@N9AcxV
zLR|tWskEopu2%{^kDT6Cbi*P$V`{iZbo6%${!FRt?M4@&@q(}EHht4AI0_!&KVBD!
ze>S14`wvW1#wAYX5z1}ukxZiwIyw%&mBBU7n0MCXbsM$-Sv3ZNUKj{?Q!l5jP|`<^
ztDR3ZDRuIzrRfb%gA|n!E3%J|MZ?%`3&k)BK1N1`a=J}zr3%s)HxVePgxkK7ck`R)
z(+)o=QtK$`cn!0RhIqB`GGoHh$2StG0>~TWFa`uoP&xYxTq=$kS(ZOqTZ`hh?=A5X
zr@YHuTlW#U#KWY_lbtb}k}lTQ`J)FJ#EkGrq!;rWIW4tPQ?TD3f1&RVdJEFX0TRn0
zAz;WZPJjfHeWS#@rnyAUMabuXWa`-tNUdB|oP;S`;x9Q)j}Qul;X{gIV0yBaz3+Oa
z8_pxHF;{<>wHiRPj%!sGaTzp}D8TER_I4erpal}<Pw7M@f>>#ZG!zkYmja3rYRFH#
zf<*sFyzV1EW%Ebw9WmxUQ}6DP2Ib~3bxplp;V$P`y?OOdI0x0mgK1FJBE2uco`su@
zFRlW#@OBM<!D*kV=3rZ{3fVI49@|*f-dG#)Ch+3lt>DX6*8lC6!`v!pH8?NHe_3SC
z#6iN1ZZQ~`l`%=HQr*v(3vRuD*lbWcqjoI5WjxSJp+}L)=mz6qmx;j7S&?xd(%YLc
zrh@oCD&={|+6+(Dp@X$L7v+J-QQTG@zR(>DFm{}%@CnwW0Qt9y>b$gxmes2$4`{|`
z+QHOaEP)9~C1>Hpk>hgMj}P)hdqZ8_)x2C58d|;~i(Te8h87`b(<<coO1MTv&cGR%
zod8R9%nhVC6%SuS0}PLG<<^B@L<j40M3t{^XlMmL3M>Q~Ute_`Sqp|d!nVq0Gii~f
z9e5ba5)1vZqqpyq;HQUvLr4HS7zmD=$(%Y4pa<x)F~6|7^lN615%utk_0EzupE26@
zFl6)?Sy#bi`OV|C()VHbybSsQRRUO(w(>gxDzBM2Sy|i85f&@F75V77ogeH(-MByL
z3p?sfn@2#Nafb7pc(Ga(M+3v-a{NCpQ15>u@eav;L=&O=IITUFuVd=gy2Im9PM3)y
z9>ScF-?7B}-_cLYujQ)Xdyg2?5c))vrRlDDb*pTN^{s`Xn`8uN>9Bw-H{0>PN^V9O
zz)&0By^a3WnNS~G*&Y)SM?Vg9<>lmz(J+Gea)DOF2&{i~Z`SK%>V1BqtCawEX0f%%
zY>~jz5^hCgP!x560aE1G2b>97fa{^Sa_NgcMJz48ja?AU=q<(!dX$ibm5}JHHWV&q
zWh<KC$cngi^EwQ^qWZw^t-@z-V?8qDM!nH<LTg}Sx>FcIfi>`+{Sh@bgE3~99B<yX
zJExS-zNchOZhH~Tb}J*E13FrlALT9n7hWczjwZXwWzycrxxrL$;4@k?yUlf(TUeg;
zz<)XPaU1?%8jzdWa9G)ZqRXwCXTD`9A3`I+fV`Bn9M8M80#c;A6nu|6gYfdKDL|#o
zZq~i43Ucx>s!@>IVhC5#TX@gT+@L+Mp2L5P#Xgx$Wn|1y>AxmP#ETNk+iq9vOY)+6
z<0R0P8`-+s3OyZdb^b+_9^qrWMsb~fgu9v9c?mUrQK+v#VYtpEhT_1_=6<$Q7*lHy
z91|fHCU}>hjPHQ*ZU_wr?L`I&>=<-_p71hrOtFrptB{sDY9f#mj38+=q{5G`C%pa#
z*r!+zFS&CJgFB01#*B0*Es;fNnDypf<0?IEe0)FUBRL>wD!MTNc}Dksci`-q^Jh41
z*^TIRfM<;AeMnzqmkSbNW-&tYeKbwdK$bZ|9ha22>e2VoEpw=#!=?5L^%?-dEaJ>O
zc5GMr8_-Bn@|M7W#(7d|LXJJk)RBqJPUx#u_s-%BLo_ga5XFi|Pw(P!nqmXexQ9=}
z9Pns0Y5+@*@{4EJn7^Js76tg%E<}c)g?7}tOCeD5Gkz|<%-7_B;g_~tT<8+oG{Yww
zhg>l!O3{)0y=54^*;mM1J=lswHGmyT_zh}f0%*J@%;o5uoe^*+`DDAe(CJn~GP(N9
zC#)8#VIj4;@48uwD_nXPFM;5E)r|<Cb`yb^f<g9Xjk4D=mtcecK>%ao^<&vZ!E*lp
z&!IZKT&CTX9F143GQn-&U|ph**Ukb+^(l8#pi-q;<#N7w^@MeeEwN%%i%ly&_iP%E
zKdQ>oAIr}PAy>xb7=GbD)SWmnh!0GF{Xm7jkf!i65Q7j2akYV`zBtbLc<XTP&K}}+
zzSBZnQ4=Pu^G%JfOI5acqWPQhm-cZ0CK?o^9d@exQ+0n}RMoc&r9uI{5sVlnW<wZ(
zYo~EA_TxZw(63`8X1_fcwes)O5pqeU$nvL6>>HZf{OzEqa61yK|17qa-Ut4bbF3#2
zhv$y$Oj1QDq4=VPKd#YdvwKL9auw5pAW_$P)Hi(Hd&dfJv3`EOsfq&fVcZx(8^~wd
z(3mmT^G_OO>i2_iTr5&kH?-K>LzpUB8o?eg(BDB7pkG0Pb!Nw<5#;r!aF&g!KEpzk
zG<3sS4O*=+lRZu<b5^u7itJU6ns`vDvaz(>eAa4%hLp6`Xi=7(l}S*LzA3~}rEu3P
zL_jzjR-(bXeHQRT0q~L#H&Nvz$ae~xl;FGU6~#C<DyBBr&u1$oYqBa6o&M1?zigVW
zRL9K=)C6+4;QasPJ?G)-XQDv8SR%5sUadc4xejI%Da&;e0i6ry>kCW4)!daglLfqG
za897AB_(WG{o>UUWMIrY<;JA5u48+~^<)g}{`jObqGpF`1|u{8J-L&iq06wd|GDN)
zVkKb=21Cy6c=<CyL`lQUa>yg(cCN2KffTPjJfqu*CT4-n+bnCgy|&$xgvQ)E=EMtM
zS;JxPQ|>2Jo~w*OBT7f-LX;AQAS(F_**++e5uZ+PsPB0;2;v3zXpaG*x>!n*9yFba
zd({k+gvm1l5!?)m!OLvw-K5CXMzm`&b_p8D>2Y7X;!77s->o2An<6fc)xht@eMg2Q
z&G9ZtSUoT|;XH~_(x?7f<j-M6zibR0nv5_%SQ{;ozBj(zooBuRJvVSA#7Sjtd4**4
z^u0|@T+=eu{=DoSHN9!a0!BzFRiP>ir?Pl_PFE>H@>S5leOrRrW=LOG4ao{yvC112
zU8YlD2sB&KJB&@}OO#o^`(08>A#wOLv9l95E*%%cSm3s;eaAX*=^Zf2iYtVzYLwU(
z&Skf6xjaR+=V5<m2HiwA&hn2FDn7d!=qhj0WW3pmPkOEhlip4n=}7DN?IxHjG5R+Q
zVVcH!@lqS7vc@;4vudH}ltz#*UWu;X0avc0%0xs;L|sM8+j*nI3RShCYQke)R$MiP
zksl5m;B`z(=XFgg6%$~0qc3e_uKvfxX_&{EOUS|~bVqvi=-Gr3yZQbrTwTU<Ce>%8
zSQmhm0e{@9gy{@>LPH{-iH$iI!63c-h2#dQU5h_hJJZ@z!J^odagI_N+6w788C5BL
zp&^Ulos<87Nw_PUgFlXx%dHVNGQ4GqRumnkspWjyMXUl@fRR>w&~@d~Hwv%bm72%Y
zDG(xB2`25?mnH?l?UzeB=0EX0uWSg4>x7Wv&z5=UGE91ERiRHvTr!1%kyYx1k=|}X
zAfw#hGId?UrVJ=VKjPgYAIXSvXbHMS40mE@96nW@jQCC1SglbVOf<dBrd35#Q`VM3
z;c)figyYI_VCkpKy~{is?+tZ>;HE${=T81yeD`hB=qvpM^_6VXj2o5u6LoTedcb}!
z%xvkt6Pt%>Zh&~4w$SK5uX)pZL{(3O3Cg5F7bU5&j`UF&aHQg-zmsm{`leOZJX?mF
zc)v37-=VSnjK@7TZvmihWB90mI`L9!VV|Yw3>nAq^Eig${F(PqkXa*3qg5*oe%+a6
z=*|6`h>SU`9&yf3pUGvHmtd6uwx!Of6F%&Kp4LAVO1f%1uwJ$Y3e9Yn6@Lz8Old<R
zCnQ^Pt}moGD;7O(gUQ;hnwBoy(rsUWyXVWg8V|9##j6O<dGVDD&YEqE;D=Hx?O;AE
z>|WsKjS*;o|MeZTi<N8JTps%^c14JXiU-H^21EAU#W8qe0qS5QUYf}naG?2h+Z9L4
zTn`ZxQuGwms8+%*bPJ$+nav+lLcvJ4^AE1enf`==(VQRp-&s4&Pcg<Iyw(*l6pUwo
zuh=znM6(rI30S&2d-x~u{Z8?)n`5#6baYwv?gP7sCHtX6{+YqgR9`$Z{Wo}pQi*NG
z+?RT<I_SESL>Zg8QEY*(b<D-8zP%@_9wH5HinjBibZHo-)aF^mM^<{Oo+D9@0nxa%
zb~u?3@dhzG`?7r|hcBsf3~I?g$78(m#~poV3CDv6X))x6@1#YA#+R%(6C&RZNzEaQ
zxY_MfFNj`0*oF<8XF3KwK}O0_e1E^jQj<HhB&>btp039jbMxCiix?4&U#=Nik0OMm
zc#a_cRL1DzyArO<o5KE&lk)|^>AGe-G3e0st0jea&jZLu75Gkaap+tK3SOxwNE5O-
zBSDKf_v0hr7kSM?e26G$s+@M5zAz%H6pE@Vbejr;zvvfkUVkhpw+!S$x_scPckp__
z+c<n)e114yHgs;^QOHxwiumYT(9G$*G0$+$!4bSVRF(gwTi6p6c<FFzl|FU34EHw4
zpmQ(eOyJ>f&9~15Z{Z8SgdUn}U<xG3p_$QH!xOt7kVp}|fY$ZeEAltCgk;<UmO8-d
zYNl=nxcRtoH@E3>rU)O;)QnymaR(Fx>OCnk#1WmyUISC5*O|)~(E!79Gvl9y`gQlL
zV3kYTQSjDVBi=h_Mo~v6QKHs{f%i3wRgwYhKASJ;)Z9Kwv-0Rbhej}HE*LSQ0D1fJ
z(?dPnroDyVC&)EH`_Ec7!airNw*3VClfie1I;N2QyqG|mw5nK8UpUJI5>y3-W|P8$
z6^gG4nz<klI85Pl+KI4OUg{JL2SY^)>8-m<Aj|owVL#|5jbLd3p0le!%{gw{93rUP
zItB)j+V$Li5%bW>Rw|}KDAGlo{lkwjF5l9$>&i5yFc`@VAZv`Pl0Md3I&NEOa5mpq
zeQt`R<`>64|4?RT6Gy`Ewbe}Uw5DU;y>5xt$O}L{*1#hApk#YuKHl(E`lO9Q4BTZ0
zt2IFHND98?5I#792uwS<ST4$Ko#)=acpGe5H>JhV!%2P7<&!yvjLMR@y-5Flnxxgs
z*1T4XOcJBryF@-T10fr2xV&99BFV}ybbytZPS(Ta0%+_hbN38;Dt1Shc9zwzR+(`I
z^0KFS&;<tBh!{neN3_?5#<Gbb2K317#DVis9YL~K%vcAGb4*mO9N#8iw{~qjwQ8Wj
zAO#ui6JPxWBQLy`=(JG^aOy0VD3MwPAaA@wZ+S_{NKq;EkZEQIxc}s7)8<o#>SX*h
zKp55o$S78a8nYICW+eTRo;vOqPdE|SOv0-j*O~@f`VE-3?DRAoBMJ9HD!C<)E`y5_
z@7SIh{#})}`IFu@>Mus2AQufPOH^49dZ<*3U;M?ro6C@KbGnTYeGTpUK>Xlh4nOd*
z?^=PB%7=ZbdAl>?D2f`oPY9lOSQR2EbH&OzCm?oamged8&gqH;0bMQ}Q94=&**l3H
zBPB}JiIHD}v!x`ELr!BE?Iq)WF4;Y}lg5!zM)hEWNNlU%5^EUVDJJQ7$|kJTm0Q-Q
zPe}M%s?KMA>{4J+zcAb*Fn90r7rA<Dn7)k*8Y=~z)nAV7nza%DO0Kx5WHeu=ZW61v
z?r`G*Ba2?U<F@S6KA7(^-lU~6y_b~kvOI52Nu1!<FXjE1hm;4dA;B~#0Mlnw`Yy{-
zzYY(WuUL*r5An}eu?BlvSZPBL!yTdXh-Q-pS@L{~eGAi5Qs!}VM&T&JI}LtMK;R9q
zac!4IPkK_$>X?~#(wQ`gDyhJcs*a{^B%y?@@1F}xa6NiIzC8!d_?Z)%vjIAHkR)Rp
z{Xhrw-!l&8if_JAe1v783oz|S(x^rid*k+Q>c3J#kcqFspw-38PprbM-f*+?lw?yf
zS^O#0#F8?fD4i!hKJUR@u%ApqWFoG!?B>@{c2)gxH@6<Y&IbN-Tsm%PgxY1^tb7t&
zGQW)C&;e-ojSbS$N|swl81*EEUtE-#%F^MLO2o-$3RK9?I4j&2DHmFxk>hAD{#)@O
z%F>1Wdk?P0u-Ibf0x&HSMRgq%vNZ;34?|q<m?PYd`TBGSITi09tXFunHgK-$H1Xf2
zQxb8nxISZW55`3uq?l=*@p%M{gvs+tNvjyhi}oJ|m)TAPzufo%fcTX8h7AE$ZotdZ
zwIWR=1QIa0r1Oz&Xs==c2?J#aPMfnlGE$!(a2&JG^8tQjr{usZ*$JalwFCBcL9fX7
zXR%ZUJ&zazg1D3O>E;I}Q{1&Z{YHxX>_UZaeptp-&92VlIO<E2@Pk5(TTZDyG<+Fl
zkf4IAr^8F}UbRuNZ+l<Q#eMBpKFrGD^$M=Rn#WLNVOEXxeL+@sPh+BTZZVcPJ44ya
zh0#zRNkG9P-EGDTp=7QEf25Hkm_k&sftj0f@gUNO+?kiD)LN3bs@5jtRUks4eOQol
z!5#vTfBrC<pm_B}#?Zzf9o~uhb#0B@tX|Y;sX_-l+rLfk<k$ciBaKZ7G_iFkQX!0b
zAG2`nb`d06r_%&EwUZlS(Y?RI_2bktL$@T*g@JX#aUGYvN$36eIYCxzv;th28-SsE
z%=GBhJ_teeI=&+%w}9783@&JHog&X%Ug}g}qGx*%G2-xUtUIaho`+gZ6_WUsM&gXw
zcBQp^I;cVz96(mUzjLC`^Nr?ZJV)TULD#+wT7$z?dv#9l0^_uuX2PtEoEVN)MAV2^
zh3SPR+5XuDwjIiU$1|If32gq5*fHM6)t`f>2vo;QBBP$C@*{CuFuCMLnIrbONBwy>
z@3OotJ(D{%0TeIKJNmkyPXuOOR@#a=ec4ge1&}p@rWg~{_2Lh_+fQ7n8g|%N60l9W
z+0m!da6KHmYe)v~^q3C!yva3KblT$FGcZ}?C#=ldC|8N|tnQ*MbsubNlc~~27#wwn
z#<u>e03U7LtDw4dl2i?}Z!N=HD;^HOAG|}!hcFm;_QCz=m)He^M$L70Y@O1*csg(H
zXXX@K4D%&a)>&70WhgAw(b7-cwDq0B_DgG1yS02~u8Hx^H~`=hd2VV<%t?ZHR|z1$
z6e*YA>mJYQ7=^s#xy7V6OA88$^bT^ed?#zNIRT}9A&;z14eh0|(V3+&*?9#ZeN9zo
zs<3dwjPb=C3+w>1u1&Z-O(^Wk@nQ<t6omGKxiArNr%%z5vy=W>?6Clv4q(R^-WWy+
zs58i%d>jLMj^_iA{&C2Y$QGa1j5qvqE{&5b<d88^&)v30^F+COQlq|}M|f>Q$8U{P
ztpnVwBfIR=JCwG$@xb1VRo*6#9EFp`xV+_LH?D4U;h4l#?dW%h3^^1^E)CL3FBqU*
z*Ht~***J?9O8{Y<I}Fejbkk=njYi_XQQUm=Dms7p!C@L@tUep2EM3PY#m#2wcIHe^
z3$v3pU>d;0_ER}ID=7;7r0@?20pAIsnuDw3kKYZfo#y*Ua>-)qPF$F@OkZl`Ch@0K
zEHV0|VT`2nnMeA8;_AO%{;_tq(=+XFapEm}OyWgZK)1uHIvc23uO45m4`<D(uc_B`
z(G~bj;;D{og{x1Vf|YiE?Ag87f$<3V^DQN`9U9Das`@qj0p?CNPuJx^{{%=BC?7mz
zFi%)XN&}wyLxS~B1nd81Y*H6Dn=~0r?n3L+h+A}(4kf~GNfeaLl0L4PPy)>tCRL&k
z{4avikbL4i8*1WVRF<I7{!ITlxMpho+7T4JUYxDdKd(%8JHBXTSW+-HLb`Qd^X0Fk
zvy=7nMi6&&-@;S56@_J9LJrzV0H9P1)JPBV`7Y&3zFXXKjBGYS75Z-LPFitD2C`(|
zDJmMR=;7gLoC#7Xck_o3SQGc|h;nq{^B9bP9V$0(=)^R#wgl23<_95umn|CtnUaxX
z*uF?WQT^zebFZBQHrUH%-=n$WT#3foL}qpmt4b6+g856(_?^X-QKI#FT4Vyk*G*mj
zYm5~>i+RUp%)g2Mv(aH$E*9z@oivf!)-MI}di~Rs<IN}eWuzQ6<?5ULxQ`>dnGvYD
z`be8GB9HjMR~0Dd*<wPL97Jr5B>bx5C$n%dUQ|fMiJ!WX2D~8z0w6giY*^xhQpwW<
zw@Fn0q2Zb|fnjiOzTPia(~m}3Az`E?iCFO)<By#aIMPw8WXK7Z7Wx~P;*?${%x%pE
zG!8jIuVpI#YYbOdU>{jw9en$^1oZL{Q^wBHeRWx`wXta|wBPn{H>hz2iX3sa&u=#>
z+d1!28g&;a$FR->HLpn4sdHsIsN!D|tWvqm)tbWelU(%g(Y${v^T1-CHWzQV*0o_D
zf>)rU#ne0IvODiK4wqb8=Ph4=@de@8Z3vZMq?uaW5!{4h8GCnRiV5lS6|hNwA3)lM
zlb~#puDaAL?UBL`&<#x^+5{0UD8;7U-fU=Li}P9f1`Zo8KUC>o_b@!g#4SZxRrSca
znRHW=>RhZ_t{6$>S2_bxYHl-|f{waU7bhV<@srh24z!m?bRt>@X<MvElq5MF2xdA*
zfp6;#vhA&C#n<u=@tTc6W4&Sd&T>Pf`prNKmMMD_+&}taWw?W+cyoXx<6sdj$F;)b
z2HB<V1d+|b&?0a}6h+I;+q#0g(|&67jXHU;8>lB~gWCP><yb+9_AcF=Bv9})@>}#&
zMK2*Yeh_zqCZJBDHefjW6@~-fui2P~K&7SOLkCuFhSevI#(GNlY~zVT6F)J|J~?7d
zoQd@&?R@N$9anaoiwu27?><8PaB{y?&l;NAx$r!9{?o^{U93mfD)$W;t*(yv_eTz0
zna94EF?<BW-|`^Xaf{?lODyJ8Nr2seBi6iZ*I5bcG@HaG*b<L1GB6O`L!N^&rAy8g
zn**;Fk+I#pC4<tZrWeQaje9v`8di~@St9Ep==Q&Sj_VjRuDRPQfKWcO33XfvR1@bG
zrdmZ4w9@*NqCQ7O1ym*pghdp~5)}|oHa9>f$s`$>WWp>6=o5r$Z4nX0-L_a;H`KbI
zXz__+T>$k7*orGv>r$<?VqIDK?o0-fV#_(0KX>lE|98K;%=H|P?<5uf82@Q$RqU4$
z6WS7ca}Qoqz2chQo6{O1O>QfFbD(w4xyF8#Qv+)cQC*#n*3RzHGL)JA;f$ZNhNc#O
zC7=A6q0s(tkHOogn#)hsIz5zDBzGy;*rS`r`25wC+>KjVf8Zm|6n;0vHzc}XTWDmk
zA?WMTH5WHf&lVhe>~CMSGC9oNPUZE%xwX53S{fXQmfFQl=FEr-2TBTt=U1*<S)6lZ
zVJfB@`Np|yb5q!=YrdC9J!?9$%j}q2x8`W2arL;WQ+rRjI<9m+bfLy+71pxkvx3Hi
z_<-=$4LQ0lh3Cde58X^H?;oH%GIXpW;mV7k^M`x+<Hy*l86o%GU3#EV-S+w)Pp~^U
zfYtAhm2U1@wDrSHJ+9YJh&^biGeivfL#g!%-!-!PvPnNKN!Zq9PByhEW^eM&|HVj^
zZ7tnSFN{C8cutysb98U}`upED4R@P+A%9J?=~UIA^;_=@bUO5U&y?um)}uwC{et!m
zQLk~%_Bmr$7Vfg3=6B~^J9YJszU=wz1!;1-K6OK%FvEmTBj-9e=Z+tH%wxyH!}5j+
z=NqRDQ$$TW)id@MeO6snlvKzRlaHx8zSZT@jjTt&dH4F_KBL?Z?Em28K(x=69p5i^
z8k{!$lbcICeJWSv)pz~(=IZ13>lSY|<=&Y+X`HT9;<zSz*rP<V>x02A_I;?AZkdZl
zUOznV#?|AalPbUL{acOQ7M~-_>xZWkl@(0GPk!f9$JNHIWA0@Yr#n&eXKp^|v-s}D
zxJc!mv|U5}8hhlwntAqTmr*52CXXj)3w_Xi39r6#%JluD^qbPKu(MZB+$2X2{Bzad
z^uUJEG3*b)$p`aa{#U<dmqSVgvaP%0tTTxv%}*ccZf`Gdy_my<=XfNvc{_El3p+D2
zY3T4TKI?w`=5LbjC&#aJ7^oZM=IE4s#}`pI>DE2`sij0ZCeu4$M*NhW*_SR&Q}F+-
z`KFL(lTvh)>&f*e-#Z&B%W5bsHGEzDQASfQ`3c8={e0)*i+5|gP4iwmt<m9{?sb>@
zEzA3<k5rvraCzd8mpbW~z{f@Q+`#8<$0j#x`wshl2YS`6Ii{u9&TaZ7O6FE1?{*=o
z^-9IJh0DrTs?+1`PJH|_x2$^Yk|X+CceAy#1Jwf-C}OT(v|p1jY<<L#E7zO5*G*74
zR6q5WI!p5>|L4b=6I&iH9n^1R`s5{jC!^W7dcCr{)#vWVr*4=1uzTXu*6nNOY|Zw6
zq&KCz-M(|QZq>=l>7SIHmi;b?)aPk6hh1WV7Hquyc$1Rq^*QtG<n8!gOWYr}6$abo
zK9CmlTdRKB&o94dV8tTmGg%k5Fs&Qv`wafi_HTBK$GDi{=RGHG-`_fi&S|(-mRvdD
zww661bA5Tw<@38`)p=J~Pa+askY6*b`=Q07qjF|k9n<#gK|{{Q*O$Y#g^lPNoH5Ao
zdY@;X#r55o=8Lrl=j@-+OW&}&MrYW$H^8)3^20|jCakKIUmCk~e`1~2(g)>b?&g4s
zo3eZMC9BJ%<(j^`7rHM_XuOspO?eeO`1g_(U(NQ<s?7cATkid&G0$gJ4ZNvZ9pPN?
z0ex^}YGLYyAZfU2#HQ%2^TRbu>0h0m+`O%>JlmLeqw4^>Cq*6#4;OXMohz~xss7>1
zy<4LL_CCM<u3sL}km<eR{>vMaTJzWgP4n^W$VKy7)-RA=%$|PoLeA{2lb8OIIBY<~
zlOC_PWz`0ZDE_*wJmT|)T36rJ{t-`^cQT0zmx_gj{b%h?{Ik6879YO_$u+U3e_2%S
zT(C1^KH}H>>A5H`e6MfMo(|1Bf@^1eKi@U|;YwxF#}l6I3FV?5#ZGEz$(yn7*BGfb
z=lx8Qyi#;)V&$w2n|kXrr?gcL9&r1+tPhv1^S`nCi)6!_V*ISiEjXaweC?x++8yc3
z>K?mQZ@;JQ@7VTB_?3g-8QqsN`y7n--Io{mx0U3EwCRSvxsmveYN7nWoAYc<Y;OPB
zq~>2^&X>?*N*$(GY?<D+a|QP(xv@AN)8tjVOY%zB-5KRQMuMBP3`^sfkv~!Dxe~b-
zB18DYxN!&?n_<QgbRuP95j2KVv%*(iC<iG)Gi<N{qd^~?f{AB^Qi-6`2`y`25V<Ej
zik6@%lZMg~COu+x*)EcxA%sqc)3`~4GstY-m4}I9wYUyJwG@kKG&p~^j(}eherZg@
z5|jx+4V+1j(VUUQICwsVAu+~)prr6AK`A|D!c!2G=5MunRMU7G#A75(9AhJ%rNL<4
zXQucmd`S5FNt|KeYVkA|001tXfQE+yNbyt-gA-xW5hjAo00>|r2qmZ<cx53ri;5Ye
zSpp-q5HEr<oY9C`A~DmHmeW8GX1GaDV?4rC4m>jep2Q_6W5zTf!_ze+W)z=k)%^9G
zQH|3Kp$F_*iX<Tws2Qg%lwn3VGMGs#v3R2iy@9_31FlJ2*M1bmkUpFy7@*8d^4w@B
zzy~a6m<>312Ui3Io<y26ahgI<pcp~fd2nJ4G!ADS#Q_2x!AHOl>F@{84c55jY7@a5
zXeg30fhXKZ2uBNJpk!f5<ctp!XDkF^fEE5}WDtS|BWMVMs-X4|bP|G2M$k|R8it_Z
zPyq-UfuNBH8ik-!5p)`YMnlO$1tVx2g2qElBIpbRor$2c5L69il!8Ncc^cGk7RyGI
zcW|O12n|7NIAEPKX@O1+MMDTeb}Yugks4wG{viU09Wn^4Nl-{7R144vX##5?0cbLV
zDbytJ0+fME1WiNGc?b$jLE&UtaN+H!Xo-Lmm=_X<F@T08NG)8`(IP62PXu9NaVQ+P
z%33#SjN$Q+6tsX?0Z=m*DNd4vnIQm$7Sro-0A%?TipmnOj>G{Kq@6WT9A8C1rq~vG
zjv$57v;d7#0p~_s4@}UQmeBApFf$Z&x|q=pu$>Q9SQbwLc)%bAtma4(2by3m3B8RQ
zEuSI>EZl}0F**fmN$lGWWTFQO7zn{h7NC(-CT`Nh_nBG(I#LL}1COP%wA4Ic%F;Z1
zH;}^Y&~P+AoFG3Ld?~AGDg`&G`A%bSaTHlSYAADta5qdKTC0N+x70ijjKq!nJmD*j
zr1U&bFg>gtM&o*dzZD0jjhIH5M_R}#4g-L1NpWUDyD^;5jetOCJL^|W!?6&d5$X!M
z9D|B5LnT^Tmk-P0T!U)?3{V2?YqbTqMyN0jf(92Ic-w<9@wueIgJ5RBNGNpSc2Nwa
z)ZVW=jnJ>+APW>gNzg#svI-Pa7-(fK&@!n5RnX)e2vwpJmFPre>qKQW4i)u6MFv7`
z^+K(BDi-*N6fchwjiN-OC|jebb|tH^kt12cOhO?<iPIuZ^bjX{h_m$&XSJLuK94m}
zG*AN#XM~1QlRz4uq%_DdCaPezF5vwjbUSoad&y!JnlQ)_ycxwr^1aVH!9^#ytrJjc
zHzq{9gs3OZHcP7$?Rwycg*T~5yqpq^DA9<rbxyS_wHw1EA<_sffT6@!1t;2bqCID8
z&sps<#OE+?c#<+Yo+L^F0wib<j5Es`5=j`<920|?ri_Wk^-$})I0R=jG+|~b8j&mc
zr2_#YaiQUtkoJ9Mq8t(S8{6l$`%a+Tc1KBPC&jSv28Wsq<Tn!yhRAuxXrSVs)jYgy
z6+Qn^5jH5RKd*mOlX)lwMgB7&z$Pp{Off8Ltlobo^!YoX!YUNEys4NTRxmW2(DI9z
zTn-A+NFu~eaj=33fw!-SHXhMne#rseT_Z5oOu~pG)JU4Ahg{|%8;uxP*367WVWPD^
z^Pp(Gn*^OIEIlw!Y|YGC9m2nj;hJ_ApekGdJ6{N&3D=OZ3YkJFlgqv2UNV`|Z<I_i
zQYIVeX2VQ6jq4;bMD8n*b^IXSN~Na~(IFi&IdH}QASO}f>*FnZTjr_oeM`^NTlSVr
z=>>Jx$)D02YOAx%7ZUdu8GP}6TjuErB>zRv)6dVZGd`J2A?u6}^n5zs3sk)wuTtg%
zN&d?nUUC_~w0Dvzl%4f}D{mVt@Gd4uoEHAd5JP0*z&-3=6ujRoRfL9`bQB`z%SPxt
WhJ}}jkS1RrUxiXKe7GuVqU8UI(|DNx
literal 0
HcmV?d00001
diff --git a/5_dpnn/results/gpu_4/slurm-25137704.out b/5_dpnn/results/gpu_4/slurm-25137704.out
new file mode 100644
index 0000000..9bef85a
--- /dev/null
+++ b/5_dpnn/results/gpu_4/slurm-25137704.out
@@ -0,0 +1,4533 @@
+GpuFreq=control_disabled
+Rank, world size, device count: 3, 4, 4
+Rank, world size, device count: 2, 4, 4
+Rank, world size, device count: 1, 4, 4
+Rank, world size, device count: 0, 4, 4
+Distributed package available...[OK]
+NCCL backend available...[OK]
+Process group initialized successfully...[OK]
+nccl backend used...[OK]
+Process group initialized successfully...[OK]
+nccl backend used...[OK]
+Process group initialized successfully...[OK]
+nccl backend used...[OK]
+Process group initialized successfully...[OK]
+nccl backend used...[OK]
+Files already downloaded and verified
+Epoch: 001/100 | Batch 0000/0043 | Averaged Loss: 2.3026
+Epoch: 001/100 | Batch 0001/0043 | Averaged Loss: 2.3028
+Epoch: 001/100 | Batch 0002/0043 | Averaged Loss: 2.3028
+Epoch: 001/100 | Batch 0003/0043 | Averaged Loss: 2.3029
+Epoch: 001/100 | Batch 0004/0043 | Averaged Loss: 2.3029
+Epoch: 001/100 | Batch 0005/0043 | Averaged Loss: 2.3028
+Epoch: 001/100 | Batch 0006/0043 | Averaged Loss: 2.3024
+Epoch: 001/100 | Batch 0007/0043 | Averaged Loss: 2.3024
+Epoch: 001/100 | Batch 0008/0043 | Averaged Loss: 2.3027
+Epoch: 001/100 | Batch 0009/0043 | Averaged Loss: 2.3033
+Epoch: 001/100 | Batch 0010/0043 | Averaged Loss: 2.3021
+Epoch: 001/100 | Batch 0011/0043 | Averaged Loss: 2.3030
+Epoch: 001/100 | Batch 0012/0043 | Averaged Loss: 2.3023
+Epoch: 001/100 | Batch 0013/0043 | Averaged Loss: 2.3028
+Epoch: 001/100 | Batch 0014/0043 | Averaged Loss: 2.3026
+Epoch: 001/100 | Batch 0015/0043 | Averaged Loss: 2.3022
+Epoch: 001/100 | Batch 0016/0043 | Averaged Loss: 2.3020
+Epoch: 001/100 | Batch 0017/0043 | Averaged Loss: 2.3025
+Epoch: 001/100 | Batch 0018/0043 | Averaged Loss: 2.3019
+Epoch: 001/100 | Batch 0019/0043 | Averaged Loss: 2.3031
+Epoch: 001/100 | Batch 0020/0043 | Averaged Loss: 2.3025
+Epoch: 001/100 | Batch 0021/0043 | Averaged Loss: 2.3033
+Epoch: 001/100 | Batch 0022/0043 | Averaged Loss: 2.3024
+Epoch: 001/100 | Batch 0023/0043 | Averaged Loss: 2.3029
+Epoch: 001/100 | Batch 0024/0043 | Averaged Loss: 2.3007
+Epoch: 001/100 | Batch 0025/0043 | Averaged Loss: 2.3027
+Epoch: 001/100 | Batch 0026/0043 | Averaged Loss: 2.3020
+Epoch: 001/100 | Batch 0027/0043 | Averaged Loss: 2.3024
+Epoch: 001/100 | Batch 0028/0043 | Averaged Loss: 2.3014
+Epoch: 001/100 | Batch 0029/0043 | Averaged Loss: 2.3013
+Epoch: 001/100 | Batch 0030/0043 | Averaged Loss: 2.3013
+Epoch: 001/100 | Batch 0031/0043 | Averaged Loss: 2.3008
+Epoch: 001/100 | Batch 0032/0043 | Averaged Loss: 2.2999
+Epoch: 001/100 | Batch 0033/0043 | Averaged Loss: 2.3011
+Epoch: 001/100 | Batch 0034/0043 | Averaged Loss: 2.3001
+Epoch: 001/100 | Batch 0035/0043 | Averaged Loss: 2.3004
+Epoch: 001/100 | Batch 0036/0043 | Averaged Loss: 2.2987
+Epoch: 001/100 | Batch 0037/0043 | Averaged Loss: 2.2994
+Epoch: 001/100 | Batch 0038/0043 | Averaged Loss: 2.2990
+Epoch: 001/100 | Batch 0039/0043 | Averaged Loss: 2.2976
+Epoch: 001/100 | Batch 0040/0043 | Averaged Loss: 2.2968
+Epoch: 001/100 | Batch 0041/0043 | Averaged Loss: 2.2962
+Epoch: 001/100 | Batch 0042/0043 | Averaged Loss: 2.2980
+Epoch: 001/100 | Train: 9.99% | Validation: 9.81%
+Time elapsed: 0.1652088314294815 min
+Epoch: 002/100 | Batch 0000/0043 | Averaged Loss: 2.2924
+Epoch: 002/100 | Batch 0001/0043 | Averaged Loss: 2.2916
+Epoch: 002/100 | Batch 0002/0043 | Averaged Loss: 2.2914
+Epoch: 002/100 | Batch 0003/0043 | Averaged Loss: 2.2886
+Epoch: 002/100 | Batch 0004/0043 | Averaged Loss: 2.2819
+Epoch: 002/100 | Batch 0005/0043 | Averaged Loss: 2.2720
+Epoch: 002/100 | Batch 0006/0043 | Averaged Loss: 2.2683
+Epoch: 002/100 | Batch 0007/0043 | Averaged Loss: 2.2625
+Epoch: 002/100 | Batch 0008/0043 | Averaged Loss: 2.2440
+Epoch: 002/100 | Batch 0009/0043 | Averaged Loss: 2.2357
+Epoch: 002/100 | Batch 0010/0043 | Averaged Loss: 2.2125
+Epoch: 002/100 | Batch 0011/0043 | Averaged Loss: 2.2076
+Epoch: 002/100 | Batch 0012/0043 | Averaged Loss: 2.2034
+Epoch: 002/100 | Batch 0013/0043 | Averaged Loss: 2.1835
+Epoch: 002/100 | Batch 0014/0043 | Averaged Loss: 2.1845
+Epoch: 002/100 | Batch 0015/0043 | Averaged Loss: 2.1537
+Epoch: 002/100 | Batch 0016/0043 | Averaged Loss: 2.1681
+Epoch: 002/100 | Batch 0017/0043 | Averaged Loss: 2.1414
+Epoch: 002/100 | Batch 0018/0043 | Averaged Loss: 2.1275
+Epoch: 002/100 | Batch 0019/0043 | Averaged Loss: 2.1038
+Epoch: 002/100 | Batch 0020/0043 | Averaged Loss: 2.1233
+Epoch: 002/100 | Batch 0021/0043 | Averaged Loss: 2.0754
+Epoch: 002/100 | Batch 0022/0043 | Averaged Loss: 2.0939
+Epoch: 002/100 | Batch 0023/0043 | Averaged Loss: 2.1617
+Epoch: 002/100 | Batch 0024/0043 | Averaged Loss: 2.0722
+Epoch: 002/100 | Batch 0025/0043 | Averaged Loss: 2.0494
+Epoch: 002/100 | Batch 0026/0043 | Averaged Loss: 2.0706
+Epoch: 002/100 | Batch 0027/0043 | Averaged Loss: 2.0697
+Epoch: 002/100 | Batch 0028/0043 | Averaged Loss: 2.0589
+Epoch: 002/100 | Batch 0029/0043 | Averaged Loss: 1.9921
+Epoch: 002/100 | Batch 0030/0043 | Averaged Loss: 2.2638
+Epoch: 002/100 | Batch 0031/0043 | Averaged Loss: 2.4497
+Epoch: 002/100 | Batch 0032/0043 | Averaged Loss: 2.4843
+Epoch: 002/100 | Batch 0033/0043 | Averaged Loss: 2.4583
+Epoch: 002/100 | Batch 0034/0043 | Averaged Loss: 2.3793
+Epoch: 002/100 | Batch 0035/0043 | Averaged Loss: 2.3701
+Epoch: 002/100 | Batch 0036/0043 | Averaged Loss: 2.3343
+Epoch: 002/100 | Batch 0037/0043 | Averaged Loss: 2.3449
+Epoch: 002/100 | Batch 0038/0043 | Averaged Loss: 2.3433
+Epoch: 002/100 | Batch 0039/0043 | Averaged Loss: 2.3145
+Epoch: 002/100 | Batch 0040/0043 | Averaged Loss: 2.3397
+Epoch: 002/100 | Batch 0041/0043 | Averaged Loss: 2.3128
+Epoch: 002/100 | Batch 0042/0043 | Averaged Loss: 2.3062
+Epoch: 002/100 | Train: 10.45% | Validation: 10.62%
+Time elapsed: 0.3136664927005768 min
+Epoch: 003/100 | Batch 0000/0043 | Averaged Loss: 2.2884
+Epoch: 003/100 | Batch 0001/0043 | Averaged Loss: 2.2631
+Epoch: 003/100 | Batch 0002/0043 | Averaged Loss: 2.2518
+Epoch: 003/100 | Batch 0003/0043 | Averaged Loss: 2.4154
+Epoch: 003/100 | Batch 0004/0043 | Averaged Loss: 2.3330
+Epoch: 003/100 | Batch 0005/0043 | Averaged Loss: 2.2992
+Epoch: 003/100 | Batch 0006/0043 | Averaged Loss: 2.3043
+Epoch: 003/100 | Batch 0007/0043 | Averaged Loss: 2.2936
+Epoch: 003/100 | Batch 0008/0043 | Averaged Loss: 2.2717
+Epoch: 003/100 | Batch 0009/0043 | Averaged Loss: 2.2917
+Epoch: 003/100 | Batch 0010/0043 | Averaged Loss: 2.2813
+Epoch: 003/100 | Batch 0011/0043 | Averaged Loss: 2.2694
+Epoch: 003/100 | Batch 0012/0043 | Averaged Loss: 2.2600
+Epoch: 003/100 | Batch 0013/0043 | Averaged Loss: 2.2676
+Epoch: 003/100 | Batch 0014/0043 | Averaged Loss: 2.2738
+Epoch: 003/100 | Batch 0015/0043 | Averaged Loss: 2.2637
+Epoch: 003/100 | Batch 0016/0043 | Averaged Loss: 2.2872
+Epoch: 003/100 | Batch 0017/0043 | Averaged Loss: 2.2833
+Epoch: 003/100 | Batch 0018/0043 | Averaged Loss: 2.3006
+Epoch: 003/100 | Batch 0019/0043 | Averaged Loss: 2.3095
+Epoch: 003/100 | Batch 0020/0043 | Averaged Loss: 2.2829
+Epoch: 003/100 | Batch 0021/0043 | Averaged Loss: 2.2541
+Epoch: 003/100 | Batch 0022/0043 | Averaged Loss: 2.2304
+Epoch: 003/100 | Batch 0023/0043 | Averaged Loss: 2.2193
+Epoch: 003/100 | Batch 0024/0043 | Averaged Loss: 2.2462
+Epoch: 003/100 | Batch 0025/0043 | Averaged Loss: 2.2758
+Epoch: 003/100 | Batch 0026/0043 | Averaged Loss: 2.2629
+Epoch: 003/100 | Batch 0027/0043 | Averaged Loss: 2.2809
+Epoch: 003/100 | Batch 0028/0043 | Averaged Loss: 2.2762
+Epoch: 003/100 | Batch 0029/0043 | Averaged Loss: 2.2958
+Epoch: 003/100 | Batch 0030/0043 | Averaged Loss: 2.2730
+Epoch: 003/100 | Batch 0031/0043 | Averaged Loss: 2.2857
+Epoch: 003/100 | Batch 0032/0043 | Averaged Loss: 2.2297
+Epoch: 003/100 | Batch 0033/0043 | Averaged Loss: 2.2150
+Epoch: 003/100 | Batch 0034/0043 | Averaged Loss: 2.1955
+Epoch: 003/100 | Batch 0035/0043 | Averaged Loss: 2.1994
+Epoch: 003/100 | Batch 0036/0043 | Averaged Loss: 2.2026
+Epoch: 003/100 | Batch 0037/0043 | Averaged Loss: 2.1891
+Epoch: 003/100 | Batch 0038/0043 | Averaged Loss: 2.1737
+Epoch: 003/100 | Batch 0039/0043 | Averaged Loss: 2.2221
+Epoch: 003/100 | Batch 0040/0043 | Averaged Loss: 2.2104
+Epoch: 003/100 | Batch 0041/0043 | Averaged Loss: 2.1858
+Epoch: 003/100 | Batch 0042/0043 | Averaged Loss: 2.1990
+Epoch: 003/100 | Train: 15.85% | Validation: 16.06%
+Time elapsed: 0.46184664964675903 min
+Epoch: 004/100 | Batch 0000/0043 | Averaged Loss: 2.1834
+Epoch: 004/100 | Batch 0001/0043 | Averaged Loss: 2.1733
+Epoch: 004/100 | Batch 0002/0043 | Averaged Loss: 2.1277
+Epoch: 004/100 | Batch 0003/0043 | Averaged Loss: 2.1711
+Epoch: 004/100 | Batch 0004/0043 | Averaged Loss: 2.1740
+Epoch: 004/100 | Batch 0005/0043 | Averaged Loss: 2.1777
+Epoch: 004/100 | Batch 0006/0043 | Averaged Loss: 2.1184
+Epoch: 004/100 | Batch 0007/0043 | Averaged Loss: 2.1536
+Epoch: 004/100 | Batch 0008/0043 | Averaged Loss: 2.1973
+Epoch: 004/100 | Batch 0009/0043 | Averaged Loss: 2.1479
+Epoch: 004/100 | Batch 0010/0043 | Averaged Loss: 2.2709
+Epoch: 004/100 | Batch 0011/0043 | Averaged Loss: 2.2843
+Epoch: 004/100 | Batch 0012/0043 | Averaged Loss: 2.3528
+Epoch: 004/100 | Batch 0013/0043 | Averaged Loss: 2.3912
+Epoch: 004/100 | Batch 0014/0043 | Averaged Loss: 2.3658
+Epoch: 004/100 | Batch 0015/0043 | Averaged Loss: 2.3644
+Epoch: 004/100 | Batch 0016/0043 | Averaged Loss: 2.3632
+Epoch: 004/100 | Batch 0017/0043 | Averaged Loss: 2.3544
+Epoch: 004/100 | Batch 0018/0043 | Averaged Loss: 2.3361
+Epoch: 004/100 | Batch 0019/0043 | Averaged Loss: 2.3291
+Epoch: 004/100 | Batch 0020/0043 | Averaged Loss: 2.3285
+Epoch: 004/100 | Batch 0021/0043 | Averaged Loss: 2.3003
+Epoch: 004/100 | Batch 0022/0043 | Averaged Loss: 2.2738
+Epoch: 004/100 | Batch 0023/0043 | Averaged Loss: 2.2558
+Epoch: 004/100 | Batch 0024/0043 | Averaged Loss: 2.2184
+Epoch: 004/100 | Batch 0025/0043 | Averaged Loss: 2.2260
+Epoch: 004/100 | Batch 0026/0043 | Averaged Loss: 2.2459
+Epoch: 004/100 | Batch 0027/0043 | Averaged Loss: 2.2084
+Epoch: 004/100 | Batch 0028/0043 | Averaged Loss: 2.1881
+Epoch: 004/100 | Batch 0029/0043 | Averaged Loss: 2.2240
+Epoch: 004/100 | Batch 0030/0043 | Averaged Loss: 2.2863
+Epoch: 004/100 | Batch 0031/0043 | Averaged Loss: 2.3089
+Epoch: 004/100 | Batch 0032/0043 | Averaged Loss: 2.2979
+Epoch: 004/100 | Batch 0033/0043 | Averaged Loss: 2.2643
+Epoch: 004/100 | Batch 0034/0043 | Averaged Loss: 2.2536
+Epoch: 004/100 | Batch 0035/0043 | Averaged Loss: 2.2235
+Epoch: 004/100 | Batch 0036/0043 | Averaged Loss: 2.1966
+Epoch: 004/100 | Batch 0037/0043 | Averaged Loss: 2.1359
+Epoch: 004/100 | Batch 0038/0043 | Averaged Loss: 2.4440
+Epoch: 004/100 | Batch 0039/0043 | Averaged Loss: 2.3121
+Epoch: 004/100 | Batch 0040/0043 | Averaged Loss: 2.3021
+Epoch: 004/100 | Batch 0041/0043 | Averaged Loss: 2.3010
+Epoch: 004/100 | Batch 0042/0043 | Averaged Loss: 2.3122
+Epoch: 004/100 | Train: 12.38% | Validation: 12.57%
+Time elapsed: 0.6100757122039795 min
+Epoch: 005/100 | Batch 0000/0043 | Averaged Loss: 2.2898
+Epoch: 005/100 | Batch 0001/0043 | Averaged Loss: 2.2917
+Epoch: 005/100 | Batch 0002/0043 | Averaged Loss: 2.2888
+Epoch: 005/100 | Batch 0003/0043 | Averaged Loss: 2.2697
+Epoch: 005/100 | Batch 0004/0043 | Averaged Loss: 2.2627
+Epoch: 005/100 | Batch 0005/0043 | Averaged Loss: 2.2586
+Epoch: 005/100 | Batch 0006/0043 | Averaged Loss: 2.2685
+Epoch: 005/100 | Batch 0007/0043 | Averaged Loss: 2.2834
+Epoch: 005/100 | Batch 0008/0043 | Averaged Loss: 2.2846
+Epoch: 005/100 | Batch 0009/0043 | Averaged Loss: 2.2378
+Epoch: 005/100 | Batch 0010/0043 | Averaged Loss: 2.2573
+Epoch: 005/100 | Batch 0011/0043 | Averaged Loss: 2.2514
+Epoch: 005/100 | Batch 0012/0043 | Averaged Loss: 2.2541
+Epoch: 005/100 | Batch 0013/0043 | Averaged Loss: 2.2816
+Epoch: 005/100 | Batch 0014/0043 | Averaged Loss: 2.2883
+Epoch: 005/100 | Batch 0015/0043 | Averaged Loss: 2.3075
+Epoch: 005/100 | Batch 0016/0043 | Averaged Loss: 2.2799
+Epoch: 005/100 | Batch 0017/0043 | Averaged Loss: 2.2759
+Epoch: 005/100 | Batch 0018/0043 | Averaged Loss: 2.2325
+Epoch: 005/100 | Batch 0019/0043 | Averaged Loss: 2.2253
+Epoch: 005/100 | Batch 0020/0043 | Averaged Loss: 2.2928
+Epoch: 005/100 | Batch 0021/0043 | Averaged Loss: 2.3204
+Epoch: 005/100 | Batch 0022/0043 | Averaged Loss: 2.3101
+Epoch: 005/100 | Batch 0023/0043 | Averaged Loss: 2.2624
+Epoch: 005/100 | Batch 0024/0043 | Averaged Loss: 2.3925
+Epoch: 005/100 | Batch 0025/0043 | Averaged Loss: 2.3244
+Epoch: 005/100 | Batch 0026/0043 | Averaged Loss: 2.3057
+Epoch: 005/100 | Batch 0027/0043 | Averaged Loss: 2.2821
+Epoch: 005/100 | Batch 0028/0043 | Averaged Loss: 2.2857
+Epoch: 005/100 | Batch 0029/0043 | Averaged Loss: 2.3215
+Epoch: 005/100 | Batch 0030/0043 | Averaged Loss: 2.2909
+Epoch: 005/100 | Batch 0031/0043 | Averaged Loss: 2.3026
+Epoch: 005/100 | Batch 0032/0043 | Averaged Loss: 2.2798
+Epoch: 005/100 | Batch 0033/0043 | Averaged Loss: 2.2592
+Epoch: 005/100 | Batch 0034/0043 | Averaged Loss: 2.2695
+Epoch: 005/100 | Batch 0035/0043 | Averaged Loss: 2.2369
+Epoch: 005/100 | Batch 0036/0043 | Averaged Loss: 2.1995
+Epoch: 005/100 | Batch 0037/0043 | Averaged Loss: 2.2087
+Epoch: 005/100 | Batch 0038/0043 | Averaged Loss: 2.1972
+Epoch: 005/100 | Batch 0039/0043 | Averaged Loss: 2.2144
+Epoch: 005/100 | Batch 0040/0043 | Averaged Loss: 2.2517
+Epoch: 005/100 | Batch 0041/0043 | Averaged Loss: 2.2670
+Epoch: 005/100 | Batch 0042/0043 | Averaged Loss: 2.3374
+Epoch: 005/100 | Train: 9.99% | Validation: 10.01%
+Time elapsed: 0.7582553625106812 min
+Epoch: 006/100 | Batch 0000/0043 | Averaged Loss: 2.3456
+Epoch: 006/100 | Batch 0001/0043 | Averaged Loss: 2.3378
+Epoch: 006/100 | Batch 0002/0043 | Averaged Loss: 2.3432
+Epoch: 006/100 | Batch 0003/0043 | Averaged Loss: 2.3105
+Epoch: 006/100 | Batch 0004/0043 | Averaged Loss: 2.2831
+Epoch: 006/100 | Batch 0005/0043 | Averaged Loss: 2.2871
+Epoch: 006/100 | Batch 0006/0043 | Averaged Loss: 2.2644
+Epoch: 006/100 | Batch 0007/0043 | Averaged Loss: 2.2675
+Epoch: 006/100 | Batch 0008/0043 | Averaged Loss: 2.2420
+Epoch: 006/100 | Batch 0009/0043 | Averaged Loss: 2.2156
+Epoch: 006/100 | Batch 0010/0043 | Averaged Loss: 2.1847
+Epoch: 006/100 | Batch 0011/0043 | Averaged Loss: 2.1642
+Epoch: 006/100 | Batch 0012/0043 | Averaged Loss: 2.2089
+Epoch: 006/100 | Batch 0013/0043 | Averaged Loss: 2.1880
+Epoch: 006/100 | Batch 0014/0043 | Averaged Loss: 2.2147
+Epoch: 006/100 | Batch 0015/0043 | Averaged Loss: 2.1878
+Epoch: 006/100 | Batch 0016/0043 | Averaged Loss: 2.2016
+Epoch: 006/100 | Batch 0017/0043 | Averaged Loss: 2.1109
+Epoch: 006/100 | Batch 0018/0043 | Averaged Loss: 2.1342
+Epoch: 006/100 | Batch 0019/0043 | Averaged Loss: 2.1370
+Epoch: 006/100 | Batch 0020/0043 | Averaged Loss: 2.1458
+Epoch: 006/100 | Batch 0021/0043 | Averaged Loss: 2.1194
+Epoch: 006/100 | Batch 0022/0043 | Averaged Loss: 2.1207
+Epoch: 006/100 | Batch 0023/0043 | Averaged Loss: 2.0959
+Epoch: 006/100 | Batch 0024/0043 | Averaged Loss: 2.1041
+Epoch: 006/100 | Batch 0025/0043 | Averaged Loss: 2.1128
+Epoch: 006/100 | Batch 0026/0043 | Averaged Loss: 2.0876
+Epoch: 006/100 | Batch 0027/0043 | Averaged Loss: 2.0895
+Epoch: 006/100 | Batch 0028/0043 | Averaged Loss: 2.1164
+Epoch: 006/100 | Batch 0029/0043 | Averaged Loss: 2.1080
+Epoch: 006/100 | Batch 0030/0043 | Averaged Loss: 2.0799
+Epoch: 006/100 | Batch 0031/0043 | Averaged Loss: 2.0558
+Epoch: 006/100 | Batch 0032/0043 | Averaged Loss: 2.1377
+Epoch: 006/100 | Batch 0033/0043 | Averaged Loss: 2.1114
+Epoch: 006/100 | Batch 0034/0043 | Averaged Loss: 2.1761
+Epoch: 006/100 | Batch 0035/0043 | Averaged Loss: 2.1193
+Epoch: 006/100 | Batch 0036/0043 | Averaged Loss: 2.0728
+Epoch: 006/100 | Batch 0037/0043 | Averaged Loss: 2.1474
+Epoch: 006/100 | Batch 0038/0043 | Averaged Loss: 2.1097
+Epoch: 006/100 | Batch 0039/0043 | Averaged Loss: 2.1259
+Epoch: 006/100 | Batch 0040/0043 | Averaged Loss: 2.0837
+Epoch: 006/100 | Batch 0041/0043 | Averaged Loss: 2.0732
+Epoch: 006/100 | Batch 0042/0043 | Averaged Loss: 2.0781
+Epoch: 006/100 | Train: 17.60% | Validation: 17.75%
+Time elapsed: 0.9062973856925964 min
+Epoch: 007/100 | Batch 0000/0043 | Averaged Loss: 2.0640
+Epoch: 007/100 | Batch 0001/0043 | Averaged Loss: 2.0814
+Epoch: 007/100 | Batch 0002/0043 | Averaged Loss: 2.0088
+Epoch: 007/100 | Batch 0003/0043 | Averaged Loss: 2.0949
+Epoch: 007/100 | Batch 0004/0043 | Averaged Loss: 2.0075
+Epoch: 007/100 | Batch 0005/0043 | Averaged Loss: 2.0750
+Epoch: 007/100 | Batch 0006/0043 | Averaged Loss: 2.0164
+Epoch: 007/100 | Batch 0007/0043 | Averaged Loss: 2.0465
+Epoch: 007/100 | Batch 0008/0043 | Averaged Loss: 2.0038
+Epoch: 007/100 | Batch 0009/0043 | Averaged Loss: 2.0232
+Epoch: 007/100 | Batch 0010/0043 | Averaged Loss: 2.0104
+Epoch: 007/100 | Batch 0011/0043 | Averaged Loss: 2.0099
+Epoch: 007/100 | Batch 0012/0043 | Averaged Loss: 1.9894
+Epoch: 007/100 | Batch 0013/0043 | Averaged Loss: 1.9826
+Epoch: 007/100 | Batch 0014/0043 | Averaged Loss: 1.9901
+Epoch: 007/100 | Batch 0015/0043 | Averaged Loss: 1.9899
+Epoch: 007/100 | Batch 0016/0043 | Averaged Loss: 1.9863
+Epoch: 007/100 | Batch 0017/0043 | Averaged Loss: 1.9630
+Epoch: 007/100 | Batch 0018/0043 | Averaged Loss: 1.9693
+Epoch: 007/100 | Batch 0019/0043 | Averaged Loss: 1.9583
+Epoch: 007/100 | Batch 0020/0043 | Averaged Loss: 1.9686
+Epoch: 007/100 | Batch 0021/0043 | Averaged Loss: 1.9817
+Epoch: 007/100 | Batch 0022/0043 | Averaged Loss: 1.9498
+Epoch: 007/100 | Batch 0023/0043 | Averaged Loss: 2.0651
+Epoch: 007/100 | Batch 0024/0043 | Averaged Loss: 2.0604
+Epoch: 007/100 | Batch 0025/0043 | Averaged Loss: 2.0194
+Epoch: 007/100 | Batch 0026/0043 | Averaged Loss: 2.0948
+Epoch: 007/100 | Batch 0027/0043 | Averaged Loss: 1.9420
+Epoch: 007/100 | Batch 0028/0043 | Averaged Loss: 2.0318
+Epoch: 007/100 | Batch 0029/0043 | Averaged Loss: 1.9742
+Epoch: 007/100 | Batch 0030/0043 | Averaged Loss: 2.0427
+Epoch: 007/100 | Batch 0031/0043 | Averaged Loss: 1.9531
+Epoch: 007/100 | Batch 0032/0043 | Averaged Loss: 1.9724
+Epoch: 007/100 | Batch 0033/0043 | Averaged Loss: 1.9436
+Epoch: 007/100 | Batch 0034/0043 | Averaged Loss: 1.9730
+Epoch: 007/100 | Batch 0035/0043 | Averaged Loss: 1.9632
+Epoch: 007/100 | Batch 0036/0043 | Averaged Loss: 1.9508
+Epoch: 007/100 | Batch 0037/0043 | Averaged Loss: 1.9812
+Epoch: 007/100 | Batch 0038/0043 | Averaged Loss: 1.9144
+Epoch: 007/100 | Batch 0039/0043 | Averaged Loss: 1.9944
+Epoch: 007/100 | Batch 0040/0043 | Averaged Loss: 1.9858
+Epoch: 007/100 | Batch 0041/0043 | Averaged Loss: 1.9551
+Epoch: 007/100 | Batch 0042/0043 | Averaged Loss: 1.9049
+Epoch: 007/100 | Train: 18.47% | Validation: 17.55%
+Time elapsed: 1.0544540882110596 min
+Epoch: 008/100 | Batch 0000/0043 | Averaged Loss: 2.0407
+Epoch: 008/100 | Batch 0001/0043 | Averaged Loss: 1.9570
+Epoch: 008/100 | Batch 0002/0043 | Averaged Loss: 2.0130
+Epoch: 008/100 | Batch 0003/0043 | Averaged Loss: 1.9222
+Epoch: 008/100 | Batch 0004/0043 | Averaged Loss: 2.0517
+Epoch: 008/100 | Batch 0005/0043 | Averaged Loss: 1.9549
+Epoch: 008/100 | Batch 0006/0043 | Averaged Loss: 2.0241
+Epoch: 008/100 | Batch 0007/0043 | Averaged Loss: 1.9633
+Epoch: 008/100 | Batch 0008/0043 | Averaged Loss: 1.9792
+Epoch: 008/100 | Batch 0009/0043 | Averaged Loss: 1.9213
+Epoch: 008/100 | Batch 0010/0043 | Averaged Loss: 1.9214
+Epoch: 008/100 | Batch 0011/0043 | Averaged Loss: 1.9617
+Epoch: 008/100 | Batch 0012/0043 | Averaged Loss: 1.9310
+Epoch: 008/100 | Batch 0013/0043 | Averaged Loss: 2.0450
+Epoch: 008/100 | Batch 0014/0043 | Averaged Loss: 1.9343
+Epoch: 008/100 | Batch 0015/0043 | Averaged Loss: 2.0212
+Epoch: 008/100 | Batch 0016/0043 | Averaged Loss: 1.9231
+Epoch: 008/100 | Batch 0017/0043 | Averaged Loss: 1.9489
+Epoch: 008/100 | Batch 0018/0043 | Averaged Loss: 1.9532
+Epoch: 008/100 | Batch 0019/0043 | Averaged Loss: 1.9456
+Epoch: 008/100 | Batch 0020/0043 | Averaged Loss: 1.9508
+Epoch: 008/100 | Batch 0021/0043 | Averaged Loss: 1.9375
+Epoch: 008/100 | Batch 0022/0043 | Averaged Loss: 1.9423
+Epoch: 008/100 | Batch 0023/0043 | Averaged Loss: 1.9132
+Epoch: 008/100 | Batch 0024/0043 | Averaged Loss: 1.9575
+Epoch: 008/100 | Batch 0025/0043 | Averaged Loss: 1.9038
+Epoch: 008/100 | Batch 0026/0043 | Averaged Loss: 1.9259
+Epoch: 008/100 | Batch 0027/0043 | Averaged Loss: 1.9106
+Epoch: 008/100 | Batch 0028/0043 | Averaged Loss: 1.9284
+Epoch: 008/100 | Batch 0029/0043 | Averaged Loss: 1.9638
+Epoch: 008/100 | Batch 0030/0043 | Averaged Loss: 1.9044
+Epoch: 008/100 | Batch 0031/0043 | Averaged Loss: 1.9152
+Epoch: 008/100 | Batch 0032/0043 | Averaged Loss: 1.8942
+Epoch: 008/100 | Batch 0033/0043 | Averaged Loss: 1.9192
+Epoch: 008/100 | Batch 0034/0043 | Averaged Loss: 1.8967
+Epoch: 008/100 | Batch 0035/0043 | Averaged Loss: 1.8949
+Epoch: 008/100 | Batch 0036/0043 | Averaged Loss: 1.8697
+Epoch: 008/100 | Batch 0037/0043 | Averaged Loss: 1.9062
+Epoch: 008/100 | Batch 0038/0043 | Averaged Loss: 1.8893
+Epoch: 008/100 | Batch 0039/0043 | Averaged Loss: 1.9272
+Epoch: 008/100 | Batch 0040/0043 | Averaged Loss: 1.8620
+Epoch: 008/100 | Batch 0041/0043 | Averaged Loss: 1.8829
+Epoch: 008/100 | Batch 0042/0043 | Averaged Loss: 1.9079
+Epoch: 008/100 | Train: 20.61% | Validation: 19.85%
+Time elapsed: 1.202406883239746 min
+Epoch: 009/100 | Batch 0000/0043 | Averaged Loss: 1.9022
+Epoch: 009/100 | Batch 0001/0043 | Averaged Loss: 1.8865
+Epoch: 009/100 | Batch 0002/0043 | Averaged Loss: 1.8659
+Epoch: 009/100 | Batch 0003/0043 | Averaged Loss: 1.8859
+Epoch: 009/100 | Batch 0004/0043 | Averaged Loss: 1.8730
+Epoch: 009/100 | Batch 0005/0043 | Averaged Loss: 1.9070
+Epoch: 009/100 | Batch 0006/0043 | Averaged Loss: 1.8973
+Epoch: 009/100 | Batch 0007/0043 | Averaged Loss: 1.8867
+Epoch: 009/100 | Batch 0008/0043 | Averaged Loss: 1.8689
+Epoch: 009/100 | Batch 0009/0043 | Averaged Loss: 1.9180
+Epoch: 009/100 | Batch 0010/0043 | Averaged Loss: 1.8882
+Epoch: 009/100 | Batch 0011/0043 | Averaged Loss: 1.8846
+Epoch: 009/100 | Batch 0012/0043 | Averaged Loss: 1.8737
+Epoch: 009/100 | Batch 0013/0043 | Averaged Loss: 1.8777
+Epoch: 009/100 | Batch 0014/0043 | Averaged Loss: 1.9092
+Epoch: 009/100 | Batch 0015/0043 | Averaged Loss: 1.8663
+Epoch: 009/100 | Batch 0016/0043 | Averaged Loss: 1.8424
+Epoch: 009/100 | Batch 0017/0043 | Averaged Loss: 1.8666
+Epoch: 009/100 | Batch 0018/0043 | Averaged Loss: 1.8863
+Epoch: 009/100 | Batch 0019/0043 | Averaged Loss: 1.9282
+Epoch: 009/100 | Batch 0020/0043 | Averaged Loss: 1.9171
+Epoch: 009/100 | Batch 0021/0043 | Averaged Loss: 1.8866
+Epoch: 009/100 | Batch 0022/0043 | Averaged Loss: 1.8759
+Epoch: 009/100 | Batch 0023/0043 | Averaged Loss: 1.8467
+Epoch: 009/100 | Batch 0024/0043 | Averaged Loss: 1.8696
+Epoch: 009/100 | Batch 0025/0043 | Averaged Loss: 1.8844
+Epoch: 009/100 | Batch 0026/0043 | Averaged Loss: 1.8636
+Epoch: 009/100 | Batch 0027/0043 | Averaged Loss: 1.8843
+Epoch: 009/100 | Batch 0028/0043 | Averaged Loss: 1.8293
+Epoch: 009/100 | Batch 0029/0043 | Averaged Loss: 1.8153
+Epoch: 009/100 | Batch 0030/0043 | Averaged Loss: 1.8626
+Epoch: 009/100 | Batch 0031/0043 | Averaged Loss: 1.8337
+Epoch: 009/100 | Batch 0032/0043 | Averaged Loss: 1.8192
+Epoch: 009/100 | Batch 0033/0043 | Averaged Loss: 1.8398
+Epoch: 009/100 | Batch 0034/0043 | Averaged Loss: 1.7981
+Epoch: 009/100 | Batch 0035/0043 | Averaged Loss: 1.8320
+Epoch: 009/100 | Batch 0036/0043 | Averaged Loss: 1.8706
+Epoch: 009/100 | Batch 0037/0043 | Averaged Loss: 1.8640
+Epoch: 009/100 | Batch 0038/0043 | Averaged Loss: 1.8736
+Epoch: 009/100 | Batch 0039/0043 | Averaged Loss: 1.8115
+Epoch: 009/100 | Batch 0040/0043 | Averaged Loss: 1.8799
+Epoch: 009/100 | Batch 0041/0043 | Averaged Loss: 1.9146
+Epoch: 009/100 | Batch 0042/0043 | Averaged Loss: 1.8270
+Epoch: 009/100 | Train: 27.60% | Validation: 26.61%
+Time elapsed: 1.350715160369873 min
+Epoch: 010/100 | Batch 0000/0043 | Averaged Loss: 1.8260
+Epoch: 010/100 | Batch 0001/0043 | Averaged Loss: 1.8194
+Epoch: 010/100 | Batch 0002/0043 | Averaged Loss: 1.8264
+Epoch: 010/100 | Batch 0003/0043 | Averaged Loss: 1.8218
+Epoch: 010/100 | Batch 0004/0043 | Averaged Loss: 1.7929
+Epoch: 010/100 | Batch 0005/0043 | Averaged Loss: 1.8605
+Epoch: 010/100 | Batch 0006/0043 | Averaged Loss: 1.8384
+Epoch: 010/100 | Batch 0007/0043 | Averaged Loss: 1.9025
+Epoch: 010/100 | Batch 0008/0043 | Averaged Loss: 1.9293
+Epoch: 010/100 | Batch 0009/0043 | Averaged Loss: 1.8479
+Epoch: 010/100 | Batch 0010/0043 | Averaged Loss: 1.7909
+Epoch: 010/100 | Batch 0011/0043 | Averaged Loss: 1.9244
+Epoch: 010/100 | Batch 0012/0043 | Averaged Loss: 1.9547
+Epoch: 010/100 | Batch 0013/0043 | Averaged Loss: 1.9115
+Epoch: 010/100 | Batch 0014/0043 | Averaged Loss: 1.8280
+Epoch: 010/100 | Batch 0015/0043 | Averaged Loss: 1.8803
+Epoch: 010/100 | Batch 0016/0043 | Averaged Loss: 1.8205
+Epoch: 010/100 | Batch 0017/0043 | Averaged Loss: 1.8420
+Epoch: 010/100 | Batch 0018/0043 | Averaged Loss: 1.8231
+Epoch: 010/100 | Batch 0019/0043 | Averaged Loss: 1.7899
+Epoch: 010/100 | Batch 0020/0043 | Averaged Loss: 1.8025
+Epoch: 010/100 | Batch 0021/0043 | Averaged Loss: 1.7308
+Epoch: 010/100 | Batch 0022/0043 | Averaged Loss: 1.8012
+Epoch: 010/100 | Batch 0023/0043 | Averaged Loss: 1.8159
+Epoch: 010/100 | Batch 0024/0043 | Averaged Loss: 1.8037
+Epoch: 010/100 | Batch 0025/0043 | Averaged Loss: 1.7766
+Epoch: 010/100 | Batch 0026/0043 | Averaged Loss: 1.7517
+Epoch: 010/100 | Batch 0027/0043 | Averaged Loss: 1.7659
+Epoch: 010/100 | Batch 0028/0043 | Averaged Loss: 1.7742
+Epoch: 010/100 | Batch 0029/0043 | Averaged Loss: 1.7789
+Epoch: 010/100 | Batch 0030/0043 | Averaged Loss: 1.7023
+Epoch: 010/100 | Batch 0031/0043 | Averaged Loss: 1.7976
+Epoch: 010/100 | Batch 0032/0043 | Averaged Loss: 1.8201
+Epoch: 010/100 | Batch 0033/0043 | Averaged Loss: 1.7535
+Epoch: 010/100 | Batch 0034/0043 | Averaged Loss: 1.8058
+Epoch: 010/100 | Batch 0035/0043 | Averaged Loss: 1.7665
+Epoch: 010/100 | Batch 0036/0043 | Averaged Loss: 1.7061
+Epoch: 010/100 | Batch 0037/0043 | Averaged Loss: 1.7648
+Epoch: 010/100 | Batch 0038/0043 | Averaged Loss: 1.7667
+Epoch: 010/100 | Batch 0039/0043 | Averaged Loss: 1.7179
+Epoch: 010/100 | Batch 0040/0043 | Averaged Loss: 1.7875
+Epoch: 010/100 | Batch 0041/0043 | Averaged Loss: 1.7134
+Epoch: 010/100 | Batch 0042/0043 | Averaged Loss: 1.7701
+Epoch: 010/100 | Train: 31.80% | Validation: 31.37%
+Time elapsed: 1.4986793994903564 min
+Epoch: 011/100 | Batch 0000/0043 | Averaged Loss: 1.7827
+Epoch: 011/100 | Batch 0001/0043 | Averaged Loss: 1.7671
+Epoch: 011/100 | Batch 0002/0043 | Averaged Loss: 1.7308
+Epoch: 011/100 | Batch 0003/0043 | Averaged Loss: 1.8194
+Epoch: 011/100 | Batch 0004/0043 | Averaged Loss: 1.7649
+Epoch: 011/100 | Batch 0005/0043 | Averaged Loss: 1.7122
+Epoch: 011/100 | Batch 0006/0043 | Averaged Loss: 1.7354
+Epoch: 011/100 | Batch 0007/0043 | Averaged Loss: 1.6937
+Epoch: 011/100 | Batch 0008/0043 | Averaged Loss: 1.6843
+Epoch: 011/100 | Batch 0009/0043 | Averaged Loss: 1.6866
+Epoch: 011/100 | Batch 0010/0043 | Averaged Loss: 1.7127
+Epoch: 011/100 | Batch 0011/0043 | Averaged Loss: 1.6913
+Epoch: 011/100 | Batch 0012/0043 | Averaged Loss: 1.6470
+Epoch: 011/100 | Batch 0013/0043 | Averaged Loss: 1.6923
+Epoch: 011/100 | Batch 0014/0043 | Averaged Loss: 1.6591
+Epoch: 011/100 | Batch 0015/0043 | Averaged Loss: 1.6957
+Epoch: 011/100 | Batch 0016/0043 | Averaged Loss: 1.7599
+Epoch: 011/100 | Batch 0017/0043 | Averaged Loss: 1.6854
+Epoch: 011/100 | Batch 0018/0043 | Averaged Loss: 1.7147
+Epoch: 011/100 | Batch 0019/0043 | Averaged Loss: 1.6736
+Epoch: 011/100 | Batch 0020/0043 | Averaged Loss: 1.7226
+Epoch: 011/100 | Batch 0021/0043 | Averaged Loss: 1.6338
+Epoch: 011/100 | Batch 0022/0043 | Averaged Loss: 1.6790
+Epoch: 011/100 | Batch 0023/0043 | Averaged Loss: 1.6465
+Epoch: 011/100 | Batch 0024/0043 | Averaged Loss: 1.6959
+Epoch: 011/100 | Batch 0025/0043 | Averaged Loss: 1.7000
+Epoch: 011/100 | Batch 0026/0043 | Averaged Loss: 1.7023
+Epoch: 011/100 | Batch 0027/0043 | Averaged Loss: 1.6617
+Epoch: 011/100 | Batch 0028/0043 | Averaged Loss: 1.6549
+Epoch: 011/100 | Batch 0029/0043 | Averaged Loss: 1.6956
+Epoch: 011/100 | Batch 0030/0043 | Averaged Loss: 1.7240
+Epoch: 011/100 | Batch 0031/0043 | Averaged Loss: 1.6802
+Epoch: 011/100 | Batch 0032/0043 | Averaged Loss: 1.6875
+Epoch: 011/100 | Batch 0033/0043 | Averaged Loss: 1.6977
+Epoch: 011/100 | Batch 0034/0043 | Averaged Loss: 1.6718
+Epoch: 011/100 | Batch 0035/0043 | Averaged Loss: 1.6807
+Epoch: 011/100 | Batch 0036/0043 | Averaged Loss: 1.6723
+Epoch: 011/100 | Batch 0037/0043 | Averaged Loss: 1.6291
+Epoch: 011/100 | Batch 0038/0043 | Averaged Loss: 1.5944
+Epoch: 011/100 | Batch 0039/0043 | Averaged Loss: 1.6302
+Epoch: 011/100 | Batch 0040/0043 | Averaged Loss: 1.6552
+Epoch: 011/100 | Batch 0041/0043 | Averaged Loss: 1.5904
+Epoch: 011/100 | Batch 0042/0043 | Averaged Loss: 1.6464
+Epoch: 011/100 | Train: 35.80% | Validation: 34.20%
+Time elapsed: 1.6468377113342285 min
+Epoch: 012/100 | Batch 0000/0043 | Averaged Loss: 1.6891
+Epoch: 012/100 | Batch 0001/0043 | Averaged Loss: 1.7018
+Epoch: 012/100 | Batch 0002/0043 | Averaged Loss: 1.6279
+Epoch: 012/100 | Batch 0003/0043 | Averaged Loss: 1.6547
+Epoch: 012/100 | Batch 0004/0043 | Averaged Loss: 1.7025
+Epoch: 012/100 | Batch 0005/0043 | Averaged Loss: 1.6643
+Epoch: 012/100 | Batch 0006/0043 | Averaged Loss: 1.7378
+Epoch: 012/100 | Batch 0007/0043 | Averaged Loss: 1.6039
+Epoch: 012/100 | Batch 0008/0043 | Averaged Loss: 1.6546
+Epoch: 012/100 | Batch 0009/0043 | Averaged Loss: 1.6252
+Epoch: 012/100 | Batch 0010/0043 | Averaged Loss: 1.6375
+Epoch: 012/100 | Batch 0011/0043 | Averaged Loss: 1.6466
+Epoch: 012/100 | Batch 0012/0043 | Averaged Loss: 1.5937
+Epoch: 012/100 | Batch 0013/0043 | Averaged Loss: 1.6084
+Epoch: 012/100 | Batch 0014/0043 | Averaged Loss: 1.5582
+Epoch: 012/100 | Batch 0015/0043 | Averaged Loss: 1.5184
+Epoch: 012/100 | Batch 0016/0043 | Averaged Loss: 1.5396
+Epoch: 012/100 | Batch 0017/0043 | Averaged Loss: 1.5963
+Epoch: 012/100 | Batch 0018/0043 | Averaged Loss: 1.5348
+Epoch: 012/100 | Batch 0019/0043 | Averaged Loss: 1.6132
+Epoch: 012/100 | Batch 0020/0043 | Averaged Loss: 1.6176
+Epoch: 012/100 | Batch 0021/0043 | Averaged Loss: 1.6215
+Epoch: 012/100 | Batch 0022/0043 | Averaged Loss: 1.5994
+Epoch: 012/100 | Batch 0023/0043 | Averaged Loss: 1.6611
+Epoch: 012/100 | Batch 0024/0043 | Averaged Loss: 1.6953
+Epoch: 012/100 | Batch 0025/0043 | Averaged Loss: 1.5595
+Epoch: 012/100 | Batch 0026/0043 | Averaged Loss: 1.6756
+Epoch: 012/100 | Batch 0027/0043 | Averaged Loss: 1.5748
+Epoch: 012/100 | Batch 0028/0043 | Averaged Loss: 1.6042
+Epoch: 012/100 | Batch 0029/0043 | Averaged Loss: 1.5504
+Epoch: 012/100 | Batch 0030/0043 | Averaged Loss: 1.5255
+Epoch: 012/100 | Batch 0031/0043 | Averaged Loss: 1.5497
+Epoch: 012/100 | Batch 0032/0043 | Averaged Loss: 1.5562
+Epoch: 012/100 | Batch 0033/0043 | Averaged Loss: 1.5153
+Epoch: 012/100 | Batch 0034/0043 | Averaged Loss: 1.5760
+Epoch: 012/100 | Batch 0035/0043 | Averaged Loss: 1.6034
+Epoch: 012/100 | Batch 0036/0043 | Averaged Loss: 1.5181
+Epoch: 012/100 | Batch 0037/0043 | Averaged Loss: 1.5310
+Epoch: 012/100 | Batch 0038/0043 | Averaged Loss: 1.5612
+Epoch: 012/100 | Batch 0039/0043 | Averaged Loss: 1.5091
+Epoch: 012/100 | Batch 0040/0043 | Averaged Loss: 1.5031
+Epoch: 012/100 | Batch 0041/0043 | Averaged Loss: 1.4776
+Epoch: 012/100 | Batch 0042/0043 | Averaged Loss: 1.5509
+Epoch: 012/100 | Train: 40.81% | Validation: 41.21%
+Time elapsed: 1.7951159477233887 min
+Epoch: 013/100 | Batch 0000/0043 | Averaged Loss: 1.5019
+Epoch: 013/100 | Batch 0001/0043 | Averaged Loss: 1.5006
+Epoch: 013/100 | Batch 0002/0043 | Averaged Loss: 1.5019
+Epoch: 013/100 | Batch 0003/0043 | Averaged Loss: 1.4479
+Epoch: 013/100 | Batch 0004/0043 | Averaged Loss: 1.4811
+Epoch: 013/100 | Batch 0005/0043 | Averaged Loss: 1.5830
+Epoch: 013/100 | Batch 0006/0043 | Averaged Loss: 1.5563
+Epoch: 013/100 | Batch 0007/0043 | Averaged Loss: 1.4932
+Epoch: 013/100 | Batch 0008/0043 | Averaged Loss: 1.5633
+Epoch: 013/100 | Batch 0009/0043 | Averaged Loss: 1.5008
+Epoch: 013/100 | Batch 0010/0043 | Averaged Loss: 1.5265
+Epoch: 013/100 | Batch 0011/0043 | Averaged Loss: 1.5106
+Epoch: 013/100 | Batch 0012/0043 | Averaged Loss: 1.4470
+Epoch: 013/100 | Batch 0013/0043 | Averaged Loss: 1.4770
+Epoch: 013/100 | Batch 0014/0043 | Averaged Loss: 1.5321
+Epoch: 013/100 | Batch 0015/0043 | Averaged Loss: 1.5232
+Epoch: 013/100 | Batch 0016/0043 | Averaged Loss: 1.5311
+Epoch: 013/100 | Batch 0017/0043 | Averaged Loss: 1.4668
+Epoch: 013/100 | Batch 0018/0043 | Averaged Loss: 1.4935
+Epoch: 013/100 | Batch 0019/0043 | Averaged Loss: 1.5060
+Epoch: 013/100 | Batch 0020/0043 | Averaged Loss: 1.4847
+Epoch: 013/100 | Batch 0021/0043 | Averaged Loss: 1.4529
+Epoch: 013/100 | Batch 0022/0043 | Averaged Loss: 1.5070
+Epoch: 013/100 | Batch 0023/0043 | Averaged Loss: 1.4991
+Epoch: 013/100 | Batch 0024/0043 | Averaged Loss: 1.4769
+Epoch: 013/100 | Batch 0025/0043 | Averaged Loss: 1.4878
+Epoch: 013/100 | Batch 0026/0043 | Averaged Loss: 1.4236
+Epoch: 013/100 | Batch 0027/0043 | Averaged Loss: 1.5525
+Epoch: 013/100 | Batch 0028/0043 | Averaged Loss: 1.4935
+Epoch: 013/100 | Batch 0029/0043 | Averaged Loss: 1.4295
+Epoch: 013/100 | Batch 0030/0043 | Averaged Loss: 1.4374
+Epoch: 013/100 | Batch 0031/0043 | Averaged Loss: 1.4539
+Epoch: 013/100 | Batch 0032/0043 | Averaged Loss: 1.4389
+Epoch: 013/100 | Batch 0033/0043 | Averaged Loss: 1.4655
+Epoch: 013/100 | Batch 0034/0043 | Averaged Loss: 1.4321
+Epoch: 013/100 | Batch 0035/0043 | Averaged Loss: 1.5267
+Epoch: 013/100 | Batch 0036/0043 | Averaged Loss: 1.4349
+Epoch: 013/100 | Batch 0037/0043 | Averaged Loss: 1.4139
+Epoch: 013/100 | Batch 0038/0043 | Averaged Loss: 1.4086
+Epoch: 013/100 | Batch 0039/0043 | Averaged Loss: 1.4136
+Epoch: 013/100 | Batch 0040/0043 | Averaged Loss: 1.4061
+Epoch: 013/100 | Batch 0041/0043 | Averaged Loss: 1.4754
+Epoch: 013/100 | Batch 0042/0043 | Averaged Loss: 1.4591
+Epoch: 013/100 | Train: 46.82% | Validation: 46.68%
+Time elapsed: 1.9434208869934082 min
+Epoch: 014/100 | Batch 0000/0043 | Averaged Loss: 1.3885
+Epoch: 014/100 | Batch 0001/0043 | Averaged Loss: 1.4237
+Epoch: 014/100 | Batch 0002/0043 | Averaged Loss: 1.4461
+Epoch: 014/100 | Batch 0003/0043 | Averaged Loss: 1.4312
+Epoch: 014/100 | Batch 0004/0043 | Averaged Loss: 1.4353
+Epoch: 014/100 | Batch 0005/0043 | Averaged Loss: 1.4400
+Epoch: 014/100 | Batch 0006/0043 | Averaged Loss: 1.4692
+Epoch: 014/100 | Batch 0007/0043 | Averaged Loss: 1.4564
+Epoch: 014/100 | Batch 0008/0043 | Averaged Loss: 1.5626
+Epoch: 014/100 | Batch 0009/0043 | Averaged Loss: 1.4211
+Epoch: 014/100 | Batch 0010/0043 | Averaged Loss: 1.5957
+Epoch: 014/100 | Batch 0011/0043 | Averaged Loss: 1.5399
+Epoch: 014/100 | Batch 0012/0043 | Averaged Loss: 1.4904
+Epoch: 014/100 | Batch 0013/0043 | Averaged Loss: 1.5451
+Epoch: 014/100 | Batch 0014/0043 | Averaged Loss: 1.5073
+Epoch: 014/100 | Batch 0015/0043 | Averaged Loss: 1.4363
+Epoch: 014/100 | Batch 0016/0043 | Averaged Loss: 1.4266
+Epoch: 014/100 | Batch 0017/0043 | Averaged Loss: 1.4527
+Epoch: 014/100 | Batch 0018/0043 | Averaged Loss: 1.4443
+Epoch: 014/100 | Batch 0019/0043 | Averaged Loss: 1.4718
+Epoch: 014/100 | Batch 0020/0043 | Averaged Loss: 1.4243
+Epoch: 014/100 | Batch 0021/0043 | Averaged Loss: 1.4060
+Epoch: 014/100 | Batch 0022/0043 | Averaged Loss: 1.4475
+Epoch: 014/100 | Batch 0023/0043 | Averaged Loss: 1.4001
+Epoch: 014/100 | Batch 0024/0043 | Averaged Loss: 1.3764
+Epoch: 014/100 | Batch 0025/0043 | Averaged Loss: 1.4353
+Epoch: 014/100 | Batch 0026/0043 | Averaged Loss: 1.4671
+Epoch: 014/100 | Batch 0027/0043 | Averaged Loss: 1.4276
+Epoch: 014/100 | Batch 0028/0043 | Averaged Loss: 1.4123
+Epoch: 014/100 | Batch 0029/0043 | Averaged Loss: 1.3857
+Epoch: 014/100 | Batch 0030/0043 | Averaged Loss: 1.3051
+Epoch: 014/100 | Batch 0031/0043 | Averaged Loss: 1.4229
+Epoch: 014/100 | Batch 0032/0043 | Averaged Loss: 1.4282
+Epoch: 014/100 | Batch 0033/0043 | Averaged Loss: 1.3612
+Epoch: 014/100 | Batch 0034/0043 | Averaged Loss: 1.4391
+Epoch: 014/100 | Batch 0035/0043 | Averaged Loss: 1.3486
+Epoch: 014/100 | Batch 0036/0043 | Averaged Loss: 1.3622
+Epoch: 014/100 | Batch 0037/0043 | Averaged Loss: 1.3601
+Epoch: 014/100 | Batch 0038/0043 | Averaged Loss: 1.4568
+Epoch: 014/100 | Batch 0039/0043 | Averaged Loss: 1.4918
+Epoch: 014/100 | Batch 0040/0043 | Averaged Loss: 1.4155
+Epoch: 014/100 | Batch 0041/0043 | Averaged Loss: 1.4148
+Epoch: 014/100 | Batch 0042/0043 | Averaged Loss: 1.4094
+Epoch: 014/100 | Train: 50.25% | Validation: 50.93%
+Time elapsed: 2.091334104537964 min
+Epoch: 015/100 | Batch 0000/0043 | Averaged Loss: 1.3984
+Epoch: 015/100 | Batch 0001/0043 | Averaged Loss: 1.3317
+Epoch: 015/100 | Batch 0002/0043 | Averaged Loss: 1.4048
+Epoch: 015/100 | Batch 0003/0043 | Averaged Loss: 1.3065
+Epoch: 015/100 | Batch 0004/0043 | Averaged Loss: 1.3430
+Epoch: 015/100 | Batch 0005/0043 | Averaged Loss: 1.3542
+Epoch: 015/100 | Batch 0006/0043 | Averaged Loss: 1.2684
+Epoch: 015/100 | Batch 0007/0043 | Averaged Loss: 1.2889
+Epoch: 015/100 | Batch 0008/0043 | Averaged Loss: 1.3963
+Epoch: 015/100 | Batch 0009/0043 | Averaged Loss: 1.3172
+Epoch: 015/100 | Batch 0010/0043 | Averaged Loss: 1.3943
+Epoch: 015/100 | Batch 0011/0043 | Averaged Loss: 1.3039
+Epoch: 015/100 | Batch 0012/0043 | Averaged Loss: 1.3553
+Epoch: 015/100 | Batch 0013/0043 | Averaged Loss: 1.3360
+Epoch: 015/100 | Batch 0014/0043 | Averaged Loss: 1.3124
+Epoch: 015/100 | Batch 0015/0043 | Averaged Loss: 1.2946
+Epoch: 015/100 | Batch 0016/0043 | Averaged Loss: 1.2983
+Epoch: 015/100 | Batch 0017/0043 | Averaged Loss: 1.2696
+Epoch: 015/100 | Batch 0018/0043 | Averaged Loss: 1.2549
+Epoch: 015/100 | Batch 0019/0043 | Averaged Loss: 1.2803
+Epoch: 015/100 | Batch 0020/0043 | Averaged Loss: 1.3209
+Epoch: 015/100 | Batch 0021/0043 | Averaged Loss: 1.2952
+Epoch: 015/100 | Batch 0022/0043 | Averaged Loss: 1.3788
+Epoch: 015/100 | Batch 0023/0043 | Averaged Loss: 1.3899
+Epoch: 015/100 | Batch 0024/0043 | Averaged Loss: 1.3495
+Epoch: 015/100 | Batch 0025/0043 | Averaged Loss: 1.3666
+Epoch: 015/100 | Batch 0026/0043 | Averaged Loss: 1.3818
+Epoch: 015/100 | Batch 0027/0043 | Averaged Loss: 1.3758
+Epoch: 015/100 | Batch 0028/0043 | Averaged Loss: 1.3323
+Epoch: 015/100 | Batch 0029/0043 | Averaged Loss: 1.3847
+Epoch: 015/100 | Batch 0030/0043 | Averaged Loss: 1.3673
+Epoch: 015/100 | Batch 0031/0043 | Averaged Loss: 1.3834
+Epoch: 015/100 | Batch 0032/0043 | Averaged Loss: 1.3465
+Epoch: 015/100 | Batch 0033/0043 | Averaged Loss: 1.3252
+Epoch: 015/100 | Batch 0034/0043 | Averaged Loss: 1.2852
+Epoch: 015/100 | Batch 0035/0043 | Averaged Loss: 1.2727
+Epoch: 015/100 | Batch 0036/0043 | Averaged Loss: 1.2855
+Epoch: 015/100 | Batch 0037/0043 | Averaged Loss: 1.2512
+Epoch: 015/100 | Batch 0038/0043 | Averaged Loss: 1.2585
+Epoch: 015/100 | Batch 0039/0043 | Averaged Loss: 1.3271
+Epoch: 015/100 | Batch 0040/0043 | Averaged Loss: 1.2616
+Epoch: 015/100 | Batch 0041/0043 | Averaged Loss: 1.2979
+Epoch: 015/100 | Batch 0042/0043 | Averaged Loss: 1.2454
+Epoch: 015/100 | Train: 53.33% | Validation: 53.91%
+Time elapsed: 2.23929500579834 min
+Epoch: 016/100 | Batch 0000/0043 | Averaged Loss: 1.2755
+Epoch: 016/100 | Batch 0001/0043 | Averaged Loss: 1.2360
+Epoch: 016/100 | Batch 0002/0043 | Averaged Loss: 1.1675
+Epoch: 016/100 | Batch 0003/0043 | Averaged Loss: 1.3003
+Epoch: 016/100 | Batch 0004/0043 | Averaged Loss: 1.2597
+Epoch: 016/100 | Batch 0005/0043 | Averaged Loss: 1.2389
+Epoch: 016/100 | Batch 0006/0043 | Averaged Loss: 1.2367
+Epoch: 016/100 | Batch 0007/0043 | Averaged Loss: 1.3365
+Epoch: 016/100 | Batch 0008/0043 | Averaged Loss: 1.2969
+Epoch: 016/100 | Batch 0009/0043 | Averaged Loss: 1.2593
+Epoch: 016/100 | Batch 0010/0043 | Averaged Loss: 1.2737
+Epoch: 016/100 | Batch 0011/0043 | Averaged Loss: 1.2654
+Epoch: 016/100 | Batch 0012/0043 | Averaged Loss: 1.1929
+Epoch: 016/100 | Batch 0013/0043 | Averaged Loss: 1.1865
+Epoch: 016/100 | Batch 0014/0043 | Averaged Loss: 1.3070
+Epoch: 016/100 | Batch 0015/0043 | Averaged Loss: 1.2503
+Epoch: 016/100 | Batch 0016/0043 | Averaged Loss: 1.2495
+Epoch: 016/100 | Batch 0017/0043 | Averaged Loss: 1.2079
+Epoch: 016/100 | Batch 0018/0043 | Averaged Loss: 1.1679
+Epoch: 016/100 | Batch 0019/0043 | Averaged Loss: 1.2463
+Epoch: 016/100 | Batch 0020/0043 | Averaged Loss: 1.2139
+Epoch: 016/100 | Batch 0021/0043 | Averaged Loss: 1.2398
+Epoch: 016/100 | Batch 0022/0043 | Averaged Loss: 1.2940
+Epoch: 016/100 | Batch 0023/0043 | Averaged Loss: 1.1663
+Epoch: 016/100 | Batch 0024/0043 | Averaged Loss: 1.2019
+Epoch: 016/100 | Batch 0025/0043 | Averaged Loss: 1.2525
+Epoch: 016/100 | Batch 0026/0043 | Averaged Loss: 1.1444
+Epoch: 016/100 | Batch 0027/0043 | Averaged Loss: 1.2248
+Epoch: 016/100 | Batch 0028/0043 | Averaged Loss: 1.2360
+Epoch: 016/100 | Batch 0029/0043 | Averaged Loss: 1.2075
+Epoch: 016/100 | Batch 0030/0043 | Averaged Loss: 1.2378
+Epoch: 016/100 | Batch 0031/0043 | Averaged Loss: 1.2499
+Epoch: 016/100 | Batch 0032/0043 | Averaged Loss: 1.2860
+Epoch: 016/100 | Batch 0033/0043 | Averaged Loss: 1.3099
+Epoch: 016/100 | Batch 0034/0043 | Averaged Loss: 1.3133
+Epoch: 016/100 | Batch 0035/0043 | Averaged Loss: 1.2353
+Epoch: 016/100 | Batch 0036/0043 | Averaged Loss: 1.3471
+Epoch: 016/100 | Batch 0037/0043 | Averaged Loss: 1.2985
+Epoch: 016/100 | Batch 0038/0043 | Averaged Loss: 1.2902
+Epoch: 016/100 | Batch 0039/0043 | Averaged Loss: 1.2379
+Epoch: 016/100 | Batch 0040/0043 | Averaged Loss: 1.1803
+Epoch: 016/100 | Batch 0041/0043 | Averaged Loss: 1.2872
+Epoch: 016/100 | Batch 0042/0043 | Averaged Loss: 1.1555
+Epoch: 016/100 | Train: 56.63% | Validation: 56.57%
+Time elapsed: 2.387202739715576 min
+Epoch: 017/100 | Batch 0000/0043 | Averaged Loss: 1.1844
+Epoch: 017/100 | Batch 0001/0043 | Averaged Loss: 1.2990
+Epoch: 017/100 | Batch 0002/0043 | Averaged Loss: 1.2281
+Epoch: 017/100 | Batch 0003/0043 | Averaged Loss: 1.1415
+Epoch: 017/100 | Batch 0004/0043 | Averaged Loss: 1.2781
+Epoch: 017/100 | Batch 0005/0043 | Averaged Loss: 1.2331
+Epoch: 017/100 | Batch 0006/0043 | Averaged Loss: 1.2061
+Epoch: 017/100 | Batch 0007/0043 | Averaged Loss: 1.2210
+Epoch: 017/100 | Batch 0008/0043 | Averaged Loss: 1.1589
+Epoch: 017/100 | Batch 0009/0043 | Averaged Loss: 1.2043
+Epoch: 017/100 | Batch 0010/0043 | Averaged Loss: 1.2336
+Epoch: 017/100 | Batch 0011/0043 | Averaged Loss: 1.1533
+Epoch: 017/100 | Batch 0012/0043 | Averaged Loss: 1.1512
+Epoch: 017/100 | Batch 0013/0043 | Averaged Loss: 1.1244
+Epoch: 017/100 | Batch 0014/0043 | Averaged Loss: 1.1863
+Epoch: 017/100 | Batch 0015/0043 | Averaged Loss: 1.2103
+Epoch: 017/100 | Batch 0016/0043 | Averaged Loss: 1.1674
+Epoch: 017/100 | Batch 0017/0043 | Averaged Loss: 1.1688
+Epoch: 017/100 | Batch 0018/0043 | Averaged Loss: 1.1339
+Epoch: 017/100 | Batch 0019/0043 | Averaged Loss: 1.1645
+Epoch: 017/100 | Batch 0020/0043 | Averaged Loss: 1.1698
+Epoch: 017/100 | Batch 0021/0043 | Averaged Loss: 1.1892
+Epoch: 017/100 | Batch 0022/0043 | Averaged Loss: 1.1202
+Epoch: 017/100 | Batch 0023/0043 | Averaged Loss: 1.0956
+Epoch: 017/100 | Batch 0024/0043 | Averaged Loss: 1.1147
+Epoch: 017/100 | Batch 0025/0043 | Averaged Loss: 1.1696
+Epoch: 017/100 | Batch 0026/0043 | Averaged Loss: 1.1443
+Epoch: 017/100 | Batch 0027/0043 | Averaged Loss: 1.0666
+Epoch: 017/100 | Batch 0028/0043 | Averaged Loss: 1.1182
+Epoch: 017/100 | Batch 0029/0043 | Averaged Loss: 1.1898
+Epoch: 017/100 | Batch 0030/0043 | Averaged Loss: 1.1652
+Epoch: 017/100 | Batch 0031/0043 | Averaged Loss: 1.2868
+Epoch: 017/100 | Batch 0032/0043 | Averaged Loss: 1.3278
+Epoch: 017/100 | Batch 0033/0043 | Averaged Loss: 1.2949
+Epoch: 017/100 | Batch 0034/0043 | Averaged Loss: 1.3118
+Epoch: 017/100 | Batch 0035/0043 | Averaged Loss: 1.3424
+Epoch: 017/100 | Batch 0036/0043 | Averaged Loss: 1.2666
+Epoch: 017/100 | Batch 0037/0043 | Averaged Loss: 1.1985
+Epoch: 017/100 | Batch 0038/0043 | Averaged Loss: 1.3131
+Epoch: 017/100 | Batch 0039/0043 | Averaged Loss: 1.3381
+Epoch: 017/100 | Batch 0040/0043 | Averaged Loss: 1.2632
+Epoch: 017/100 | Batch 0041/0043 | Averaged Loss: 1.2775
+Epoch: 017/100 | Batch 0042/0043 | Averaged Loss: 1.1668
+Epoch: 017/100 | Train: 59.54% | Validation: 59.23%
+Time elapsed: 2.5353550910949707 min
+Epoch: 018/100 | Batch 0000/0043 | Averaged Loss: 1.1593
+Epoch: 018/100 | Batch 0001/0043 | Averaged Loss: 1.1198
+Epoch: 018/100 | Batch 0002/0043 | Averaged Loss: 1.1597
+Epoch: 018/100 | Batch 0003/0043 | Averaged Loss: 1.1918
+Epoch: 018/100 | Batch 0004/0043 | Averaged Loss: 1.2301
+Epoch: 018/100 | Batch 0005/0043 | Averaged Loss: 1.0621
+Epoch: 018/100 | Batch 0006/0043 | Averaged Loss: 1.2444
+Epoch: 018/100 | Batch 0007/0043 | Averaged Loss: 1.1190
+Epoch: 018/100 | Batch 0008/0043 | Averaged Loss: 1.1656
+Epoch: 018/100 | Batch 0009/0043 | Averaged Loss: 1.1043
+Epoch: 018/100 | Batch 0010/0043 | Averaged Loss: 1.1476
+Epoch: 018/100 | Batch 0011/0043 | Averaged Loss: 1.1865
+Epoch: 018/100 | Batch 0012/0043 | Averaged Loss: 1.0954
+Epoch: 018/100 | Batch 0013/0043 | Averaged Loss: 1.1729
+Epoch: 018/100 | Batch 0014/0043 | Averaged Loss: 1.1718
+Epoch: 018/100 | Batch 0015/0043 | Averaged Loss: 1.1844
+Epoch: 018/100 | Batch 0016/0043 | Averaged Loss: 1.1116
+Epoch: 018/100 | Batch 0017/0043 | Averaged Loss: 1.2656
+Epoch: 018/100 | Batch 0018/0043 | Averaged Loss: 1.1640
+Epoch: 018/100 | Batch 0019/0043 | Averaged Loss: 1.2473
+Epoch: 018/100 | Batch 0020/0043 | Averaged Loss: 1.1639
+Epoch: 018/100 | Batch 0021/0043 | Averaged Loss: 1.1216
+Epoch: 018/100 | Batch 0022/0043 | Averaged Loss: 1.1881
+Epoch: 018/100 | Batch 0023/0043 | Averaged Loss: 1.1952
+Epoch: 018/100 | Batch 0024/0043 | Averaged Loss: 1.1544
+Epoch: 018/100 | Batch 0025/0043 | Averaged Loss: 1.2064
+Epoch: 018/100 | Batch 0026/0043 | Averaged Loss: 1.1471
+Epoch: 018/100 | Batch 0027/0043 | Averaged Loss: 1.1738
+Epoch: 018/100 | Batch 0028/0043 | Averaged Loss: 1.1478
+Epoch: 018/100 | Batch 0029/0043 | Averaged Loss: 1.0991
+Epoch: 018/100 | Batch 0030/0043 | Averaged Loss: 1.1968
+Epoch: 018/100 | Batch 0031/0043 | Averaged Loss: 1.1595
+Epoch: 018/100 | Batch 0032/0043 | Averaged Loss: 1.1708
+Epoch: 018/100 | Batch 0033/0043 | Averaged Loss: 1.1714
+Epoch: 018/100 | Batch 0034/0043 | Averaged Loss: 1.1266
+Epoch: 018/100 | Batch 0035/0043 | Averaged Loss: 1.0562
+Epoch: 018/100 | Batch 0036/0043 | Averaged Loss: 1.1644
+Epoch: 018/100 | Batch 0037/0043 | Averaged Loss: 1.1287
+Epoch: 018/100 | Batch 0038/0043 | Averaged Loss: 1.1971
+Epoch: 018/100 | Batch 0039/0043 | Averaged Loss: 1.0698
+Epoch: 018/100 | Batch 0040/0043 | Averaged Loss: 1.1764
+Epoch: 018/100 | Batch 0041/0043 | Averaged Loss: 1.0887
+Epoch: 018/100 | Batch 0042/0043 | Averaged Loss: 1.2024
+Epoch: 018/100 | Train: 62.89% | Validation: 63.23%
+Time elapsed: 2.683157205581665 min
+Epoch: 019/100 | Batch 0000/0043 | Averaged Loss: 1.1180
+Epoch: 019/100 | Batch 0001/0043 | Averaged Loss: 1.0738
+Epoch: 019/100 | Batch 0002/0043 | Averaged Loss: 1.0003
+Epoch: 019/100 | Batch 0003/0043 | Averaged Loss: 1.0827
+Epoch: 019/100 | Batch 0004/0043 | Averaged Loss: 1.0269
+Epoch: 019/100 | Batch 0005/0043 | Averaged Loss: 1.0592
+Epoch: 019/100 | Batch 0006/0043 | Averaged Loss: 1.0106
+Epoch: 019/100 | Batch 0007/0043 | Averaged Loss: 1.0858
+Epoch: 019/100 | Batch 0008/0043 | Averaged Loss: 0.9705
+Epoch: 019/100 | Batch 0009/0043 | Averaged Loss: 1.0402
+Epoch: 019/100 | Batch 0010/0043 | Averaged Loss: 1.0106
+Epoch: 019/100 | Batch 0011/0043 | Averaged Loss: 0.9966
+Epoch: 019/100 | Batch 0012/0043 | Averaged Loss: 1.0512
+Epoch: 019/100 | Batch 0013/0043 | Averaged Loss: 1.0457
+Epoch: 019/100 | Batch 0014/0043 | Averaged Loss: 1.0698
+Epoch: 019/100 | Batch 0015/0043 | Averaged Loss: 1.0710
+Epoch: 019/100 | Batch 0016/0043 | Averaged Loss: 1.0839
+Epoch: 019/100 | Batch 0017/0043 | Averaged Loss: 1.0494
+Epoch: 019/100 | Batch 0018/0043 | Averaged Loss: 1.1020
+Epoch: 019/100 | Batch 0019/0043 | Averaged Loss: 1.1239
+Epoch: 019/100 | Batch 0020/0043 | Averaged Loss: 1.0191
+Epoch: 019/100 | Batch 0021/0043 | Averaged Loss: 1.0660
+Epoch: 019/100 | Batch 0022/0043 | Averaged Loss: 1.1044
+Epoch: 019/100 | Batch 0023/0043 | Averaged Loss: 1.0236
+Epoch: 019/100 | Batch 0024/0043 | Averaged Loss: 1.0720
+Epoch: 019/100 | Batch 0025/0043 | Averaged Loss: 1.1243
+Epoch: 019/100 | Batch 0026/0043 | Averaged Loss: 1.0264
+Epoch: 019/100 | Batch 0027/0043 | Averaged Loss: 1.0383
+Epoch: 019/100 | Batch 0028/0043 | Averaged Loss: 1.0229
+Epoch: 019/100 | Batch 0029/0043 | Averaged Loss: 1.0800
+Epoch: 019/100 | Batch 0030/0043 | Averaged Loss: 1.1080
+Epoch: 019/100 | Batch 0031/0043 | Averaged Loss: 1.0793
+Epoch: 019/100 | Batch 0032/0043 | Averaged Loss: 1.1637
+Epoch: 019/100 | Batch 0033/0043 | Averaged Loss: 1.0885
+Epoch: 019/100 | Batch 0034/0043 | Averaged Loss: 1.1619
+Epoch: 019/100 | Batch 0035/0043 | Averaged Loss: 1.0817
+Epoch: 019/100 | Batch 0036/0043 | Averaged Loss: 1.1280
+Epoch: 019/100 | Batch 0037/0043 | Averaged Loss: 1.1293
+Epoch: 019/100 | Batch 0038/0043 | Averaged Loss: 1.0065
+Epoch: 019/100 | Batch 0039/0043 | Averaged Loss: 1.1436
+Epoch: 019/100 | Batch 0040/0043 | Averaged Loss: 1.0714
+Epoch: 019/100 | Batch 0041/0043 | Averaged Loss: 1.0498
+Epoch: 019/100 | Batch 0042/0043 | Averaged Loss: 0.9821
+Epoch: 019/100 | Train: 65.18% | Validation: 64.43%
+Time elapsed: 2.831535577774048 min
+Epoch: 020/100 | Batch 0000/0043 | Averaged Loss: 1.0332
+Epoch: 020/100 | Batch 0001/0043 | Averaged Loss: 0.9778
+Epoch: 020/100 | Batch 0002/0043 | Averaged Loss: 0.9774
+Epoch: 020/100 | Batch 0003/0043 | Averaged Loss: 1.0565
+Epoch: 020/100 | Batch 0004/0043 | Averaged Loss: 0.9749
+Epoch: 020/100 | Batch 0005/0043 | Averaged Loss: 0.9771
+Epoch: 020/100 | Batch 0006/0043 | Averaged Loss: 1.0518
+Epoch: 020/100 | Batch 0007/0043 | Averaged Loss: 1.0306
+Epoch: 020/100 | Batch 0008/0043 | Averaged Loss: 1.0361
+Epoch: 020/100 | Batch 0009/0043 | Averaged Loss: 1.0450
+Epoch: 020/100 | Batch 0010/0043 | Averaged Loss: 0.9526
+Epoch: 020/100 | Batch 0011/0043 | Averaged Loss: 1.0929
+Epoch: 020/100 | Batch 0012/0043 | Averaged Loss: 1.0391
+Epoch: 020/100 | Batch 0013/0043 | Averaged Loss: 1.1110
+Epoch: 020/100 | Batch 0014/0043 | Averaged Loss: 0.9865
+Epoch: 020/100 | Batch 0015/0043 | Averaged Loss: 1.0747
+Epoch: 020/100 | Batch 0016/0043 | Averaged Loss: 1.0359
+Epoch: 020/100 | Batch 0017/0043 | Averaged Loss: 1.0600
+Epoch: 020/100 | Batch 0018/0043 | Averaged Loss: 0.9902
+Epoch: 020/100 | Batch 0019/0043 | Averaged Loss: 1.0023
+Epoch: 020/100 | Batch 0020/0043 | Averaged Loss: 0.9953
+Epoch: 020/100 | Batch 0021/0043 | Averaged Loss: 1.0543
+Epoch: 020/100 | Batch 0022/0043 | Averaged Loss: 0.9509
+Epoch: 020/100 | Batch 0023/0043 | Averaged Loss: 1.0502
+Epoch: 020/100 | Batch 0024/0043 | Averaged Loss: 0.9854
+Epoch: 020/100 | Batch 0025/0043 | Averaged Loss: 0.9187
+Epoch: 020/100 | Batch 0026/0043 | Averaged Loss: 0.9772
+Epoch: 020/100 | Batch 0027/0043 | Averaged Loss: 0.9923
+Epoch: 020/100 | Batch 0028/0043 | Averaged Loss: 1.0108
+Epoch: 020/100 | Batch 0029/0043 | Averaged Loss: 0.9008
+Epoch: 020/100 | Batch 0030/0043 | Averaged Loss: 0.9613
+Epoch: 020/100 | Batch 0031/0043 | Averaged Loss: 0.9331
+Epoch: 020/100 | Batch 0032/0043 | Averaged Loss: 1.0030
+Epoch: 020/100 | Batch 0033/0043 | Averaged Loss: 1.0492
+Epoch: 020/100 | Batch 0034/0043 | Averaged Loss: 0.9250
+Epoch: 020/100 | Batch 0035/0043 | Averaged Loss: 0.9831
+Epoch: 020/100 | Batch 0036/0043 | Averaged Loss: 0.9275
+Epoch: 020/100 | Batch 0037/0043 | Averaged Loss: 0.9767
+Epoch: 020/100 | Batch 0038/0043 | Averaged Loss: 0.9524
+Epoch: 020/100 | Batch 0039/0043 | Averaged Loss: 0.9162
+Epoch: 020/100 | Batch 0040/0043 | Averaged Loss: 0.9379
+Epoch: 020/100 | Batch 0041/0043 | Averaged Loss: 1.0023
+Epoch: 020/100 | Batch 0042/0043 | Averaged Loss: 0.9166
+Epoch: 020/100 | Train: 69.90% | Validation: 68.33%
+Time elapsed: 2.9799304008483887 min
+Epoch: 021/100 | Batch 0000/0043 | Averaged Loss: 0.8291
+Epoch: 021/100 | Batch 0001/0043 | Averaged Loss: 0.9202
+Epoch: 021/100 | Batch 0002/0043 | Averaged Loss: 0.9129
+Epoch: 021/100 | Batch 0003/0043 | Averaged Loss: 0.9448
+Epoch: 021/100 | Batch 0004/0043 | Averaged Loss: 0.9334
+Epoch: 021/100 | Batch 0005/0043 | Averaged Loss: 0.9240
+Epoch: 021/100 | Batch 0006/0043 | Averaged Loss: 0.8719
+Epoch: 021/100 | Batch 0007/0043 | Averaged Loss: 0.9118
+Epoch: 021/100 | Batch 0008/0043 | Averaged Loss: 0.9263
+Epoch: 021/100 | Batch 0009/0043 | Averaged Loss: 0.8966
+Epoch: 021/100 | Batch 0010/0043 | Averaged Loss: 0.8823
+Epoch: 021/100 | Batch 0011/0043 | Averaged Loss: 0.9947
+Epoch: 021/100 | Batch 0012/0043 | Averaged Loss: 0.9713
+Epoch: 021/100 | Batch 0013/0043 | Averaged Loss: 0.9969
+Epoch: 021/100 | Batch 0014/0043 | Averaged Loss: 0.9222
+Epoch: 021/100 | Batch 0015/0043 | Averaged Loss: 1.0652
+Epoch: 021/100 | Batch 0016/0043 | Averaged Loss: 0.9711
+Epoch: 021/100 | Batch 0017/0043 | Averaged Loss: 1.0621
+Epoch: 021/100 | Batch 0018/0043 | Averaged Loss: 0.9414
+Epoch: 021/100 | Batch 0019/0043 | Averaged Loss: 0.9429
+Epoch: 021/100 | Batch 0020/0043 | Averaged Loss: 0.9586
+Epoch: 021/100 | Batch 0021/0043 | Averaged Loss: 0.8889
+Epoch: 021/100 | Batch 0022/0043 | Averaged Loss: 0.9842
+Epoch: 021/100 | Batch 0023/0043 | Averaged Loss: 0.9320
+Epoch: 021/100 | Batch 0024/0043 | Averaged Loss: 1.0526
+Epoch: 021/100 | Batch 0025/0043 | Averaged Loss: 1.0393
+Epoch: 021/100 | Batch 0026/0043 | Averaged Loss: 0.9840
+Epoch: 021/100 | Batch 0027/0043 | Averaged Loss: 0.9807
+Epoch: 021/100 | Batch 0028/0043 | Averaged Loss: 0.9549
+Epoch: 021/100 | Batch 0029/0043 | Averaged Loss: 0.9376
+Epoch: 021/100 | Batch 0030/0043 | Averaged Loss: 0.9526
+Epoch: 021/100 | Batch 0031/0043 | Averaged Loss: 0.9143
+Epoch: 021/100 | Batch 0032/0043 | Averaged Loss: 0.9587
+Epoch: 021/100 | Batch 0033/0043 | Averaged Loss: 0.9608
+Epoch: 021/100 | Batch 0034/0043 | Averaged Loss: 1.0046
+Epoch: 021/100 | Batch 0035/0043 | Averaged Loss: 0.8997
+Epoch: 021/100 | Batch 0036/0043 | Averaged Loss: 0.9489
+Epoch: 021/100 | Batch 0037/0043 | Averaged Loss: 0.8459
+Epoch: 021/100 | Batch 0038/0043 | Averaged Loss: 0.9968
+Epoch: 021/100 | Batch 0039/0043 | Averaged Loss: 0.9225
+Epoch: 021/100 | Batch 0040/0043 | Averaged Loss: 0.9801
+Epoch: 021/100 | Batch 0041/0043 | Averaged Loss: 0.9117
+Epoch: 021/100 | Batch 0042/0043 | Averaged Loss: 0.9427
+Epoch: 021/100 | Train: 70.14% | Validation: 68.41%
+Time elapsed: 3.128114938735962 min
+Epoch: 022/100 | Batch 0000/0043 | Averaged Loss: 0.8808
+Epoch: 022/100 | Batch 0001/0043 | Averaged Loss: 0.8109
+Epoch: 022/100 | Batch 0002/0043 | Averaged Loss: 0.8905
+Epoch: 022/100 | Batch 0003/0043 | Averaged Loss: 0.8633
+Epoch: 022/100 | Batch 0004/0043 | Averaged Loss: 0.9087
+Epoch: 022/100 | Batch 0005/0043 | Averaged Loss: 0.8402
+Epoch: 022/100 | Batch 0006/0043 | Averaged Loss: 0.8552
+Epoch: 022/100 | Batch 0007/0043 | Averaged Loss: 0.8464
+Epoch: 022/100 | Batch 0008/0043 | Averaged Loss: 0.8262
+Epoch: 022/100 | Batch 0009/0043 | Averaged Loss: 0.8569
+Epoch: 022/100 | Batch 0010/0043 | Averaged Loss: 0.8599
+Epoch: 022/100 | Batch 0011/0043 | Averaged Loss: 0.8307
+Epoch: 022/100 | Batch 0012/0043 | Averaged Loss: 0.8451
+Epoch: 022/100 | Batch 0013/0043 | Averaged Loss: 0.8500
+Epoch: 022/100 | Batch 0014/0043 | Averaged Loss: 0.9144
+Epoch: 022/100 | Batch 0015/0043 | Averaged Loss: 0.8652
+Epoch: 022/100 | Batch 0016/0043 | Averaged Loss: 0.9052
+Epoch: 022/100 | Batch 0017/0043 | Averaged Loss: 0.8867
+Epoch: 022/100 | Batch 0018/0043 | Averaged Loss: 0.8432
+Epoch: 022/100 | Batch 0019/0043 | Averaged Loss: 1.0047
+Epoch: 022/100 | Batch 0020/0043 | Averaged Loss: 0.9010
+Epoch: 022/100 | Batch 0021/0043 | Averaged Loss: 0.9602
+Epoch: 022/100 | Batch 0022/0043 | Averaged Loss: 0.9392
+Epoch: 022/100 | Batch 0023/0043 | Averaged Loss: 0.8842
+Epoch: 022/100 | Batch 0024/0043 | Averaged Loss: 0.8756
+Epoch: 022/100 | Batch 0025/0043 | Averaged Loss: 0.9205
+Epoch: 022/100 | Batch 0026/0043 | Averaged Loss: 0.8779
+Epoch: 022/100 | Batch 0027/0043 | Averaged Loss: 0.8682
+Epoch: 022/100 | Batch 0028/0043 | Averaged Loss: 0.8529
+Epoch: 022/100 | Batch 0029/0043 | Averaged Loss: 0.8526
+Epoch: 022/100 | Batch 0030/0043 | Averaged Loss: 0.9737
+Epoch: 022/100 | Batch 0031/0043 | Averaged Loss: 0.8433
+Epoch: 022/100 | Batch 0032/0043 | Averaged Loss: 0.8750
+Epoch: 022/100 | Batch 0033/0043 | Averaged Loss: 0.8892
+Epoch: 022/100 | Batch 0034/0043 | Averaged Loss: 0.9476
+Epoch: 022/100 | Batch 0035/0043 | Averaged Loss: 0.8692
+Epoch: 022/100 | Batch 0036/0043 | Averaged Loss: 0.9002
+Epoch: 022/100 | Batch 0037/0043 | Averaged Loss: 0.8686
+Epoch: 022/100 | Batch 0038/0043 | Averaged Loss: 0.8967
+Epoch: 022/100 | Batch 0039/0043 | Averaged Loss: 0.9241
+Epoch: 022/100 | Batch 0040/0043 | Averaged Loss: 0.9489
+Epoch: 022/100 | Batch 0041/0043 | Averaged Loss: 0.8607
+Epoch: 022/100 | Batch 0042/0043 | Averaged Loss: 0.9280
+Epoch: 022/100 | Train: 69.22% | Validation: 68.24%
+Time elapsed: 3.276364803314209 min
+Epoch: 023/100 | Batch 0000/0043 | Averaged Loss: 0.8216
+Epoch: 023/100 | Batch 0001/0043 | Averaged Loss: 0.8286
+Epoch: 023/100 | Batch 0002/0043 | Averaged Loss: 0.7862
+Epoch: 023/100 | Batch 0003/0043 | Averaged Loss: 0.7956
+Epoch: 023/100 | Batch 0004/0043 | Averaged Loss: 0.7999
+Epoch: 023/100 | Batch 0005/0043 | Averaged Loss: 0.8592
+Epoch: 023/100 | Batch 0006/0043 | Averaged Loss: 0.8422
+Epoch: 023/100 | Batch 0007/0043 | Averaged Loss: 0.8792
+Epoch: 023/100 | Batch 0008/0043 | Averaged Loss: 0.8471
+Epoch: 023/100 | Batch 0009/0043 | Averaged Loss: 0.9237
+Epoch: 023/100 | Batch 0010/0043 | Averaged Loss: 0.8960
+Epoch: 023/100 | Batch 0011/0043 | Averaged Loss: 0.9536
+Epoch: 023/100 | Batch 0012/0043 | Averaged Loss: 0.8457
+Epoch: 023/100 | Batch 0013/0043 | Averaged Loss: 0.9523
+Epoch: 023/100 | Batch 0014/0043 | Averaged Loss: 0.9043
+Epoch: 023/100 | Batch 0015/0043 | Averaged Loss: 0.8608
+Epoch: 023/100 | Batch 0016/0043 | Averaged Loss: 0.9270
+Epoch: 023/100 | Batch 0017/0043 | Averaged Loss: 0.7960
+Epoch: 023/100 | Batch 0018/0043 | Averaged Loss: 0.8127
+Epoch: 023/100 | Batch 0019/0043 | Averaged Loss: 0.8015
+Epoch: 023/100 | Batch 0020/0043 | Averaged Loss: 0.7765
+Epoch: 023/100 | Batch 0021/0043 | Averaged Loss: 0.8771
+Epoch: 023/100 | Batch 0022/0043 | Averaged Loss: 0.8714
+Epoch: 023/100 | Batch 0023/0043 | Averaged Loss: 0.8586
+Epoch: 023/100 | Batch 0024/0043 | Averaged Loss: 0.8868
+Epoch: 023/100 | Batch 0025/0043 | Averaged Loss: 0.7980
+Epoch: 023/100 | Batch 0026/0043 | Averaged Loss: 0.9008
+Epoch: 023/100 | Batch 0027/0043 | Averaged Loss: 0.9940
+Epoch: 023/100 | Batch 0028/0043 | Averaged Loss: 0.9522
+Epoch: 023/100 | Batch 0029/0043 | Averaged Loss: 0.9037
+Epoch: 023/100 | Batch 0030/0043 | Averaged Loss: 0.9232
+Epoch: 023/100 | Batch 0031/0043 | Averaged Loss: 0.9725
+Epoch: 023/100 | Batch 0032/0043 | Averaged Loss: 0.9193
+Epoch: 023/100 | Batch 0033/0043 | Averaged Loss: 0.8966
+Epoch: 023/100 | Batch 0034/0043 | Averaged Loss: 0.9265
+Epoch: 023/100 | Batch 0035/0043 | Averaged Loss: 0.9100
+Epoch: 023/100 | Batch 0036/0043 | Averaged Loss: 0.8856
+Epoch: 023/100 | Batch 0037/0043 | Averaged Loss: 0.8642
+Epoch: 023/100 | Batch 0038/0043 | Averaged Loss: 0.8399
+Epoch: 023/100 | Batch 0039/0043 | Averaged Loss: 0.8443
+Epoch: 023/100 | Batch 0040/0043 | Averaged Loss: 0.9369
+Epoch: 023/100 | Batch 0041/0043 | Averaged Loss: 0.8663
+Epoch: 023/100 | Batch 0042/0043 | Averaged Loss: 0.8103
+Epoch: 023/100 | Train: 72.37% | Validation: 69.58%
+Time elapsed: 3.424492120742798 min
+Epoch: 024/100 | Batch 0000/0043 | Averaged Loss: 0.7704
+Epoch: 024/100 | Batch 0001/0043 | Averaged Loss: 0.8298
+Epoch: 024/100 | Batch 0002/0043 | Averaged Loss: 0.7371
+Epoch: 024/100 | Batch 0003/0043 | Averaged Loss: 0.7324
+Epoch: 024/100 | Batch 0004/0043 | Averaged Loss: 0.8090
+Epoch: 024/100 | Batch 0005/0043 | Averaged Loss: 0.8282
+Epoch: 024/100 | Batch 0006/0043 | Averaged Loss: 0.8621
+Epoch: 024/100 | Batch 0007/0043 | Averaged Loss: 0.8538
+Epoch: 024/100 | Batch 0008/0043 | Averaged Loss: 0.8166
+Epoch: 024/100 | Batch 0009/0043 | Averaged Loss: 0.8504
+Epoch: 024/100 | Batch 0010/0043 | Averaged Loss: 0.8473
+Epoch: 024/100 | Batch 0011/0043 | Averaged Loss: 0.7913
+Epoch: 024/100 | Batch 0012/0043 | Averaged Loss: 0.8380
+Epoch: 024/100 | Batch 0013/0043 | Averaged Loss: 0.8203
+Epoch: 024/100 | Batch 0014/0043 | Averaged Loss: 0.8583
+Epoch: 024/100 | Batch 0015/0043 | Averaged Loss: 0.8079
+Epoch: 024/100 | Batch 0016/0043 | Averaged Loss: 0.7940
+Epoch: 024/100 | Batch 0017/0043 | Averaged Loss: 0.8238
+Epoch: 024/100 | Batch 0018/0043 | Averaged Loss: 0.8563
+Epoch: 024/100 | Batch 0019/0043 | Averaged Loss: 0.8904
+Epoch: 024/100 | Batch 0020/0043 | Averaged Loss: 0.7268
+Epoch: 024/100 | Batch 0021/0043 | Averaged Loss: 0.9609
+Epoch: 024/100 | Batch 0022/0043 | Averaged Loss: 0.7957
+Epoch: 024/100 | Batch 0023/0043 | Averaged Loss: 0.8440
+Epoch: 024/100 | Batch 0024/0043 | Averaged Loss: 0.8183
+Epoch: 024/100 | Batch 0025/0043 | Averaged Loss: 0.8325
+Epoch: 024/100 | Batch 0026/0043 | Averaged Loss: 0.8122
+Epoch: 024/100 | Batch 0027/0043 | Averaged Loss: 0.7781
+Epoch: 024/100 | Batch 0028/0043 | Averaged Loss: 0.7920
+Epoch: 024/100 | Batch 0029/0043 | Averaged Loss: 0.8069
+Epoch: 024/100 | Batch 0030/0043 | Averaged Loss: 0.8071
+Epoch: 024/100 | Batch 0031/0043 | Averaged Loss: 0.8392
+Epoch: 024/100 | Batch 0032/0043 | Averaged Loss: 0.7525
+Epoch: 024/100 | Batch 0033/0043 | Averaged Loss: 0.8473
+Epoch: 024/100 | Batch 0034/0043 | Averaged Loss: 0.7776
+Epoch: 024/100 | Batch 0035/0043 | Averaged Loss: 0.8548
+Epoch: 024/100 | Batch 0036/0043 | Averaged Loss: 0.8621
+Epoch: 024/100 | Batch 0037/0043 | Averaged Loss: 0.8710
+Epoch: 024/100 | Batch 0038/0043 | Averaged Loss: 0.8029
+Epoch: 024/100 | Batch 0039/0043 | Averaged Loss: 0.8096
+Epoch: 024/100 | Batch 0040/0043 | Averaged Loss: 0.7743
+Epoch: 024/100 | Batch 0041/0043 | Averaged Loss: 0.8444
+Epoch: 024/100 | Batch 0042/0043 | Averaged Loss: 0.7601
+Epoch: 024/100 | Train: 73.04% | Validation: 69.43%
+Time elapsed: 3.57242488861084 min
+Epoch: 025/100 | Batch 0000/0043 | Averaged Loss: 0.8080
+Epoch: 025/100 | Batch 0001/0043 | Averaged Loss: 0.7974
+Epoch: 025/100 | Batch 0002/0043 | Averaged Loss: 0.7678
+Epoch: 025/100 | Batch 0003/0043 | Averaged Loss: 0.7708
+Epoch: 025/100 | Batch 0004/0043 | Averaged Loss: 0.7590
+Epoch: 025/100 | Batch 0005/0043 | Averaged Loss: 0.7451
+Epoch: 025/100 | Batch 0006/0043 | Averaged Loss: 0.7833
+Epoch: 025/100 | Batch 0007/0043 | Averaged Loss: 0.7728
+Epoch: 025/100 | Batch 0008/0043 | Averaged Loss: 0.7622
+Epoch: 025/100 | Batch 0009/0043 | Averaged Loss: 0.7213
+Epoch: 025/100 | Batch 0010/0043 | Averaged Loss: 0.7140
+Epoch: 025/100 | Batch 0011/0043 | Averaged Loss: 0.7336
+Epoch: 025/100 | Batch 0012/0043 | Averaged Loss: 0.7811
+Epoch: 025/100 | Batch 0013/0043 | Averaged Loss: 0.8169
+Epoch: 025/100 | Batch 0014/0043 | Averaged Loss: 0.7189
+Epoch: 025/100 | Batch 0015/0043 | Averaged Loss: 0.7710
+Epoch: 025/100 | Batch 0016/0043 | Averaged Loss: 0.6960
+Epoch: 025/100 | Batch 0017/0043 | Averaged Loss: 0.7007
+Epoch: 025/100 | Batch 0018/0043 | Averaged Loss: 0.7582
+Epoch: 025/100 | Batch 0019/0043 | Averaged Loss: 0.7436
+Epoch: 025/100 | Batch 0020/0043 | Averaged Loss: 0.6987
+Epoch: 025/100 | Batch 0021/0043 | Averaged Loss: 0.7493
+Epoch: 025/100 | Batch 0022/0043 | Averaged Loss: 0.7485
+Epoch: 025/100 | Batch 0023/0043 | Averaged Loss: 0.7607
+Epoch: 025/100 | Batch 0024/0043 | Averaged Loss: 0.7250
+Epoch: 025/100 | Batch 0025/0043 | Averaged Loss: 0.7780
+Epoch: 025/100 | Batch 0026/0043 | Averaged Loss: 0.7761
+Epoch: 025/100 | Batch 0027/0043 | Averaged Loss: 0.7818
+Epoch: 025/100 | Batch 0028/0043 | Averaged Loss: 0.7893
+Epoch: 025/100 | Batch 0029/0043 | Averaged Loss: 0.7904
+Epoch: 025/100 | Batch 0030/0043 | Averaged Loss: 0.7893
+Epoch: 025/100 | Batch 0031/0043 | Averaged Loss: 0.7474
+Epoch: 025/100 | Batch 0032/0043 | Averaged Loss: 0.7310
+Epoch: 025/100 | Batch 0033/0043 | Averaged Loss: 0.7526
+Epoch: 025/100 | Batch 0034/0043 | Averaged Loss: 0.7581
+Epoch: 025/100 | Batch 0035/0043 | Averaged Loss: 0.7380
+Epoch: 025/100 | Batch 0036/0043 | Averaged Loss: 0.7876
+Epoch: 025/100 | Batch 0037/0043 | Averaged Loss: 0.7347
+Epoch: 025/100 | Batch 0038/0043 | Averaged Loss: 0.7529
+Epoch: 025/100 | Batch 0039/0043 | Averaged Loss: 0.7900
+Epoch: 025/100 | Batch 0040/0043 | Averaged Loss: 0.7638
+Epoch: 025/100 | Batch 0041/0043 | Averaged Loss: 0.8098
+Epoch: 025/100 | Batch 0042/0043 | Averaged Loss: 0.7206
+Epoch: 025/100 | Train: 77.19% | Validation: 72.14%
+Time elapsed: 3.720778703689575 min
+Epoch: 026/100 | Batch 0000/0043 | Averaged Loss: 0.7191
+Epoch: 026/100 | Batch 0001/0043 | Averaged Loss: 0.7169
+Epoch: 026/100 | Batch 0002/0043 | Averaged Loss: 0.7123
+Epoch: 026/100 | Batch 0003/0043 | Averaged Loss: 0.6921
+Epoch: 026/100 | Batch 0004/0043 | Averaged Loss: 0.6637
+Epoch: 026/100 | Batch 0005/0043 | Averaged Loss: 0.6502
+Epoch: 026/100 | Batch 0006/0043 | Averaged Loss: 0.6541
+Epoch: 026/100 | Batch 0007/0043 | Averaged Loss: 0.6804
+Epoch: 026/100 | Batch 0008/0043 | Averaged Loss: 0.6769
+Epoch: 026/100 | Batch 0009/0043 | Averaged Loss: 0.6514
+Epoch: 026/100 | Batch 0010/0043 | Averaged Loss: 0.6933
+Epoch: 026/100 | Batch 0011/0043 | Averaged Loss: 0.6562
+Epoch: 026/100 | Batch 0012/0043 | Averaged Loss: 0.6427
+Epoch: 026/100 | Batch 0013/0043 | Averaged Loss: 0.6272
+Epoch: 026/100 | Batch 0014/0043 | Averaged Loss: 0.7246
+Epoch: 026/100 | Batch 0015/0043 | Averaged Loss: 0.6711
+Epoch: 026/100 | Batch 0016/0043 | Averaged Loss: 0.7093
+Epoch: 026/100 | Batch 0017/0043 | Averaged Loss: 0.7112
+Epoch: 026/100 | Batch 0018/0043 | Averaged Loss: 0.6716
+Epoch: 026/100 | Batch 0019/0043 | Averaged Loss: 0.7188
+Epoch: 026/100 | Batch 0020/0043 | Averaged Loss: 0.7435
+Epoch: 026/100 | Batch 0021/0043 | Averaged Loss: 0.7644
+Epoch: 026/100 | Batch 0022/0043 | Averaged Loss: 0.7161
+Epoch: 026/100 | Batch 0023/0043 | Averaged Loss: 0.7145
+Epoch: 026/100 | Batch 0024/0043 | Averaged Loss: 0.7071
+Epoch: 026/100 | Batch 0025/0043 | Averaged Loss: 0.6912
+Epoch: 026/100 | Batch 0026/0043 | Averaged Loss: 0.7237
+Epoch: 026/100 | Batch 0027/0043 | Averaged Loss: 0.6817
+Epoch: 026/100 | Batch 0028/0043 | Averaged Loss: 0.7810
+Epoch: 026/100 | Batch 0029/0043 | Averaged Loss: 0.6995
+Epoch: 026/100 | Batch 0030/0043 | Averaged Loss: 0.7031
+Epoch: 026/100 | Batch 0031/0043 | Averaged Loss: 0.7395
+Epoch: 026/100 | Batch 0032/0043 | Averaged Loss: 0.6637
+Epoch: 026/100 | Batch 0033/0043 | Averaged Loss: 0.7331
+Epoch: 026/100 | Batch 0034/0043 | Averaged Loss: 0.7511
+Epoch: 026/100 | Batch 0035/0043 | Averaged Loss: 0.7579
+Epoch: 026/100 | Batch 0036/0043 | Averaged Loss: 0.7191
+Epoch: 026/100 | Batch 0037/0043 | Averaged Loss: 0.6664
+Epoch: 026/100 | Batch 0038/0043 | Averaged Loss: 0.8635
+Epoch: 026/100 | Batch 0039/0043 | Averaged Loss: 0.7273
+Epoch: 026/100 | Batch 0040/0043 | Averaged Loss: 0.7980
+Epoch: 026/100 | Batch 0041/0043 | Averaged Loss: 0.7993
+Epoch: 026/100 | Batch 0042/0043 | Averaged Loss: 0.7051
+Epoch: 026/100 | Train: 77.05% | Validation: 70.68%
+Time elapsed: 3.8690428733825684 min
+Epoch: 027/100 | Batch 0000/0043 | Averaged Loss: 0.6210
+Epoch: 027/100 | Batch 0001/0043 | Averaged Loss: 0.6801
+Epoch: 027/100 | Batch 0002/0043 | Averaged Loss: 0.6486
+Epoch: 027/100 | Batch 0003/0043 | Averaged Loss: 0.6365
+Epoch: 027/100 | Batch 0004/0043 | Averaged Loss: 0.6382
+Epoch: 027/100 | Batch 0005/0043 | Averaged Loss: 0.6320
+Epoch: 027/100 | Batch 0006/0043 | Averaged Loss: 0.6167
+Epoch: 027/100 | Batch 0007/0043 | Averaged Loss: 0.6007
+Epoch: 027/100 | Batch 0008/0043 | Averaged Loss: 0.6862
+Epoch: 027/100 | Batch 0009/0043 | Averaged Loss: 0.5828
+Epoch: 027/100 | Batch 0010/0043 | Averaged Loss: 0.7402
+Epoch: 027/100 | Batch 0011/0043 | Averaged Loss: 0.6793
+Epoch: 027/100 | Batch 0012/0043 | Averaged Loss: 0.6845
+Epoch: 027/100 | Batch 0013/0043 | Averaged Loss: 0.6547
+Epoch: 027/100 | Batch 0014/0043 | Averaged Loss: 0.6597
+Epoch: 027/100 | Batch 0015/0043 | Averaged Loss: 0.6958
+Epoch: 027/100 | Batch 0016/0043 | Averaged Loss: 0.6694
+Epoch: 027/100 | Batch 0017/0043 | Averaged Loss: 0.6837
+Epoch: 027/100 | Batch 0018/0043 | Averaged Loss: 0.7167
+Epoch: 027/100 | Batch 0019/0043 | Averaged Loss: 0.7483
+Epoch: 027/100 | Batch 0020/0043 | Averaged Loss: 0.7035
+Epoch: 027/100 | Batch 0021/0043 | Averaged Loss: 0.6503
+Epoch: 027/100 | Batch 0022/0043 | Averaged Loss: 0.6961
+Epoch: 027/100 | Batch 0023/0043 | Averaged Loss: 0.7019
+Epoch: 027/100 | Batch 0024/0043 | Averaged Loss: 0.6923
+Epoch: 027/100 | Batch 0025/0043 | Averaged Loss: 0.6536
+Epoch: 027/100 | Batch 0026/0043 | Averaged Loss: 0.6800
+Epoch: 027/100 | Batch 0027/0043 | Averaged Loss: 0.6976
+Epoch: 027/100 | Batch 0028/0043 | Averaged Loss: 0.6992
+Epoch: 027/100 | Batch 0029/0043 | Averaged Loss: 0.6920
+Epoch: 027/100 | Batch 0030/0043 | Averaged Loss: 0.7017
+Epoch: 027/100 | Batch 0031/0043 | Averaged Loss: 0.6681
+Epoch: 027/100 | Batch 0032/0043 | Averaged Loss: 0.8013
+Epoch: 027/100 | Batch 0033/0043 | Averaged Loss: 0.7081
+Epoch: 027/100 | Batch 0034/0043 | Averaged Loss: 0.7170
+Epoch: 027/100 | Batch 0035/0043 | Averaged Loss: 0.6852
+Epoch: 027/100 | Batch 0036/0043 | Averaged Loss: 0.7434
+Epoch: 027/100 | Batch 0037/0043 | Averaged Loss: 0.7167
+Epoch: 027/100 | Batch 0038/0043 | Averaged Loss: 0.7426
+Epoch: 027/100 | Batch 0039/0043 | Averaged Loss: 0.7558
+Epoch: 027/100 | Batch 0040/0043 | Averaged Loss: 0.7606
+Epoch: 027/100 | Batch 0041/0043 | Averaged Loss: 0.6685
+Epoch: 027/100 | Batch 0042/0043 | Averaged Loss: 0.7373
+Epoch: 027/100 | Train: 78.00% | Validation: 70.92%
+Time elapsed: 4.01759147644043 min
+Epoch: 028/100 | Batch 0000/0043 | Averaged Loss: 0.6658
+Epoch: 028/100 | Batch 0001/0043 | Averaged Loss: 0.6250
+Epoch: 028/100 | Batch 0002/0043 | Averaged Loss: 0.6007
+Epoch: 028/100 | Batch 0003/0043 | Averaged Loss: 0.6258
+Epoch: 028/100 | Batch 0004/0043 | Averaged Loss: 0.6902
+Epoch: 028/100 | Batch 0005/0043 | Averaged Loss: 0.7058
+Epoch: 028/100 | Batch 0006/0043 | Averaged Loss: 0.6684
+Epoch: 028/100 | Batch 0007/0043 | Averaged Loss: 0.6633
+Epoch: 028/100 | Batch 0008/0043 | Averaged Loss: 0.6892
+Epoch: 028/100 | Batch 0009/0043 | Averaged Loss: 0.6528
+Epoch: 028/100 | Batch 0010/0043 | Averaged Loss: 0.7069
+Epoch: 028/100 | Batch 0011/0043 | Averaged Loss: 0.7321
+Epoch: 028/100 | Batch 0012/0043 | Averaged Loss: 0.6291
+Epoch: 028/100 | Batch 0013/0043 | Averaged Loss: 0.6267
+Epoch: 028/100 | Batch 0014/0043 | Averaged Loss: 0.7017
+Epoch: 028/100 | Batch 0015/0043 | Averaged Loss: 0.7787
+Epoch: 028/100 | Batch 0016/0043 | Averaged Loss: 0.6480
+Epoch: 028/100 | Batch 0017/0043 | Averaged Loss: 0.6392
+Epoch: 028/100 | Batch 0018/0043 | Averaged Loss: 0.6288
+Epoch: 028/100 | Batch 0019/0043 | Averaged Loss: 0.6339
+Epoch: 028/100 | Batch 0020/0043 | Averaged Loss: 0.7106
+Epoch: 028/100 | Batch 0021/0043 | Averaged Loss: 0.6579
+Epoch: 028/100 | Batch 0022/0043 | Averaged Loss: 0.6113
+Epoch: 028/100 | Batch 0023/0043 | Averaged Loss: 0.6467
+Epoch: 028/100 | Batch 0024/0043 | Averaged Loss: 0.5552
+Epoch: 028/100 | Batch 0025/0043 | Averaged Loss: 0.6067
+Epoch: 028/100 | Batch 0026/0043 | Averaged Loss: 0.5958
+Epoch: 028/100 | Batch 0027/0043 | Averaged Loss: 0.6256
+Epoch: 028/100 | Batch 0028/0043 | Averaged Loss: 0.5919
+Epoch: 028/100 | Batch 0029/0043 | Averaged Loss: 0.6075
+Epoch: 028/100 | Batch 0030/0043 | Averaged Loss: 0.6801
+Epoch: 028/100 | Batch 0031/0043 | Averaged Loss: 0.6511
+Epoch: 028/100 | Batch 0032/0043 | Averaged Loss: 0.6279
+Epoch: 028/100 | Batch 0033/0043 | Averaged Loss: 0.6147
+Epoch: 028/100 | Batch 0034/0043 | Averaged Loss: 0.6897
+Epoch: 028/100 | Batch 0035/0043 | Averaged Loss: 0.6570
+Epoch: 028/100 | Batch 0036/0043 | Averaged Loss: 0.6683
+Epoch: 028/100 | Batch 0037/0043 | Averaged Loss: 0.6555
+Epoch: 028/100 | Batch 0038/0043 | Averaged Loss: 0.6656
+Epoch: 028/100 | Batch 0039/0043 | Averaged Loss: 0.6537
+Epoch: 028/100 | Batch 0040/0043 | Averaged Loss: 0.6846
+Epoch: 028/100 | Batch 0041/0043 | Averaged Loss: 0.5748
+Epoch: 028/100 | Batch 0042/0043 | Averaged Loss: 0.6876
+Epoch: 028/100 | Train: 80.42% | Validation: 73.78%
+Time elapsed: 4.165477275848389 min
+Epoch: 029/100 | Batch 0000/0043 | Averaged Loss: 0.6001
+Epoch: 029/100 | Batch 0001/0043 | Averaged Loss: 0.5825
+Epoch: 029/100 | Batch 0002/0043 | Averaged Loss: 0.5821
+Epoch: 029/100 | Batch 0003/0043 | Averaged Loss: 0.5153
+Epoch: 029/100 | Batch 0004/0043 | Averaged Loss: 0.5632
+Epoch: 029/100 | Batch 0005/0043 | Averaged Loss: 0.5656
+Epoch: 029/100 | Batch 0006/0043 | Averaged Loss: 0.5670
+Epoch: 029/100 | Batch 0007/0043 | Averaged Loss: 0.5649
+Epoch: 029/100 | Batch 0008/0043 | Averaged Loss: 0.5805
+Epoch: 029/100 | Batch 0009/0043 | Averaged Loss: 0.5557
+Epoch: 029/100 | Batch 0010/0043 | Averaged Loss: 0.5467
+Epoch: 029/100 | Batch 0011/0043 | Averaged Loss: 0.5628
+Epoch: 029/100 | Batch 0012/0043 | Averaged Loss: 0.5598
+Epoch: 029/100 | Batch 0013/0043 | Averaged Loss: 0.5480
+Epoch: 029/100 | Batch 0014/0043 | Averaged Loss: 0.5715
+Epoch: 029/100 | Batch 0015/0043 | Averaged Loss: 0.6375
+Epoch: 029/100 | Batch 0016/0043 | Averaged Loss: 0.5493
+Epoch: 029/100 | Batch 0017/0043 | Averaged Loss: 0.5029
+Epoch: 029/100 | Batch 0018/0043 | Averaged Loss: 0.6528
+Epoch: 029/100 | Batch 0019/0043 | Averaged Loss: 0.6010
+Epoch: 029/100 | Batch 0020/0043 | Averaged Loss: 0.6323
+Epoch: 029/100 | Batch 0021/0043 | Averaged Loss: 0.6435
+Epoch: 029/100 | Batch 0022/0043 | Averaged Loss: 0.5900
+Epoch: 029/100 | Batch 0023/0043 | Averaged Loss: 0.6704
+Epoch: 029/100 | Batch 0024/0043 | Averaged Loss: 0.5776
+Epoch: 029/100 | Batch 0025/0043 | Averaged Loss: 0.6035
+Epoch: 029/100 | Batch 0026/0043 | Averaged Loss: 0.6184
+Epoch: 029/100 | Batch 0027/0043 | Averaged Loss: 0.5900
+Epoch: 029/100 | Batch 0028/0043 | Averaged Loss: 0.6431
+Epoch: 029/100 | Batch 0029/0043 | Averaged Loss: 0.6060
+Epoch: 029/100 | Batch 0030/0043 | Averaged Loss: 0.6547
+Epoch: 029/100 | Batch 0031/0043 | Averaged Loss: 0.6191
+Epoch: 029/100 | Batch 0032/0043 | Averaged Loss: 0.6788
+Epoch: 029/100 | Batch 0033/0043 | Averaged Loss: 0.6450
+Epoch: 029/100 | Batch 0034/0043 | Averaged Loss: 0.6431
+Epoch: 029/100 | Batch 0035/0043 | Averaged Loss: 0.5977
+Epoch: 029/100 | Batch 0036/0043 | Averaged Loss: 0.6184
+Epoch: 029/100 | Batch 0037/0043 | Averaged Loss: 0.5858
+Epoch: 029/100 | Batch 0038/0043 | Averaged Loss: 0.5782
+Epoch: 029/100 | Batch 0039/0043 | Averaged Loss: 0.6420
+Epoch: 029/100 | Batch 0040/0043 | Averaged Loss: 0.6946
+Epoch: 029/100 | Batch 0041/0043 | Averaged Loss: 0.6434
+Epoch: 029/100 | Batch 0042/0043 | Averaged Loss: 0.6036
+Epoch: 029/100 | Train: 80.12% | Validation: 71.58%
+Time elapsed: 4.3133368492126465 min
+Epoch: 030/100 | Batch 0000/0043 | Averaged Loss: 0.6247
+Epoch: 030/100 | Batch 0001/0043 | Averaged Loss: 0.5961
+Epoch: 030/100 | Batch 0002/0043 | Averaged Loss: 0.5856
+Epoch: 030/100 | Batch 0003/0043 | Averaged Loss: 0.6245
+Epoch: 030/100 | Batch 0004/0043 | Averaged Loss: 0.5444
+Epoch: 030/100 | Batch 0005/0043 | Averaged Loss: 0.5652
+Epoch: 030/100 | Batch 0006/0043 | Averaged Loss: 0.5593
+Epoch: 030/100 | Batch 0007/0043 | Averaged Loss: 0.5559
+Epoch: 030/100 | Batch 0008/0043 | Averaged Loss: 0.5013
+Epoch: 030/100 | Batch 0009/0043 | Averaged Loss: 0.5219
+Epoch: 030/100 | Batch 0010/0043 | Averaged Loss: 0.5200
+Epoch: 030/100 | Batch 0011/0043 | Averaged Loss: 0.5852
+Epoch: 030/100 | Batch 0012/0043 | Averaged Loss: 0.5174
+Epoch: 030/100 | Batch 0013/0043 | Averaged Loss: 0.5502
+Epoch: 030/100 | Batch 0014/0043 | Averaged Loss: 0.5612
+Epoch: 030/100 | Batch 0015/0043 | Averaged Loss: 0.6064
+Epoch: 030/100 | Batch 0016/0043 | Averaged Loss: 0.5271
+Epoch: 030/100 | Batch 0017/0043 | Averaged Loss: 0.5623
+Epoch: 030/100 | Batch 0018/0043 | Averaged Loss: 0.5241
+Epoch: 030/100 | Batch 0019/0043 | Averaged Loss: 0.5563
+Epoch: 030/100 | Batch 0020/0043 | Averaged Loss: 0.5476
+Epoch: 030/100 | Batch 0021/0043 | Averaged Loss: 0.5624
+Epoch: 030/100 | Batch 0022/0043 | Averaged Loss: 0.5495
+Epoch: 030/100 | Batch 0023/0043 | Averaged Loss: 0.5672
+Epoch: 030/100 | Batch 0024/0043 | Averaged Loss: 0.5548
+Epoch: 030/100 | Batch 0025/0043 | Averaged Loss: 0.5601
+Epoch: 030/100 | Batch 0026/0043 | Averaged Loss: 0.6044
+Epoch: 030/100 | Batch 0027/0043 | Averaged Loss: 0.5345
+Epoch: 030/100 | Batch 0028/0043 | Averaged Loss: 0.5534
+Epoch: 030/100 | Batch 0029/0043 | Averaged Loss: 0.5905
+Epoch: 030/100 | Batch 0030/0043 | Averaged Loss: 0.5562
+Epoch: 030/100 | Batch 0031/0043 | Averaged Loss: 0.5837
+Epoch: 030/100 | Batch 0032/0043 | Averaged Loss: 0.5663
+Epoch: 030/100 | Batch 0033/0043 | Averaged Loss: 0.5776
+Epoch: 030/100 | Batch 0034/0043 | Averaged Loss: 0.5521
+Epoch: 030/100 | Batch 0035/0043 | Averaged Loss: 0.6232
+Epoch: 030/100 | Batch 0036/0043 | Averaged Loss: 0.5512
+Epoch: 030/100 | Batch 0037/0043 | Averaged Loss: 0.6382
+Epoch: 030/100 | Batch 0038/0043 | Averaged Loss: 0.5571
+Epoch: 030/100 | Batch 0039/0043 | Averaged Loss: 0.5473
+Epoch: 030/100 | Batch 0040/0043 | Averaged Loss: 0.6140
+Epoch: 030/100 | Batch 0041/0043 | Averaged Loss: 0.6310
+Epoch: 030/100 | Batch 0042/0043 | Averaged Loss: 0.6364
+Epoch: 030/100 | Train: 81.40% | Validation: 72.56%
+Time elapsed: 4.461178779602051 min
+Epoch: 031/100 | Batch 0000/0043 | Averaged Loss: 0.5423
+Epoch: 031/100 | Batch 0001/0043 | Averaged Loss: 0.5479
+Epoch: 031/100 | Batch 0002/0043 | Averaged Loss: 0.4999
+Epoch: 031/100 | Batch 0003/0043 | Averaged Loss: 0.5514
+Epoch: 031/100 | Batch 0004/0043 | Averaged Loss: 0.5414
+Epoch: 031/100 | Batch 0005/0043 | Averaged Loss: 0.4446
+Epoch: 031/100 | Batch 0006/0043 | Averaged Loss: 0.5066
+Epoch: 031/100 | Batch 0007/0043 | Averaged Loss: 0.5014
+Epoch: 031/100 | Batch 0008/0043 | Averaged Loss: 0.4842
+Epoch: 031/100 | Batch 0009/0043 | Averaged Loss: 0.5588
+Epoch: 031/100 | Batch 0010/0043 | Averaged Loss: 0.5347
+Epoch: 031/100 | Batch 0011/0043 | Averaged Loss: 0.5280
+Epoch: 031/100 | Batch 0012/0043 | Averaged Loss: 0.5286
+Epoch: 031/100 | Batch 0013/0043 | Averaged Loss: 0.4769
+Epoch: 031/100 | Batch 0014/0043 | Averaged Loss: 0.5170
+Epoch: 031/100 | Batch 0015/0043 | Averaged Loss: 0.5292
+Epoch: 031/100 | Batch 0016/0043 | Averaged Loss: 0.5078
+Epoch: 031/100 | Batch 0017/0043 | Averaged Loss: 0.5087
+Epoch: 031/100 | Batch 0018/0043 | Averaged Loss: 0.5279
+Epoch: 031/100 | Batch 0019/0043 | Averaged Loss: 0.5365
+Epoch: 031/100 | Batch 0020/0043 | Averaged Loss: 0.5641
+Epoch: 031/100 | Batch 0021/0043 | Averaged Loss: 0.6098
+Epoch: 031/100 | Batch 0022/0043 | Averaged Loss: 0.6117
+Epoch: 031/100 | Batch 0023/0043 | Averaged Loss: 0.5584
+Epoch: 031/100 | Batch 0024/0043 | Averaged Loss: 0.5356
+Epoch: 031/100 | Batch 0025/0043 | Averaged Loss: 0.5619
+Epoch: 031/100 | Batch 0026/0043 | Averaged Loss: 0.5441
+Epoch: 031/100 | Batch 0027/0043 | Averaged Loss: 0.5762
+Epoch: 031/100 | Batch 0028/0043 | Averaged Loss: 0.5637
+Epoch: 031/100 | Batch 0029/0043 | Averaged Loss: 0.5832
+Epoch: 031/100 | Batch 0030/0043 | Averaged Loss: 0.5069
+Epoch: 031/100 | Batch 0031/0043 | Averaged Loss: 0.6274
+Epoch: 031/100 | Batch 0032/0043 | Averaged Loss: 0.5635
+Epoch: 031/100 | Batch 0033/0043 | Averaged Loss: 0.5263
+Epoch: 031/100 | Batch 0034/0043 | Averaged Loss: 0.5401
+Epoch: 031/100 | Batch 0035/0043 | Averaged Loss: 0.5228
+Epoch: 031/100 | Batch 0036/0043 | Averaged Loss: 0.5029
+Epoch: 031/100 | Batch 0037/0043 | Averaged Loss: 0.5780
+Epoch: 031/100 | Batch 0038/0043 | Averaged Loss: 0.5663
+Epoch: 031/100 | Batch 0039/0043 | Averaged Loss: 0.5171
+Epoch: 031/100 | Batch 0040/0043 | Averaged Loss: 0.5898
+Epoch: 031/100 | Batch 0041/0043 | Averaged Loss: 0.5359
+Epoch: 031/100 | Batch 0042/0043 | Averaged Loss: 0.5533
+Epoch: 031/100 | Train: 83.85% | Validation: 73.58%
+Time elapsed: 4.609407424926758 min
+Epoch: 032/100 | Batch 0000/0043 | Averaged Loss: 0.5280
+Epoch: 032/100 | Batch 0001/0043 | Averaged Loss: 0.5018
+Epoch: 032/100 | Batch 0002/0043 | Averaged Loss: 0.4990
+Epoch: 032/100 | Batch 0003/0043 | Averaged Loss: 0.4697
+Epoch: 032/100 | Batch 0004/0043 | Averaged Loss: 0.5003
+Epoch: 032/100 | Batch 0005/0043 | Averaged Loss: 0.4693
+Epoch: 032/100 | Batch 0006/0043 | Averaged Loss: 0.4881
+Epoch: 032/100 | Batch 0007/0043 | Averaged Loss: 0.4428
+Epoch: 032/100 | Batch 0008/0043 | Averaged Loss: 0.4796
+Epoch: 032/100 | Batch 0009/0043 | Averaged Loss: 0.4832
+Epoch: 032/100 | Batch 0010/0043 | Averaged Loss: 0.4850
+Epoch: 032/100 | Batch 0011/0043 | Averaged Loss: 0.4579
+Epoch: 032/100 | Batch 0012/0043 | Averaged Loss: 0.4573
+Epoch: 032/100 | Batch 0013/0043 | Averaged Loss: 0.4990
+Epoch: 032/100 | Batch 0014/0043 | Averaged Loss: 0.5522
+Epoch: 032/100 | Batch 0015/0043 | Averaged Loss: 0.5180
+Epoch: 032/100 | Batch 0016/0043 | Averaged Loss: 0.5233
+Epoch: 032/100 | Batch 0017/0043 | Averaged Loss: 0.5008
+Epoch: 032/100 | Batch 0018/0043 | Averaged Loss: 0.4817
+Epoch: 032/100 | Batch 0019/0043 | Averaged Loss: 0.4586
+Epoch: 032/100 | Batch 0020/0043 | Averaged Loss: 0.5195
+Epoch: 032/100 | Batch 0021/0043 | Averaged Loss: 0.5461
+Epoch: 032/100 | Batch 0022/0043 | Averaged Loss: 0.4782
+Epoch: 032/100 | Batch 0023/0043 | Averaged Loss: 0.5282
+Epoch: 032/100 | Batch 0024/0043 | Averaged Loss: 0.5387
+Epoch: 032/100 | Batch 0025/0043 | Averaged Loss: 0.5306
+Epoch: 032/100 | Batch 0026/0043 | Averaged Loss: 0.4788
+Epoch: 032/100 | Batch 0027/0043 | Averaged Loss: 0.5160
+Epoch: 032/100 | Batch 0028/0043 | Averaged Loss: 0.5011
+Epoch: 032/100 | Batch 0029/0043 | Averaged Loss: 0.4721
+Epoch: 032/100 | Batch 0030/0043 | Averaged Loss: 0.4731
+Epoch: 032/100 | Batch 0031/0043 | Averaged Loss: 0.5276
+Epoch: 032/100 | Batch 0032/0043 | Averaged Loss: 0.5250
+Epoch: 032/100 | Batch 0033/0043 | Averaged Loss: 0.5866
+Epoch: 032/100 | Batch 0034/0043 | Averaged Loss: 0.5500
+Epoch: 032/100 | Batch 0035/0043 | Averaged Loss: 0.5939
+Epoch: 032/100 | Batch 0036/0043 | Averaged Loss: 0.5641
+Epoch: 032/100 | Batch 0037/0043 | Averaged Loss: 0.5209
+Epoch: 032/100 | Batch 0038/0043 | Averaged Loss: 0.5317
+Epoch: 032/100 | Batch 0039/0043 | Averaged Loss: 0.5167
+Epoch: 032/100 | Batch 0040/0043 | Averaged Loss: 0.5705
+Epoch: 032/100 | Batch 0041/0043 | Averaged Loss: 0.5689
+Epoch: 032/100 | Batch 0042/0043 | Averaged Loss: 0.5264
+Epoch: 032/100 | Train: 83.80% | Validation: 73.61%
+Time elapsed: 4.757781982421875 min
+Epoch: 033/100 | Batch 0000/0043 | Averaged Loss: 0.4975
+Epoch: 033/100 | Batch 0001/0043 | Averaged Loss: 0.4118
+Epoch: 033/100 | Batch 0002/0043 | Averaged Loss: 0.4350
+Epoch: 033/100 | Batch 0003/0043 | Averaged Loss: 0.4484
+Epoch: 033/100 | Batch 0004/0043 | Averaged Loss: 0.4519
+Epoch: 033/100 | Batch 0005/0043 | Averaged Loss: 0.4506
+Epoch: 033/100 | Batch 0006/0043 | Averaged Loss: 0.4537
+Epoch: 033/100 | Batch 0007/0043 | Averaged Loss: 0.4476
+Epoch: 033/100 | Batch 0008/0043 | Averaged Loss: 0.5139
+Epoch: 033/100 | Batch 0009/0043 | Averaged Loss: 0.4468
+Epoch: 033/100 | Batch 0010/0043 | Averaged Loss: 0.4993
+Epoch: 033/100 | Batch 0011/0043 | Averaged Loss: 0.4021
+Epoch: 033/100 | Batch 0012/0043 | Averaged Loss: 0.5146
+Epoch: 033/100 | Batch 0013/0043 | Averaged Loss: 0.4781
+Epoch: 033/100 | Batch 0014/0043 | Averaged Loss: 0.4955
+Epoch: 033/100 | Batch 0015/0043 | Averaged Loss: 0.5023
+Epoch: 033/100 | Batch 0016/0043 | Averaged Loss: 0.4633
+Epoch: 033/100 | Batch 0017/0043 | Averaged Loss: 0.5217
+Epoch: 033/100 | Batch 0018/0043 | Averaged Loss: 0.4696
+Epoch: 033/100 | Batch 0019/0043 | Averaged Loss: 0.4944
+Epoch: 033/100 | Batch 0020/0043 | Averaged Loss: 0.4789
+Epoch: 033/100 | Batch 0021/0043 | Averaged Loss: 0.4407
+Epoch: 033/100 | Batch 0022/0043 | Averaged Loss: 0.4174
+Epoch: 033/100 | Batch 0023/0043 | Averaged Loss: 0.4216
+Epoch: 033/100 | Batch 0024/0043 | Averaged Loss: 0.4820
+Epoch: 033/100 | Batch 0025/0043 | Averaged Loss: 0.4830
+Epoch: 033/100 | Batch 0026/0043 | Averaged Loss: 0.4914
+Epoch: 033/100 | Batch 0027/0043 | Averaged Loss: 0.4756
+Epoch: 033/100 | Batch 0028/0043 | Averaged Loss: 0.4618
+Epoch: 033/100 | Batch 0029/0043 | Averaged Loss: 0.4248
+Epoch: 033/100 | Batch 0030/0043 | Averaged Loss: 0.5387
+Epoch: 033/100 | Batch 0031/0043 | Averaged Loss: 0.4957
+Epoch: 033/100 | Batch 0032/0043 | Averaged Loss: 0.5121
+Epoch: 033/100 | Batch 0033/0043 | Averaged Loss: 0.4668
+Epoch: 033/100 | Batch 0034/0043 | Averaged Loss: 0.5287
+Epoch: 033/100 | Batch 0035/0043 | Averaged Loss: 0.5382
+Epoch: 033/100 | Batch 0036/0043 | Averaged Loss: 0.5267
+Epoch: 033/100 | Batch 0037/0043 | Averaged Loss: 0.4404
+Epoch: 033/100 | Batch 0038/0043 | Averaged Loss: 0.5216
+Epoch: 033/100 | Batch 0039/0043 | Averaged Loss: 0.5147
+Epoch: 033/100 | Batch 0040/0043 | Averaged Loss: 0.4743
+Epoch: 033/100 | Batch 0041/0043 | Averaged Loss: 0.4878
+Epoch: 033/100 | Batch 0042/0043 | Averaged Loss: 0.5023
+Epoch: 033/100 | Train: 86.25% | Validation: 74.51%
+Time elapsed: 4.905848503112793 min
+Epoch: 034/100 | Batch 0000/0043 | Averaged Loss: 0.4042
+Epoch: 034/100 | Batch 0001/0043 | Averaged Loss: 0.4020
+Epoch: 034/100 | Batch 0002/0043 | Averaged Loss: 0.3923
+Epoch: 034/100 | Batch 0003/0043 | Averaged Loss: 0.3851
+Epoch: 034/100 | Batch 0004/0043 | Averaged Loss: 0.4062
+Epoch: 034/100 | Batch 0005/0043 | Averaged Loss: 0.4057
+Epoch: 034/100 | Batch 0006/0043 | Averaged Loss: 0.4723
+Epoch: 034/100 | Batch 0007/0043 | Averaged Loss: 0.3664
+Epoch: 034/100 | Batch 0008/0043 | Averaged Loss: 0.4592
+Epoch: 034/100 | Batch 0009/0043 | Averaged Loss: 0.3980
+Epoch: 034/100 | Batch 0010/0043 | Averaged Loss: 0.4290
+Epoch: 034/100 | Batch 0011/0043 | Averaged Loss: 0.3968
+Epoch: 034/100 | Batch 0012/0043 | Averaged Loss: 0.4358
+Epoch: 034/100 | Batch 0013/0043 | Averaged Loss: 0.3701
+Epoch: 034/100 | Batch 0014/0043 | Averaged Loss: 0.4647
+Epoch: 034/100 | Batch 0015/0043 | Averaged Loss: 0.4549
+Epoch: 034/100 | Batch 0016/0043 | Averaged Loss: 0.4011
+Epoch: 034/100 | Batch 0017/0043 | Averaged Loss: 0.4310
+Epoch: 034/100 | Batch 0018/0043 | Averaged Loss: 0.5005
+Epoch: 034/100 | Batch 0019/0043 | Averaged Loss: 0.4220
+Epoch: 034/100 | Batch 0020/0043 | Averaged Loss: 0.4303
+Epoch: 034/100 | Batch 0021/0043 | Averaged Loss: 0.4285
+Epoch: 034/100 | Batch 0022/0043 | Averaged Loss: 0.4804
+Epoch: 034/100 | Batch 0023/0043 | Averaged Loss: 0.4753
+Epoch: 034/100 | Batch 0024/0043 | Averaged Loss: 0.5111
+Epoch: 034/100 | Batch 0025/0043 | Averaged Loss: 0.4216
+Epoch: 034/100 | Batch 0026/0043 | Averaged Loss: 0.4408
+Epoch: 034/100 | Batch 0027/0043 | Averaged Loss: 0.3804
+Epoch: 034/100 | Batch 0028/0043 | Averaged Loss: 0.5570
+Epoch: 034/100 | Batch 0029/0043 | Averaged Loss: 0.4912
+Epoch: 034/100 | Batch 0030/0043 | Averaged Loss: 0.4215
+Epoch: 034/100 | Batch 0031/0043 | Averaged Loss: 0.3882
+Epoch: 034/100 | Batch 0032/0043 | Averaged Loss: 0.4141
+Epoch: 034/100 | Batch 0033/0043 | Averaged Loss: 0.4034
+Epoch: 034/100 | Batch 0034/0043 | Averaged Loss: 0.4826
+Epoch: 034/100 | Batch 0035/0043 | Averaged Loss: 0.4842
+Epoch: 034/100 | Batch 0036/0043 | Averaged Loss: 0.5653
+Epoch: 034/100 | Batch 0037/0043 | Averaged Loss: 0.4748
+Epoch: 034/100 | Batch 0038/0043 | Averaged Loss: 0.4513
+Epoch: 034/100 | Batch 0039/0043 | Averaged Loss: 0.5072
+Epoch: 034/100 | Batch 0040/0043 | Averaged Loss: 0.5085
+Epoch: 034/100 | Batch 0041/0043 | Averaged Loss: 0.4917
+Epoch: 034/100 | Batch 0042/0043 | Averaged Loss: 0.4562
+Epoch: 034/100 | Train: 84.58% | Validation: 72.51%
+Time elapsed: 5.054357528686523 min
+Epoch: 035/100 | Batch 0000/0043 | Averaged Loss: 0.4889
+Epoch: 035/100 | Batch 0001/0043 | Averaged Loss: 0.4343
+Epoch: 035/100 | Batch 0002/0043 | Averaged Loss: 0.4384
+Epoch: 035/100 | Batch 0003/0043 | Averaged Loss: 0.4758
+Epoch: 035/100 | Batch 0004/0043 | Averaged Loss: 0.3693
+Epoch: 035/100 | Batch 0005/0043 | Averaged Loss: 0.4114
+Epoch: 035/100 | Batch 0006/0043 | Averaged Loss: 0.3848
+Epoch: 035/100 | Batch 0007/0043 | Averaged Loss: 0.4131
+Epoch: 035/100 | Batch 0008/0043 | Averaged Loss: 0.3921
+Epoch: 035/100 | Batch 0009/0043 | Averaged Loss: 0.3849
+Epoch: 035/100 | Batch 0010/0043 | Averaged Loss: 0.4123
+Epoch: 035/100 | Batch 0011/0043 | Averaged Loss: 0.4715
+Epoch: 035/100 | Batch 0012/0043 | Averaged Loss: 0.3875
+Epoch: 035/100 | Batch 0013/0043 | Averaged Loss: 0.3763
+Epoch: 035/100 | Batch 0014/0043 | Averaged Loss: 0.4188
+Epoch: 035/100 | Batch 0015/0043 | Averaged Loss: 0.5089
+Epoch: 035/100 | Batch 0016/0043 | Averaged Loss: 0.4517
+Epoch: 035/100 | Batch 0017/0043 | Averaged Loss: 0.4550
+Epoch: 035/100 | Batch 0018/0043 | Averaged Loss: 0.4678
+Epoch: 035/100 | Batch 0019/0043 | Averaged Loss: 0.4221
+Epoch: 035/100 | Batch 0020/0043 | Averaged Loss: 0.4482
+Epoch: 035/100 | Batch 0021/0043 | Averaged Loss: 0.4489
+Epoch: 035/100 | Batch 0022/0043 | Averaged Loss: 0.4765
+Epoch: 035/100 | Batch 0023/0043 | Averaged Loss: 0.4675
+Epoch: 035/100 | Batch 0024/0043 | Averaged Loss: 0.4335
+Epoch: 035/100 | Batch 0025/0043 | Averaged Loss: 0.4234
+Epoch: 035/100 | Batch 0026/0043 | Averaged Loss: 0.4101
+Epoch: 035/100 | Batch 0027/0043 | Averaged Loss: 0.4412
+Epoch: 035/100 | Batch 0028/0043 | Averaged Loss: 0.4082
+Epoch: 035/100 | Batch 0029/0043 | Averaged Loss: 0.3665
+Epoch: 035/100 | Batch 0030/0043 | Averaged Loss: 0.4099
+Epoch: 035/100 | Batch 0031/0043 | Averaged Loss: 0.4774
+Epoch: 035/100 | Batch 0032/0043 | Averaged Loss: 0.3674
+Epoch: 035/100 | Batch 0033/0043 | Averaged Loss: 0.4283
+Epoch: 035/100 | Batch 0034/0043 | Averaged Loss: 0.4634
+Epoch: 035/100 | Batch 0035/0043 | Averaged Loss: 0.4138
+Epoch: 035/100 | Batch 0036/0043 | Averaged Loss: 0.4410
+Epoch: 035/100 | Batch 0037/0043 | Averaged Loss: 0.4630
+Epoch: 035/100 | Batch 0038/0043 | Averaged Loss: 0.4525
+Epoch: 035/100 | Batch 0039/0043 | Averaged Loss: 0.4497
+Epoch: 035/100 | Batch 0040/0043 | Averaged Loss: 0.5200
+Epoch: 035/100 | Batch 0041/0043 | Averaged Loss: 0.4078
+Epoch: 035/100 | Batch 0042/0043 | Averaged Loss: 0.4799
+Epoch: 035/100 | Train: 86.37% | Validation: 73.07%
+Time elapsed: 5.202664375305176 min
+Epoch: 036/100 | Batch 0000/0043 | Averaged Loss: 0.4163
+Epoch: 036/100 | Batch 0001/0043 | Averaged Loss: 0.4320
+Epoch: 036/100 | Batch 0002/0043 | Averaged Loss: 0.3521
+Epoch: 036/100 | Batch 0003/0043 | Averaged Loss: 0.4204
+Epoch: 036/100 | Batch 0004/0043 | Averaged Loss: 0.3460
+Epoch: 036/100 | Batch 0005/0043 | Averaged Loss: 0.3996
+Epoch: 036/100 | Batch 0006/0043 | Averaged Loss: 0.3939
+Epoch: 036/100 | Batch 0007/0043 | Averaged Loss: 0.3535
+Epoch: 036/100 | Batch 0008/0043 | Averaged Loss: 0.4081
+Epoch: 036/100 | Batch 0009/0043 | Averaged Loss: 0.3649
+Epoch: 036/100 | Batch 0010/0043 | Averaged Loss: 0.4048
+Epoch: 036/100 | Batch 0011/0043 | Averaged Loss: 0.4355
+Epoch: 036/100 | Batch 0012/0043 | Averaged Loss: 0.3602
+Epoch: 036/100 | Batch 0013/0043 | Averaged Loss: 0.4289
+Epoch: 036/100 | Batch 0014/0043 | Averaged Loss: 0.4317
+Epoch: 036/100 | Batch 0015/0043 | Averaged Loss: 0.4320
+Epoch: 036/100 | Batch 0016/0043 | Averaged Loss: 0.4357
+Epoch: 036/100 | Batch 0017/0043 | Averaged Loss: 0.4150
+Epoch: 036/100 | Batch 0018/0043 | Averaged Loss: 0.4478
+Epoch: 036/100 | Batch 0019/0043 | Averaged Loss: 0.3673
+Epoch: 036/100 | Batch 0020/0043 | Averaged Loss: 0.4196
+Epoch: 036/100 | Batch 0021/0043 | Averaged Loss: 0.4623
+Epoch: 036/100 | Batch 0022/0043 | Averaged Loss: 0.3661
+Epoch: 036/100 | Batch 0023/0043 | Averaged Loss: 0.4352
+Epoch: 036/100 | Batch 0024/0043 | Averaged Loss: 0.4283
+Epoch: 036/100 | Batch 0025/0043 | Averaged Loss: 0.3870
+Epoch: 036/100 | Batch 0026/0043 | Averaged Loss: 0.4469
+Epoch: 036/100 | Batch 0027/0043 | Averaged Loss: 0.4177
+Epoch: 036/100 | Batch 0028/0043 | Averaged Loss: 0.4562
+Epoch: 036/100 | Batch 0029/0043 | Averaged Loss: 0.4544
+Epoch: 036/100 | Batch 0030/0043 | Averaged Loss: 0.4239
+Epoch: 036/100 | Batch 0031/0043 | Averaged Loss: 0.4527
+Epoch: 036/100 | Batch 0032/0043 | Averaged Loss: 0.4087
+Epoch: 036/100 | Batch 0033/0043 | Averaged Loss: 0.4200
+Epoch: 036/100 | Batch 0034/0043 | Averaged Loss: 0.4042
+Epoch: 036/100 | Batch 0035/0043 | Averaged Loss: 0.4327
+Epoch: 036/100 | Batch 0036/0043 | Averaged Loss: 0.4322
+Epoch: 036/100 | Batch 0037/0043 | Averaged Loss: 0.4523
+Epoch: 036/100 | Batch 0038/0043 | Averaged Loss: 0.4121
+Epoch: 036/100 | Batch 0039/0043 | Averaged Loss: 0.3684
+Epoch: 036/100 | Batch 0040/0043 | Averaged Loss: 0.3977
+Epoch: 036/100 | Batch 0041/0043 | Averaged Loss: 0.3950
+Epoch: 036/100 | Batch 0042/0043 | Averaged Loss: 0.4814
+Epoch: 036/100 | Train: 87.93% | Validation: 73.88%
+Time elapsed: 5.350898265838623 min
+Epoch: 037/100 | Batch 0000/0043 | Averaged Loss: 0.3520
+Epoch: 037/100 | Batch 0001/0043 | Averaged Loss: 0.2966
+Epoch: 037/100 | Batch 0002/0043 | Averaged Loss: 0.3697
+Epoch: 037/100 | Batch 0003/0043 | Averaged Loss: 0.3779
+Epoch: 037/100 | Batch 0004/0043 | Averaged Loss: 0.3658
+Epoch: 037/100 | Batch 0005/0043 | Averaged Loss: 0.3280
+Epoch: 037/100 | Batch 0006/0043 | Averaged Loss: 0.3184
+Epoch: 037/100 | Batch 0007/0043 | Averaged Loss: 0.3212
+Epoch: 037/100 | Batch 0008/0043 | Averaged Loss: 0.3310
+Epoch: 037/100 | Batch 0009/0043 | Averaged Loss: 0.3441
+Epoch: 037/100 | Batch 0010/0043 | Averaged Loss: 0.3389
+Epoch: 037/100 | Batch 0011/0043 | Averaged Loss: 0.3551
+Epoch: 037/100 | Batch 0012/0043 | Averaged Loss: 0.3835
+Epoch: 037/100 | Batch 0013/0043 | Averaged Loss: 0.3459
+Epoch: 037/100 | Batch 0014/0043 | Averaged Loss: 0.4042
+Epoch: 037/100 | Batch 0015/0043 | Averaged Loss: 0.3565
+Epoch: 037/100 | Batch 0016/0043 | Averaged Loss: 0.3745
+Epoch: 037/100 | Batch 0017/0043 | Averaged Loss: 0.4251
+Epoch: 037/100 | Batch 0018/0043 | Averaged Loss: 0.3324
+Epoch: 037/100 | Batch 0019/0043 | Averaged Loss: 0.4005
+Epoch: 037/100 | Batch 0020/0043 | Averaged Loss: 0.4170
+Epoch: 037/100 | Batch 0021/0043 | Averaged Loss: 0.4650
+Epoch: 037/100 | Batch 0022/0043 | Averaged Loss: 0.3658
+Epoch: 037/100 | Batch 0023/0043 | Averaged Loss: 0.3865
+Epoch: 037/100 | Batch 0024/0043 | Averaged Loss: 0.4409
+Epoch: 037/100 | Batch 0025/0043 | Averaged Loss: 0.4235
+Epoch: 037/100 | Batch 0026/0043 | Averaged Loss: 0.4801
+Epoch: 037/100 | Batch 0027/0043 | Averaged Loss: 0.4225
+Epoch: 037/100 | Batch 0028/0043 | Averaged Loss: 0.3638
+Epoch: 037/100 | Batch 0029/0043 | Averaged Loss: 0.3973
+Epoch: 037/100 | Batch 0030/0043 | Averaged Loss: 0.4137
+Epoch: 037/100 | Batch 0031/0043 | Averaged Loss: 0.4196
+Epoch: 037/100 | Batch 0032/0043 | Averaged Loss: 0.4508
+Epoch: 037/100 | Batch 0033/0043 | Averaged Loss: 0.4022
+Epoch: 037/100 | Batch 0034/0043 | Averaged Loss: 0.4833
+Epoch: 037/100 | Batch 0035/0043 | Averaged Loss: 0.4275
+Epoch: 037/100 | Batch 0036/0043 | Averaged Loss: 0.3771
+Epoch: 037/100 | Batch 0037/0043 | Averaged Loss: 0.4755
+Epoch: 037/100 | Batch 0038/0043 | Averaged Loss: 0.4319
+Epoch: 037/100 | Batch 0039/0043 | Averaged Loss: 0.4554
+Epoch: 037/100 | Batch 0040/0043 | Averaged Loss: 0.3638
+Epoch: 037/100 | Batch 0041/0043 | Averaged Loss: 0.4488
+Epoch: 037/100 | Batch 0042/0043 | Averaged Loss: 0.3718
+Epoch: 037/100 | Train: 87.78% | Validation: 73.02%
+Time elapsed: 5.499383449554443 min
+Epoch: 038/100 | Batch 0000/0043 | Averaged Loss: 0.3481
+Epoch: 038/100 | Batch 0001/0043 | Averaged Loss: 0.3004
+Epoch: 038/100 | Batch 0002/0043 | Averaged Loss: 0.3513
+Epoch: 038/100 | Batch 0003/0043 | Averaged Loss: 0.2991
+Epoch: 038/100 | Batch 0004/0043 | Averaged Loss: 0.4181
+Epoch: 038/100 | Batch 0005/0043 | Averaged Loss: 0.3574
+Epoch: 038/100 | Batch 0006/0043 | Averaged Loss: 0.3562
+Epoch: 038/100 | Batch 0007/0043 | Averaged Loss: 0.3136
+Epoch: 038/100 | Batch 0008/0043 | Averaged Loss: 0.3437
+Epoch: 038/100 | Batch 0009/0043 | Averaged Loss: 0.3904
+Epoch: 038/100 | Batch 0010/0043 | Averaged Loss: 0.3537
+Epoch: 038/100 | Batch 0011/0043 | Averaged Loss: 0.3431
+Epoch: 038/100 | Batch 0012/0043 | Averaged Loss: 0.3759
+Epoch: 038/100 | Batch 0013/0043 | Averaged Loss: 0.3715
+Epoch: 038/100 | Batch 0014/0043 | Averaged Loss: 0.3296
+Epoch: 038/100 | Batch 0015/0043 | Averaged Loss: 0.3605
+Epoch: 038/100 | Batch 0016/0043 | Averaged Loss: 0.3309
+Epoch: 038/100 | Batch 0017/0043 | Averaged Loss: 0.3589
+Epoch: 038/100 | Batch 0018/0043 | Averaged Loss: 0.3376
+Epoch: 038/100 | Batch 0019/0043 | Averaged Loss: 0.3771
+Epoch: 038/100 | Batch 0020/0043 | Averaged Loss: 0.4156
+Epoch: 038/100 | Batch 0021/0043 | Averaged Loss: 0.4038
+Epoch: 038/100 | Batch 0022/0043 | Averaged Loss: 0.3550
+Epoch: 038/100 | Batch 0023/0043 | Averaged Loss: 0.3668
+Epoch: 038/100 | Batch 0024/0043 | Averaged Loss: 0.3504
+Epoch: 038/100 | Batch 0025/0043 | Averaged Loss: 0.3629
+Epoch: 038/100 | Batch 0026/0043 | Averaged Loss: 0.4381
+Epoch: 038/100 | Batch 0027/0043 | Averaged Loss: 0.3685
+Epoch: 038/100 | Batch 0028/0043 | Averaged Loss: 0.4317
+Epoch: 038/100 | Batch 0029/0043 | Averaged Loss: 0.3569
+Epoch: 038/100 | Batch 0030/0043 | Averaged Loss: 0.3725
+Epoch: 038/100 | Batch 0031/0043 | Averaged Loss: 0.4067
+Epoch: 038/100 | Batch 0032/0043 | Averaged Loss: 0.3893
+Epoch: 038/100 | Batch 0033/0043 | Averaged Loss: 0.3486
+Epoch: 038/100 | Batch 0034/0043 | Averaged Loss: 0.3790
+Epoch: 038/100 | Batch 0035/0043 | Averaged Loss: 0.3771
+Epoch: 038/100 | Batch 0036/0043 | Averaged Loss: 0.3760
+Epoch: 038/100 | Batch 0037/0043 | Averaged Loss: 0.4185
+Epoch: 038/100 | Batch 0038/0043 | Averaged Loss: 0.3396
+Epoch: 038/100 | Batch 0039/0043 | Averaged Loss: 0.3888
+Epoch: 038/100 | Batch 0040/0043 | Averaged Loss: 0.3754
+Epoch: 038/100 | Batch 0041/0043 | Averaged Loss: 0.3739
+Epoch: 038/100 | Batch 0042/0043 | Averaged Loss: 0.4311
+Epoch: 038/100 | Train: 88.86% | Validation: 73.29%
+Time elapsed: 5.647883415222168 min
+Epoch: 039/100 | Batch 0000/0043 | Averaged Loss: 0.3139
+Epoch: 039/100 | Batch 0001/0043 | Averaged Loss: 0.3284
+Epoch: 039/100 | Batch 0002/0043 | Averaged Loss: 0.3619
+Epoch: 039/100 | Batch 0003/0043 | Averaged Loss: 0.3469
+Epoch: 039/100 | Batch 0004/0043 | Averaged Loss: 0.3148
+Epoch: 039/100 | Batch 0005/0043 | Averaged Loss: 0.3150
+Epoch: 039/100 | Batch 0006/0043 | Averaged Loss: 0.3529
+Epoch: 039/100 | Batch 0007/0043 | Averaged Loss: 0.3019
+Epoch: 039/100 | Batch 0008/0043 | Averaged Loss: 0.3173
+Epoch: 039/100 | Batch 0009/0043 | Averaged Loss: 0.3725
+Epoch: 039/100 | Batch 0010/0043 | Averaged Loss: 0.3399
+Epoch: 039/100 | Batch 0011/0043 | Averaged Loss: 0.3435
+Epoch: 039/100 | Batch 0012/0043 | Averaged Loss: 0.3221
+Epoch: 039/100 | Batch 0013/0043 | Averaged Loss: 0.3458
+Epoch: 039/100 | Batch 0014/0043 | Averaged Loss: 0.3509
+Epoch: 039/100 | Batch 0015/0043 | Averaged Loss: 0.3767
+Epoch: 039/100 | Batch 0016/0043 | Averaged Loss: 0.3631
+Epoch: 039/100 | Batch 0017/0043 | Averaged Loss: 0.3267
+Epoch: 039/100 | Batch 0018/0043 | Averaged Loss: 0.2927
+Epoch: 039/100 | Batch 0019/0043 | Averaged Loss: 0.3077
+Epoch: 039/100 | Batch 0020/0043 | Averaged Loss: 0.2717
+Epoch: 039/100 | Batch 0021/0043 | Averaged Loss: 0.3260
+Epoch: 039/100 | Batch 0022/0043 | Averaged Loss: 0.2984
+Epoch: 039/100 | Batch 0023/0043 | Averaged Loss: 0.3666
+Epoch: 039/100 | Batch 0024/0043 | Averaged Loss: 0.2875
+Epoch: 039/100 | Batch 0025/0043 | Averaged Loss: 0.3314
+Epoch: 039/100 | Batch 0026/0043 | Averaged Loss: 0.3757
+Epoch: 039/100 | Batch 0027/0043 | Averaged Loss: 0.4235
+Epoch: 039/100 | Batch 0028/0043 | Averaged Loss: 0.3456
+Epoch: 039/100 | Batch 0029/0043 | Averaged Loss: 0.3873
+Epoch: 039/100 | Batch 0030/0043 | Averaged Loss: 0.3323
+Epoch: 039/100 | Batch 0031/0043 | Averaged Loss: 0.4320
+Epoch: 039/100 | Batch 0032/0043 | Averaged Loss: 0.4239
+Epoch: 039/100 | Batch 0033/0043 | Averaged Loss: 0.3918
+Epoch: 039/100 | Batch 0034/0043 | Averaged Loss: 0.4059
+Epoch: 039/100 | Batch 0035/0043 | Averaged Loss: 0.3508
+Epoch: 039/100 | Batch 0036/0043 | Averaged Loss: 0.3775
+Epoch: 039/100 | Batch 0037/0043 | Averaged Loss: 0.3858
+Epoch: 039/100 | Batch 0038/0043 | Averaged Loss: 0.3283
+Epoch: 039/100 | Batch 0039/0043 | Averaged Loss: 0.3402
+Epoch: 039/100 | Batch 0040/0043 | Averaged Loss: 0.3668
+Epoch: 039/100 | Batch 0041/0043 | Averaged Loss: 0.3446
+Epoch: 039/100 | Batch 0042/0043 | Averaged Loss: 0.3522
+Epoch: 039/100 | Train: 89.86% | Validation: 74.37%
+Time elapsed: 5.796290874481201 min
+Epoch: 040/100 | Batch 0000/0043 | Averaged Loss: 0.2937
+Epoch: 040/100 | Batch 0001/0043 | Averaged Loss: 0.2766
+Epoch: 040/100 | Batch 0002/0043 | Averaged Loss: 0.3311
+Epoch: 040/100 | Batch 0003/0043 | Averaged Loss: 0.2936
+Epoch: 040/100 | Batch 0004/0043 | Averaged Loss: 0.3061
+Epoch: 040/100 | Batch 0005/0043 | Averaged Loss: 0.3056
+Epoch: 040/100 | Batch 0006/0043 | Averaged Loss: 0.2528
+Epoch: 040/100 | Batch 0007/0043 | Averaged Loss: 0.2555
+Epoch: 040/100 | Batch 0008/0043 | Averaged Loss: 0.3182
+Epoch: 040/100 | Batch 0009/0043 | Averaged Loss: 0.2760
+Epoch: 040/100 | Batch 0010/0043 | Averaged Loss: 0.3702
+Epoch: 040/100 | Batch 0011/0043 | Averaged Loss: 0.3041
+Epoch: 040/100 | Batch 0012/0043 | Averaged Loss: 0.3070
+Epoch: 040/100 | Batch 0013/0043 | Averaged Loss: 0.3143
+Epoch: 040/100 | Batch 0014/0043 | Averaged Loss: 0.2910
+Epoch: 040/100 | Batch 0015/0043 | Averaged Loss: 0.3307
+Epoch: 040/100 | Batch 0016/0043 | Averaged Loss: 0.2701
+Epoch: 040/100 | Batch 0017/0043 | Averaged Loss: 0.3861
+Epoch: 040/100 | Batch 0018/0043 | Averaged Loss: 0.2890
+Epoch: 040/100 | Batch 0019/0043 | Averaged Loss: 0.3662
+Epoch: 040/100 | Batch 0020/0043 | Averaged Loss: 0.3398
+Epoch: 040/100 | Batch 0021/0043 | Averaged Loss: 0.3246
+Epoch: 040/100 | Batch 0022/0043 | Averaged Loss: 0.3213
+Epoch: 040/100 | Batch 0023/0043 | Averaged Loss: 0.3340
+Epoch: 040/100 | Batch 0024/0043 | Averaged Loss: 0.3618
+Epoch: 040/100 | Batch 0025/0043 | Averaged Loss: 0.3314
+Epoch: 040/100 | Batch 0026/0043 | Averaged Loss: 0.3587
+Epoch: 040/100 | Batch 0027/0043 | Averaged Loss: 0.4005
+Epoch: 040/100 | Batch 0028/0043 | Averaged Loss: 0.3402
+Epoch: 040/100 | Batch 0029/0043 | Averaged Loss: 0.3523
+Epoch: 040/100 | Batch 0030/0043 | Averaged Loss: 0.2859
+Epoch: 040/100 | Batch 0031/0043 | Averaged Loss: 0.3446
+Epoch: 040/100 | Batch 0032/0043 | Averaged Loss: 0.3185
+Epoch: 040/100 | Batch 0033/0043 | Averaged Loss: 0.3747
+Epoch: 040/100 | Batch 0034/0043 | Averaged Loss: 0.3663
+Epoch: 040/100 | Batch 0035/0043 | Averaged Loss: 0.3764
+Epoch: 040/100 | Batch 0036/0043 | Averaged Loss: 0.3662
+Epoch: 040/100 | Batch 0037/0043 | Averaged Loss: 0.3622
+Epoch: 040/100 | Batch 0038/0043 | Averaged Loss: 0.4029
+Epoch: 040/100 | Batch 0039/0043 | Averaged Loss: 0.3144
+Epoch: 040/100 | Batch 0040/0043 | Averaged Loss: 0.3404
+Epoch: 040/100 | Batch 0041/0043 | Averaged Loss: 0.3749
+Epoch: 040/100 | Batch 0042/0043 | Averaged Loss: 0.3479
+Epoch: 040/100 | Train: 88.96% | Validation: 72.83%
+Time elapsed: 5.944502830505371 min
+Epoch: 041/100 | Batch 0000/0043 | Averaged Loss: 0.3258
+Epoch: 041/100 | Batch 0001/0043 | Averaged Loss: 0.3138
+Epoch: 041/100 | Batch 0002/0043 | Averaged Loss: 0.3103
+Epoch: 041/100 | Batch 0003/0043 | Averaged Loss: 0.2468
+Epoch: 041/100 | Batch 0004/0043 | Averaged Loss: 0.3541
+Epoch: 041/100 | Batch 0005/0043 | Averaged Loss: 0.2865
+Epoch: 041/100 | Batch 0006/0043 | Averaged Loss: 0.2964
+Epoch: 041/100 | Batch 0007/0043 | Averaged Loss: 0.3067
+Epoch: 041/100 | Batch 0008/0043 | Averaged Loss: 0.2612
+Epoch: 041/100 | Batch 0009/0043 | Averaged Loss: 0.2824
+Epoch: 041/100 | Batch 0010/0043 | Averaged Loss: 0.2568
+Epoch: 041/100 | Batch 0011/0043 | Averaged Loss: 0.3478
+Epoch: 041/100 | Batch 0012/0043 | Averaged Loss: 0.2517
+Epoch: 041/100 | Batch 0013/0043 | Averaged Loss: 0.2928
+Epoch: 041/100 | Batch 0014/0043 | Averaged Loss: 0.3249
+Epoch: 041/100 | Batch 0015/0043 | Averaged Loss: 0.3352
+Epoch: 041/100 | Batch 0016/0043 | Averaged Loss: 0.2488
+Epoch: 041/100 | Batch 0017/0043 | Averaged Loss: 0.2938
+Epoch: 041/100 | Batch 0018/0043 | Averaged Loss: 0.2672
+Epoch: 041/100 | Batch 0019/0043 | Averaged Loss: 0.3064
+Epoch: 041/100 | Batch 0020/0043 | Averaged Loss: 0.3052
+Epoch: 041/100 | Batch 0021/0043 | Averaged Loss: 0.3345
+Epoch: 041/100 | Batch 0022/0043 | Averaged Loss: 0.3406
+Epoch: 041/100 | Batch 0023/0043 | Averaged Loss: 0.2683
+Epoch: 041/100 | Batch 0024/0043 | Averaged Loss: 0.3143
+Epoch: 041/100 | Batch 0025/0043 | Averaged Loss: 0.3515
+Epoch: 041/100 | Batch 0026/0043 | Averaged Loss: 0.3139
+Epoch: 041/100 | Batch 0027/0043 | Averaged Loss: 0.3050
+Epoch: 041/100 | Batch 0028/0043 | Averaged Loss: 0.3056
+Epoch: 041/100 | Batch 0029/0043 | Averaged Loss: 0.3320
+Epoch: 041/100 | Batch 0030/0043 | Averaged Loss: 0.3455
+Epoch: 041/100 | Batch 0031/0043 | Averaged Loss: 0.3150
+Epoch: 041/100 | Batch 0032/0043 | Averaged Loss: 0.3434
+Epoch: 041/100 | Batch 0033/0043 | Averaged Loss: 0.3207
+Epoch: 041/100 | Batch 0034/0043 | Averaged Loss: 0.3558
+Epoch: 041/100 | Batch 0035/0043 | Averaged Loss: 0.3136
+Epoch: 041/100 | Batch 0036/0043 | Averaged Loss: 0.3714
+Epoch: 041/100 | Batch 0037/0043 | Averaged Loss: 0.3878
+Epoch: 041/100 | Batch 0038/0043 | Averaged Loss: 0.3158
+Epoch: 041/100 | Batch 0039/0043 | Averaged Loss: 0.3396
+Epoch: 041/100 | Batch 0040/0043 | Averaged Loss: 0.3814
+Epoch: 041/100 | Batch 0041/0043 | Averaged Loss: 0.3453
+Epoch: 041/100 | Batch 0042/0043 | Averaged Loss: 0.3676
+Epoch: 041/100 | Train: 90.80% | Validation: 73.54%
+Time elapsed: 6.092668056488037 min
+Epoch: 042/100 | Batch 0000/0043 | Averaged Loss: 0.2699
+Epoch: 042/100 | Batch 0001/0043 | Averaged Loss: 0.3352
+Epoch: 042/100 | Batch 0002/0043 | Averaged Loss: 0.2782
+Epoch: 042/100 | Batch 0003/0043 | Averaged Loss: 0.2915
+Epoch: 042/100 | Batch 0004/0043 | Averaged Loss: 0.2935
+Epoch: 042/100 | Batch 0005/0043 | Averaged Loss: 0.2759
+Epoch: 042/100 | Batch 0006/0043 | Averaged Loss: 0.2592
+Epoch: 042/100 | Batch 0007/0043 | Averaged Loss: 0.3080
+Epoch: 042/100 | Batch 0008/0043 | Averaged Loss: 0.2787
+Epoch: 042/100 | Batch 0009/0043 | Averaged Loss: 0.2813
+Epoch: 042/100 | Batch 0010/0043 | Averaged Loss: 0.2820
+Epoch: 042/100 | Batch 0011/0043 | Averaged Loss: 0.2681
+Epoch: 042/100 | Batch 0012/0043 | Averaged Loss: 0.2787
+Epoch: 042/100 | Batch 0013/0043 | Averaged Loss: 0.2742
+Epoch: 042/100 | Batch 0014/0043 | Averaged Loss: 0.2433
+Epoch: 042/100 | Batch 0015/0043 | Averaged Loss: 0.2579
+Epoch: 042/100 | Batch 0016/0043 | Averaged Loss: 0.2831
+Epoch: 042/100 | Batch 0017/0043 | Averaged Loss: 0.3097
+Epoch: 042/100 | Batch 0018/0043 | Averaged Loss: 0.2745
+Epoch: 042/100 | Batch 0019/0043 | Averaged Loss: 0.2789
+Epoch: 042/100 | Batch 0020/0043 | Averaged Loss: 0.2482
+Epoch: 042/100 | Batch 0021/0043 | Averaged Loss: 0.3063
+Epoch: 042/100 | Batch 0022/0043 | Averaged Loss: 0.2891
+Epoch: 042/100 | Batch 0023/0043 | Averaged Loss: 0.2962
+Epoch: 042/100 | Batch 0024/0043 | Averaged Loss: 0.3227
+Epoch: 042/100 | Batch 0025/0043 | Averaged Loss: 0.2928
+Epoch: 042/100 | Batch 0026/0043 | Averaged Loss: 0.3042
+Epoch: 042/100 | Batch 0027/0043 | Averaged Loss: 0.3349
+Epoch: 042/100 | Batch 0028/0043 | Averaged Loss: 0.3360
+Epoch: 042/100 | Batch 0029/0043 | Averaged Loss: 0.2871
+Epoch: 042/100 | Batch 0030/0043 | Averaged Loss: 0.2897
+Epoch: 042/100 | Batch 0031/0043 | Averaged Loss: 0.2433
+Epoch: 042/100 | Batch 0032/0043 | Averaged Loss: 0.2913
+Epoch: 042/100 | Batch 0033/0043 | Averaged Loss: 0.3015
+Epoch: 042/100 | Batch 0034/0043 | Averaged Loss: 0.2840
+Epoch: 042/100 | Batch 0035/0043 | Averaged Loss: 0.3404
+Epoch: 042/100 | Batch 0036/0043 | Averaged Loss: 0.2979
+Epoch: 042/100 | Batch 0037/0043 | Averaged Loss: 0.3749
+Epoch: 042/100 | Batch 0038/0043 | Averaged Loss: 0.3156
+Epoch: 042/100 | Batch 0039/0043 | Averaged Loss: 0.3618
+Epoch: 042/100 | Batch 0040/0043 | Averaged Loss: 0.3493
+Epoch: 042/100 | Batch 0041/0043 | Averaged Loss: 0.3027
+Epoch: 042/100 | Batch 0042/0043 | Averaged Loss: 0.3815
+Epoch: 042/100 | Train: 91.33% | Validation: 73.71%
+Time elapsed: 6.2409491539001465 min
+Epoch: 043/100 | Batch 0000/0043 | Averaged Loss: 0.2843
+Epoch: 043/100 | Batch 0001/0043 | Averaged Loss: 0.2919
+Epoch: 043/100 | Batch 0002/0043 | Averaged Loss: 0.2478
+Epoch: 043/100 | Batch 0003/0043 | Averaged Loss: 0.2781
+Epoch: 043/100 | Batch 0004/0043 | Averaged Loss: 0.2800
+Epoch: 043/100 | Batch 0005/0043 | Averaged Loss: 0.2669
+Epoch: 043/100 | Batch 0006/0043 | Averaged Loss: 0.2894
+Epoch: 043/100 | Batch 0007/0043 | Averaged Loss: 0.3108
+Epoch: 043/100 | Batch 0008/0043 | Averaged Loss: 0.2388
+Epoch: 043/100 | Batch 0009/0043 | Averaged Loss: 0.2337
+Epoch: 043/100 | Batch 0010/0043 | Averaged Loss: 0.2454
+Epoch: 043/100 | Batch 0011/0043 | Averaged Loss: 0.2195
+Epoch: 043/100 | Batch 0012/0043 | Averaged Loss: 0.2869
+Epoch: 043/100 | Batch 0013/0043 | Averaged Loss: 0.2529
+Epoch: 043/100 | Batch 0014/0043 | Averaged Loss: 0.2939
+Epoch: 043/100 | Batch 0015/0043 | Averaged Loss: 0.2636
+Epoch: 043/100 | Batch 0016/0043 | Averaged Loss: 0.2680
+Epoch: 043/100 | Batch 0017/0043 | Averaged Loss: 0.2758
+Epoch: 043/100 | Batch 0018/0043 | Averaged Loss: 0.2629
+Epoch: 043/100 | Batch 0019/0043 | Averaged Loss: 0.2500
+Epoch: 043/100 | Batch 0020/0043 | Averaged Loss: 0.2118
+Epoch: 043/100 | Batch 0021/0043 | Averaged Loss: 0.2847
+Epoch: 043/100 | Batch 0022/0043 | Averaged Loss: 0.2626
+Epoch: 043/100 | Batch 0023/0043 | Averaged Loss: 0.2943
+Epoch: 043/100 | Batch 0024/0043 | Averaged Loss: 0.3008
+Epoch: 043/100 | Batch 0025/0043 | Averaged Loss: 0.2818
+Epoch: 043/100 | Batch 0026/0043 | Averaged Loss: 0.2682
+Epoch: 043/100 | Batch 0027/0043 | Averaged Loss: 0.3012
+Epoch: 043/100 | Batch 0028/0043 | Averaged Loss: 0.2553
+Epoch: 043/100 | Batch 0029/0043 | Averaged Loss: 0.2474
+Epoch: 043/100 | Batch 0030/0043 | Averaged Loss: 0.3273
+Epoch: 043/100 | Batch 0031/0043 | Averaged Loss: 0.2644
+Epoch: 043/100 | Batch 0032/0043 | Averaged Loss: 0.2611
+Epoch: 043/100 | Batch 0033/0043 | Averaged Loss: 0.3130
+Epoch: 043/100 | Batch 0034/0043 | Averaged Loss: 0.2854
+Epoch: 043/100 | Batch 0035/0043 | Averaged Loss: 0.2850
+Epoch: 043/100 | Batch 0036/0043 | Averaged Loss: 0.3280
+Epoch: 043/100 | Batch 0037/0043 | Averaged Loss: 0.3073
+Epoch: 043/100 | Batch 0038/0043 | Averaged Loss: 0.2657
+Epoch: 043/100 | Batch 0039/0043 | Averaged Loss: 0.3275
+Epoch: 043/100 | Batch 0040/0043 | Averaged Loss: 0.2618
+Epoch: 043/100 | Batch 0041/0043 | Averaged Loss: 0.3055
+Epoch: 043/100 | Batch 0042/0043 | Averaged Loss: 0.2653
+Epoch: 043/100 | Train: 92.40% | Validation: 75.02%
+Time elapsed: 6.389198303222656 min
+Epoch: 044/100 | Batch 0000/0043 | Averaged Loss: 0.1876
+Epoch: 044/100 | Batch 0001/0043 | Averaged Loss: 0.1885
+Epoch: 044/100 | Batch 0002/0043 | Averaged Loss: 0.2422
+Epoch: 044/100 | Batch 0003/0043 | Averaged Loss: 0.2084
+Epoch: 044/100 | Batch 0004/0043 | Averaged Loss: 0.1958
+Epoch: 044/100 | Batch 0005/0043 | Averaged Loss: 0.2404
+Epoch: 044/100 | Batch 0006/0043 | Averaged Loss: 0.2210
+Epoch: 044/100 | Batch 0007/0043 | Averaged Loss: 0.2372
+Epoch: 044/100 | Batch 0008/0043 | Averaged Loss: 0.2380
+Epoch: 044/100 | Batch 0009/0043 | Averaged Loss: 0.1815
+Epoch: 044/100 | Batch 0010/0043 | Averaged Loss: 0.2132
+Epoch: 044/100 | Batch 0011/0043 | Averaged Loss: 0.2104
+Epoch: 044/100 | Batch 0012/0043 | Averaged Loss: 0.2227
+Epoch: 044/100 | Batch 0013/0043 | Averaged Loss: 0.2017
+Epoch: 044/100 | Batch 0014/0043 | Averaged Loss: 0.2454
+Epoch: 044/100 | Batch 0015/0043 | Averaged Loss: 0.2549
+Epoch: 044/100 | Batch 0016/0043 | Averaged Loss: 0.2522
+Epoch: 044/100 | Batch 0017/0043 | Averaged Loss: 0.2340
+Epoch: 044/100 | Batch 0018/0043 | Averaged Loss: 0.2570
+Epoch: 044/100 | Batch 0019/0043 | Averaged Loss: 0.2294
+Epoch: 044/100 | Batch 0020/0043 | Averaged Loss: 0.2011
+Epoch: 044/100 | Batch 0021/0043 | Averaged Loss: 0.2174
+Epoch: 044/100 | Batch 0022/0043 | Averaged Loss: 0.2516
+Epoch: 044/100 | Batch 0023/0043 | Averaged Loss: 0.2040
+Epoch: 044/100 | Batch 0024/0043 | Averaged Loss: 0.2599
+Epoch: 044/100 | Batch 0025/0043 | Averaged Loss: 0.2497
+Epoch: 044/100 | Batch 0026/0043 | Averaged Loss: 0.2192
+Epoch: 044/100 | Batch 0027/0043 | Averaged Loss: 0.2583
+Epoch: 044/100 | Batch 0028/0043 | Averaged Loss: 0.2560
+Epoch: 044/100 | Batch 0029/0043 | Averaged Loss: 0.2419
+Epoch: 044/100 | Batch 0030/0043 | Averaged Loss: 0.2515
+Epoch: 044/100 | Batch 0031/0043 | Averaged Loss: 0.2582
+Epoch: 044/100 | Batch 0032/0043 | Averaged Loss: 0.3194
+Epoch: 044/100 | Batch 0033/0043 | Averaged Loss: 0.2559
+Epoch: 044/100 | Batch 0034/0043 | Averaged Loss: 0.2339
+Epoch: 044/100 | Batch 0035/0043 | Averaged Loss: 0.2973
+Epoch: 044/100 | Batch 0036/0043 | Averaged Loss: 0.2850
+Epoch: 044/100 | Batch 0037/0043 | Averaged Loss: 0.2308
+Epoch: 044/100 | Batch 0038/0043 | Averaged Loss: 0.3065
+Epoch: 044/100 | Batch 0039/0043 | Averaged Loss: 0.2884
+Epoch: 044/100 | Batch 0040/0043 | Averaged Loss: 0.3162
+Epoch: 044/100 | Batch 0041/0043 | Averaged Loss: 0.2536
+Epoch: 044/100 | Batch 0042/0043 | Averaged Loss: 0.3117
+Epoch: 044/100 | Train: 92.58% | Validation: 74.32%
+Time elapsed: 6.53717565536499 min
+Epoch: 045/100 | Batch 0000/0043 | Averaged Loss: 0.2506
+Epoch: 045/100 | Batch 0001/0043 | Averaged Loss: 0.2043
+Epoch: 045/100 | Batch 0002/0043 | Averaged Loss: 0.2217
+Epoch: 045/100 | Batch 0003/0043 | Averaged Loss: 0.1959
+Epoch: 045/100 | Batch 0004/0043 | Averaged Loss: 0.2269
+Epoch: 045/100 | Batch 0005/0043 | Averaged Loss: 0.1655
+Epoch: 045/100 | Batch 0006/0043 | Averaged Loss: 0.1945
+Epoch: 045/100 | Batch 0007/0043 | Averaged Loss: 0.2454
+Epoch: 045/100 | Batch 0008/0043 | Averaged Loss: 0.1726
+Epoch: 045/100 | Batch 0009/0043 | Averaged Loss: 0.2740
+Epoch: 045/100 | Batch 0010/0043 | Averaged Loss: 0.2029
+Epoch: 045/100 | Batch 0011/0043 | Averaged Loss: 0.2244
+Epoch: 045/100 | Batch 0012/0043 | Averaged Loss: 0.2121
+Epoch: 045/100 | Batch 0013/0043 | Averaged Loss: 0.2262
+Epoch: 045/100 | Batch 0014/0043 | Averaged Loss: 0.2789
+Epoch: 045/100 | Batch 0015/0043 | Averaged Loss: 0.2269
+Epoch: 045/100 | Batch 0016/0043 | Averaged Loss: 0.2082
+Epoch: 045/100 | Batch 0017/0043 | Averaged Loss: 0.2239
+Epoch: 045/100 | Batch 0018/0043 | Averaged Loss: 0.2197
+Epoch: 045/100 | Batch 0019/0043 | Averaged Loss: 0.2123
+Epoch: 045/100 | Batch 0020/0043 | Averaged Loss: 0.2330
+Epoch: 045/100 | Batch 0021/0043 | Averaged Loss: 0.2656
+Epoch: 045/100 | Batch 0022/0043 | Averaged Loss: 0.2326
+Epoch: 045/100 | Batch 0023/0043 | Averaged Loss: 0.2353
+Epoch: 045/100 | Batch 0024/0043 | Averaged Loss: 0.2621
+Epoch: 045/100 | Batch 0025/0043 | Averaged Loss: 0.2062
+Epoch: 045/100 | Batch 0026/0043 | Averaged Loss: 0.2366
+Epoch: 045/100 | Batch 0027/0043 | Averaged Loss: 0.2357
+Epoch: 045/100 | Batch 0028/0043 | Averaged Loss: 0.2636
+Epoch: 045/100 | Batch 0029/0043 | Averaged Loss: 0.1855
+Epoch: 045/100 | Batch 0030/0043 | Averaged Loss: 0.2743
+Epoch: 045/100 | Batch 0031/0043 | Averaged Loss: 0.1915
+Epoch: 045/100 | Batch 0032/0043 | Averaged Loss: 0.2583
+Epoch: 045/100 | Batch 0033/0043 | Averaged Loss: 0.2415
+Epoch: 045/100 | Batch 0034/0043 | Averaged Loss: 0.2121
+Epoch: 045/100 | Batch 0035/0043 | Averaged Loss: 0.3035
+Epoch: 045/100 | Batch 0036/0043 | Averaged Loss: 0.2683
+Epoch: 045/100 | Batch 0037/0043 | Averaged Loss: 0.2865
+Epoch: 045/100 | Batch 0038/0043 | Averaged Loss: 0.2701
+Epoch: 045/100 | Batch 0039/0043 | Averaged Loss: 0.3176
+Epoch: 045/100 | Batch 0040/0043 | Averaged Loss: 0.2897
+Epoch: 045/100 | Batch 0041/0043 | Averaged Loss: 0.2457
+Epoch: 045/100 | Batch 0042/0043 | Averaged Loss: 0.2740
+Epoch: 045/100 | Train: 93.33% | Validation: 74.78%
+Time elapsed: 6.685224533081055 min
+Epoch: 046/100 | Batch 0000/0043 | Averaged Loss: 0.2164
+Epoch: 046/100 | Batch 0001/0043 | Averaged Loss: 0.1996
+Epoch: 046/100 | Batch 0002/0043 | Averaged Loss: 0.2150
+Epoch: 046/100 | Batch 0003/0043 | Averaged Loss: 0.2055
+Epoch: 046/100 | Batch 0004/0043 | Averaged Loss: 0.2103
+Epoch: 046/100 | Batch 0005/0043 | Averaged Loss: 0.2141
+Epoch: 046/100 | Batch 0006/0043 | Averaged Loss: 0.2379
+Epoch: 046/100 | Batch 0007/0043 | Averaged Loss: 0.1963
+Epoch: 046/100 | Batch 0008/0043 | Averaged Loss: 0.1917
+Epoch: 046/100 | Batch 0009/0043 | Averaged Loss: 0.2346
+Epoch: 046/100 | Batch 0010/0043 | Averaged Loss: 0.1737
+Epoch: 046/100 | Batch 0011/0043 | Averaged Loss: 0.2278
+Epoch: 046/100 | Batch 0012/0043 | Averaged Loss: 0.2080
+Epoch: 046/100 | Batch 0013/0043 | Averaged Loss: 0.2292
+Epoch: 046/100 | Batch 0014/0043 | Averaged Loss: 0.1913
+Epoch: 046/100 | Batch 0015/0043 | Averaged Loss: 0.2316
+Epoch: 046/100 | Batch 0016/0043 | Averaged Loss: 0.2221
+Epoch: 046/100 | Batch 0017/0043 | Averaged Loss: 0.2125
+Epoch: 046/100 | Batch 0018/0043 | Averaged Loss: 0.2268
+Epoch: 046/100 | Batch 0019/0043 | Averaged Loss: 0.2066
+Epoch: 046/100 | Batch 0020/0043 | Averaged Loss: 0.1895
+Epoch: 046/100 | Batch 0021/0043 | Averaged Loss: 0.1915
+Epoch: 046/100 | Batch 0022/0043 | Averaged Loss: 0.2533
+Epoch: 046/100 | Batch 0023/0043 | Averaged Loss: 0.2618
+Epoch: 046/100 | Batch 0024/0043 | Averaged Loss: 0.2500
+Epoch: 046/100 | Batch 0025/0043 | Averaged Loss: 0.2099
+Epoch: 046/100 | Batch 0026/0043 | Averaged Loss: 0.2116
+Epoch: 046/100 | Batch 0027/0043 | Averaged Loss: 0.2494
+Epoch: 046/100 | Batch 0028/0043 | Averaged Loss: 0.2033
+Epoch: 046/100 | Batch 0029/0043 | Averaged Loss: 0.2287
+Epoch: 046/100 | Batch 0030/0043 | Averaged Loss: 0.2372
+Epoch: 046/100 | Batch 0031/0043 | Averaged Loss: 0.2141
+Epoch: 046/100 | Batch 0032/0043 | Averaged Loss: 0.2422
+Epoch: 046/100 | Batch 0033/0043 | Averaged Loss: 0.2367
+Epoch: 046/100 | Batch 0034/0043 | Averaged Loss: 0.2129
+Epoch: 046/100 | Batch 0035/0043 | Averaged Loss: 0.2260
+Epoch: 046/100 | Batch 0036/0043 | Averaged Loss: 0.2340
+Epoch: 046/100 | Batch 0037/0043 | Averaged Loss: 0.2159
+Epoch: 046/100 | Batch 0038/0043 | Averaged Loss: 0.2798
+Epoch: 046/100 | Batch 0039/0043 | Averaged Loss: 0.2319
+Epoch: 046/100 | Batch 0040/0043 | Averaged Loss: 0.2361
+Epoch: 046/100 | Batch 0041/0043 | Averaged Loss: 0.3125
+Epoch: 046/100 | Batch 0042/0043 | Averaged Loss: 0.2860
+Epoch: 046/100 | Train: 93.68% | Validation: 74.37%
+Time elapsed: 6.8335280418396 min
+Epoch: 047/100 | Batch 0000/0043 | Averaged Loss: 0.2004
+Epoch: 047/100 | Batch 0001/0043 | Averaged Loss: 0.2014
+Epoch: 047/100 | Batch 0002/0043 | Averaged Loss: 0.2021
+Epoch: 047/100 | Batch 0003/0043 | Averaged Loss: 0.2016
+Epoch: 047/100 | Batch 0004/0043 | Averaged Loss: 0.2448
+Epoch: 047/100 | Batch 0005/0043 | Averaged Loss: 0.2050
+Epoch: 047/100 | Batch 0006/0043 | Averaged Loss: 0.2041
+Epoch: 047/100 | Batch 0007/0043 | Averaged Loss: 0.2490
+Epoch: 047/100 | Batch 0008/0043 | Averaged Loss: 0.1717
+Epoch: 047/100 | Batch 0009/0043 | Averaged Loss: 0.2114
+Epoch: 047/100 | Batch 0010/0043 | Averaged Loss: 0.2873
+Epoch: 047/100 | Batch 0011/0043 | Averaged Loss: 0.2033
+Epoch: 047/100 | Batch 0012/0043 | Averaged Loss: 0.2683
+Epoch: 047/100 | Batch 0013/0043 | Averaged Loss: 0.2249
+Epoch: 047/100 | Batch 0014/0043 | Averaged Loss: 0.1933
+Epoch: 047/100 | Batch 0015/0043 | Averaged Loss: 0.2251
+Epoch: 047/100 | Batch 0016/0043 | Averaged Loss: 0.2106
+Epoch: 047/100 | Batch 0017/0043 | Averaged Loss: 0.2074
+Epoch: 047/100 | Batch 0018/0043 | Averaged Loss: 0.1925
+Epoch: 047/100 | Batch 0019/0043 | Averaged Loss: 0.2123
+Epoch: 047/100 | Batch 0020/0043 | Averaged Loss: 0.2052
+Epoch: 047/100 | Batch 0021/0043 | Averaged Loss: 0.2544
+Epoch: 047/100 | Batch 0022/0043 | Averaged Loss: 0.1854
+Epoch: 047/100 | Batch 0023/0043 | Averaged Loss: 0.2268
+Epoch: 047/100 | Batch 0024/0043 | Averaged Loss: 0.2337
+Epoch: 047/100 | Batch 0025/0043 | Averaged Loss: 0.1900
+Epoch: 047/100 | Batch 0026/0043 | Averaged Loss: 0.2422
+Epoch: 047/100 | Batch 0027/0043 | Averaged Loss: 0.2497
+Epoch: 047/100 | Batch 0028/0043 | Averaged Loss: 0.2237
+Epoch: 047/100 | Batch 0029/0043 | Averaged Loss: 0.2692
+Epoch: 047/100 | Batch 0030/0043 | Averaged Loss: 0.2404
+Epoch: 047/100 | Batch 0031/0043 | Averaged Loss: 0.2359
+Epoch: 047/100 | Batch 0032/0043 | Averaged Loss: 0.2024
+Epoch: 047/100 | Batch 0033/0043 | Averaged Loss: 0.2313
+Epoch: 047/100 | Batch 0034/0043 | Averaged Loss: 0.2182
+Epoch: 047/100 | Batch 0035/0043 | Averaged Loss: 0.2571
+Epoch: 047/100 | Batch 0036/0043 | Averaged Loss: 0.1836
+Epoch: 047/100 | Batch 0037/0043 | Averaged Loss: 0.2333
+Epoch: 047/100 | Batch 0038/0043 | Averaged Loss: 0.2012
+Epoch: 047/100 | Batch 0039/0043 | Averaged Loss: 0.2445
+Epoch: 047/100 | Batch 0040/0043 | Averaged Loss: 0.1817
+Epoch: 047/100 | Batch 0041/0043 | Averaged Loss: 0.1952
+Epoch: 047/100 | Batch 0042/0043 | Averaged Loss: 0.2491
+Epoch: 047/100 | Train: 93.93% | Validation: 74.83%
+Time elapsed: 6.981990814208984 min
+Epoch: 048/100 | Batch 0000/0043 | Averaged Loss: 0.1890
+Epoch: 048/100 | Batch 0001/0043 | Averaged Loss: 0.2061
+Epoch: 048/100 | Batch 0002/0043 | Averaged Loss: 0.1782
+Epoch: 048/100 | Batch 0003/0043 | Averaged Loss: 0.1743
+Epoch: 048/100 | Batch 0004/0043 | Averaged Loss: 0.1819
+Epoch: 048/100 | Batch 0005/0043 | Averaged Loss: 0.1426
+Epoch: 048/100 | Batch 0006/0043 | Averaged Loss: 0.1910
+Epoch: 048/100 | Batch 0007/0043 | Averaged Loss: 0.1717
+Epoch: 048/100 | Batch 0008/0043 | Averaged Loss: 0.1609
+Epoch: 048/100 | Batch 0009/0043 | Averaged Loss: 0.1883
+Epoch: 048/100 | Batch 0010/0043 | Averaged Loss: 0.1596
+Epoch: 048/100 | Batch 0011/0043 | Averaged Loss: 0.1680
+Epoch: 048/100 | Batch 0012/0043 | Averaged Loss: 0.1595
+Epoch: 048/100 | Batch 0013/0043 | Averaged Loss: 0.1627
+Epoch: 048/100 | Batch 0014/0043 | Averaged Loss: 0.1880
+Epoch: 048/100 | Batch 0015/0043 | Averaged Loss: 0.2006
+Epoch: 048/100 | Batch 0016/0043 | Averaged Loss: 0.1863
+Epoch: 048/100 | Batch 0017/0043 | Averaged Loss: 0.1860
+Epoch: 048/100 | Batch 0018/0043 | Averaged Loss: 0.2380
+Epoch: 048/100 | Batch 0019/0043 | Averaged Loss: 0.2227
+Epoch: 048/100 | Batch 0020/0043 | Averaged Loss: 0.1885
+Epoch: 048/100 | Batch 0021/0043 | Averaged Loss: 0.2234
+Epoch: 048/100 | Batch 0022/0043 | Averaged Loss: 0.2387
+Epoch: 048/100 | Batch 0023/0043 | Averaged Loss: 0.2216
+Epoch: 048/100 | Batch 0024/0043 | Averaged Loss: 0.2459
+Epoch: 048/100 | Batch 0025/0043 | Averaged Loss: 0.2441
+Epoch: 048/100 | Batch 0026/0043 | Averaged Loss: 0.2140
+Epoch: 048/100 | Batch 0027/0043 | Averaged Loss: 0.2202
+Epoch: 048/100 | Batch 0028/0043 | Averaged Loss: 0.1921
+Epoch: 048/100 | Batch 0029/0043 | Averaged Loss: 0.2613
+Epoch: 048/100 | Batch 0030/0043 | Averaged Loss: 0.2218
+Epoch: 048/100 | Batch 0031/0043 | Averaged Loss: 0.2319
+Epoch: 048/100 | Batch 0032/0043 | Averaged Loss: 0.2222
+Epoch: 048/100 | Batch 0033/0043 | Averaged Loss: 0.3173
+Epoch: 048/100 | Batch 0034/0043 | Averaged Loss: 0.2413
+Epoch: 048/100 | Batch 0035/0043 | Averaged Loss: 0.2568
+Epoch: 048/100 | Batch 0036/0043 | Averaged Loss: 0.2082
+Epoch: 048/100 | Batch 0037/0043 | Averaged Loss: 0.2334
+Epoch: 048/100 | Batch 0038/0043 | Averaged Loss: 0.2372
+Epoch: 048/100 | Batch 0039/0043 | Averaged Loss: 0.2626
+Epoch: 048/100 | Batch 0040/0043 | Averaged Loss: 0.2078
+Epoch: 048/100 | Batch 0041/0043 | Averaged Loss: 0.2928
+Epoch: 048/100 | Batch 0042/0043 | Averaged Loss: 0.1969
+Epoch: 048/100 | Train: 93.28% | Validation: 73.66%
+Time elapsed: 7.130038261413574 min
+Epoch: 049/100 | Batch 0000/0043 | Averaged Loss: 0.2102
+Epoch: 049/100 | Batch 0001/0043 | Averaged Loss: 0.1941
+Epoch: 049/100 | Batch 0002/0043 | Averaged Loss: 0.1708
+Epoch: 049/100 | Batch 0003/0043 | Averaged Loss: 0.1845
+Epoch: 049/100 | Batch 0004/0043 | Averaged Loss: 0.2355
+Epoch: 049/100 | Batch 0005/0043 | Averaged Loss: 0.1629
+Epoch: 049/100 | Batch 0006/0043 | Averaged Loss: 0.1850
+Epoch: 049/100 | Batch 0007/0043 | Averaged Loss: 0.2000
+Epoch: 049/100 | Batch 0008/0043 | Averaged Loss: 0.1781
+Epoch: 049/100 | Batch 0009/0043 | Averaged Loss: 0.1793
+Epoch: 049/100 | Batch 0010/0043 | Averaged Loss: 0.2109
+Epoch: 049/100 | Batch 0011/0043 | Averaged Loss: 0.1799
+Epoch: 049/100 | Batch 0012/0043 | Averaged Loss: 0.1997
+Epoch: 049/100 | Batch 0013/0043 | Averaged Loss: 0.1923
+Epoch: 049/100 | Batch 0014/0043 | Averaged Loss: 0.1767
+Epoch: 049/100 | Batch 0015/0043 | Averaged Loss: 0.1769
+Epoch: 049/100 | Batch 0016/0043 | Averaged Loss: 0.2341
+Epoch: 049/100 | Batch 0017/0043 | Averaged Loss: 0.1737
+Epoch: 049/100 | Batch 0018/0043 | Averaged Loss: 0.2006
+Epoch: 049/100 | Batch 0019/0043 | Averaged Loss: 0.2412
+Epoch: 049/100 | Batch 0020/0043 | Averaged Loss: 0.2090
+Epoch: 049/100 | Batch 0021/0043 | Averaged Loss: 0.2183
+Epoch: 049/100 | Batch 0022/0043 | Averaged Loss: 0.1613
+Epoch: 049/100 | Batch 0023/0043 | Averaged Loss: 0.1681
+Epoch: 049/100 | Batch 0024/0043 | Averaged Loss: 0.2183
+Epoch: 049/100 | Batch 0025/0043 | Averaged Loss: 0.2347
+Epoch: 049/100 | Batch 0026/0043 | Averaged Loss: 0.2140
+Epoch: 049/100 | Batch 0027/0043 | Averaged Loss: 0.1808
+Epoch: 049/100 | Batch 0028/0043 | Averaged Loss: 0.1906
+Epoch: 049/100 | Batch 0029/0043 | Averaged Loss: 0.2064
+Epoch: 049/100 | Batch 0030/0043 | Averaged Loss: 0.1798
+Epoch: 049/100 | Batch 0031/0043 | Averaged Loss: 0.1778
+Epoch: 049/100 | Batch 0032/0043 | Averaged Loss: 0.1968
+Epoch: 049/100 | Batch 0033/0043 | Averaged Loss: 0.1637
+Epoch: 049/100 | Batch 0034/0043 | Averaged Loss: 0.1899
+Epoch: 049/100 | Batch 0035/0043 | Averaged Loss: 0.1767
+Epoch: 049/100 | Batch 0036/0043 | Averaged Loss: 0.2045
+Epoch: 049/100 | Batch 0037/0043 | Averaged Loss: 0.2082
+Epoch: 049/100 | Batch 0038/0043 | Averaged Loss: 0.2483
+Epoch: 049/100 | Batch 0039/0043 | Averaged Loss: 0.1777
+Epoch: 049/100 | Batch 0040/0043 | Averaged Loss: 0.1986
+Epoch: 049/100 | Batch 0041/0043 | Averaged Loss: 0.2531
+Epoch: 049/100 | Batch 0042/0043 | Averaged Loss: 0.2222
+Epoch: 049/100 | Train: 93.83% | Validation: 73.58%
+Time elapsed: 7.27841329574585 min
+Epoch: 050/100 | Batch 0000/0043 | Averaged Loss: 0.1839
+Epoch: 050/100 | Batch 0001/0043 | Averaged Loss: 0.1815
+Epoch: 050/100 | Batch 0002/0043 | Averaged Loss: 0.1601
+Epoch: 050/100 | Batch 0003/0043 | Averaged Loss: 0.1983
+Epoch: 050/100 | Batch 0004/0043 | Averaged Loss: 0.1691
+Epoch: 050/100 | Batch 0005/0043 | Averaged Loss: 0.1719
+Epoch: 050/100 | Batch 0006/0043 | Averaged Loss: 0.1792
+Epoch: 050/100 | Batch 0007/0043 | Averaged Loss: 0.1526
+Epoch: 050/100 | Batch 0008/0043 | Averaged Loss: 0.1644
+Epoch: 050/100 | Batch 0009/0043 | Averaged Loss: 0.1626
+Epoch: 050/100 | Batch 0010/0043 | Averaged Loss: 0.1392
+Epoch: 050/100 | Batch 0011/0043 | Averaged Loss: 0.1644
+Epoch: 050/100 | Batch 0012/0043 | Averaged Loss: 0.1508
+Epoch: 050/100 | Batch 0013/0043 | Averaged Loss: 0.1809
+Epoch: 050/100 | Batch 0014/0043 | Averaged Loss: 0.1415
+Epoch: 050/100 | Batch 0015/0043 | Averaged Loss: 0.1426
+Epoch: 050/100 | Batch 0016/0043 | Averaged Loss: 0.1934
+Epoch: 050/100 | Batch 0017/0043 | Averaged Loss: 0.2006
+Epoch: 050/100 | Batch 0018/0043 | Averaged Loss: 0.1766
+Epoch: 050/100 | Batch 0019/0043 | Averaged Loss: 0.1434
+Epoch: 050/100 | Batch 0020/0043 | Averaged Loss: 0.2474
+Epoch: 050/100 | Batch 0021/0043 | Averaged Loss: 0.1707
+Epoch: 050/100 | Batch 0022/0043 | Averaged Loss: 0.1409
+Epoch: 050/100 | Batch 0023/0043 | Averaged Loss: 0.1948
+Epoch: 050/100 | Batch 0024/0043 | Averaged Loss: 0.1924
+Epoch: 050/100 | Batch 0025/0043 | Averaged Loss: 0.1899
+Epoch: 050/100 | Batch 0026/0043 | Averaged Loss: 0.1781
+Epoch: 050/100 | Batch 0027/0043 | Averaged Loss: 0.1941
+Epoch: 050/100 | Batch 0028/0043 | Averaged Loss: 0.1656
+Epoch: 050/100 | Batch 0029/0043 | Averaged Loss: 0.2248
+Epoch: 050/100 | Batch 0030/0043 | Averaged Loss: 0.1578
+Epoch: 050/100 | Batch 0031/0043 | Averaged Loss: 0.1992
+Epoch: 050/100 | Batch 0032/0043 | Averaged Loss: 0.1732
+Epoch: 050/100 | Batch 0033/0043 | Averaged Loss: 0.1917
+Epoch: 050/100 | Batch 0034/0043 | Averaged Loss: 0.2174
+Epoch: 050/100 | Batch 0035/0043 | Averaged Loss: 0.2046
+Epoch: 050/100 | Batch 0036/0043 | Averaged Loss: 0.1485
+Epoch: 050/100 | Batch 0037/0043 | Averaged Loss: 0.1984
+Epoch: 050/100 | Batch 0038/0043 | Averaged Loss: 0.1591
+Epoch: 050/100 | Batch 0039/0043 | Averaged Loss: 0.1979
+Epoch: 050/100 | Batch 0040/0043 | Averaged Loss: 0.2206
+Epoch: 050/100 | Batch 0041/0043 | Averaged Loss: 0.2112
+Epoch: 050/100 | Batch 0042/0043 | Averaged Loss: 0.2318
+Epoch: 050/100 | Train: 94.92% | Validation: 74.22%
+Time elapsed: 7.426544666290283 min
+Epoch: 051/100 | Batch 0000/0043 | Averaged Loss: 0.1430
+Epoch: 051/100 | Batch 0001/0043 | Averaged Loss: 0.1307
+Epoch: 051/100 | Batch 0002/0043 | Averaged Loss: 0.1833
+Epoch: 051/100 | Batch 0003/0043 | Averaged Loss: 0.1358
+Epoch: 051/100 | Batch 0004/0043 | Averaged Loss: 0.1726
+Epoch: 051/100 | Batch 0005/0043 | Averaged Loss: 0.1640
+Epoch: 051/100 | Batch 0006/0043 | Averaged Loss: 0.1469
+Epoch: 051/100 | Batch 0007/0043 | Averaged Loss: 0.1621
+Epoch: 051/100 | Batch 0008/0043 | Averaged Loss: 0.1730
+Epoch: 051/100 | Batch 0009/0043 | Averaged Loss: 0.1503
+Epoch: 051/100 | Batch 0010/0043 | Averaged Loss: 0.1755
+Epoch: 051/100 | Batch 0011/0043 | Averaged Loss: 0.1753
+Epoch: 051/100 | Batch 0012/0043 | Averaged Loss: 0.1313
+Epoch: 051/100 | Batch 0013/0043 | Averaged Loss: 0.1657
+Epoch: 051/100 | Batch 0014/0043 | Averaged Loss: 0.2125
+Epoch: 051/100 | Batch 0015/0043 | Averaged Loss: 0.1918
+Epoch: 051/100 | Batch 0016/0043 | Averaged Loss: 0.1455
+Epoch: 051/100 | Batch 0017/0043 | Averaged Loss: 0.1522
+Epoch: 051/100 | Batch 0018/0043 | Averaged Loss: 0.1570
+Epoch: 051/100 | Batch 0019/0043 | Averaged Loss: 0.1488
+Epoch: 051/100 | Batch 0020/0043 | Averaged Loss: 0.1884
+Epoch: 051/100 | Batch 0021/0043 | Averaged Loss: 0.1788
+Epoch: 051/100 | Batch 0022/0043 | Averaged Loss: 0.1645
+Epoch: 051/100 | Batch 0023/0043 | Averaged Loss: 0.1370
+Epoch: 051/100 | Batch 0024/0043 | Averaged Loss: 0.1699
+Epoch: 051/100 | Batch 0025/0043 | Averaged Loss: 0.1782
+Epoch: 051/100 | Batch 0026/0043 | Averaged Loss: 0.2309
+Epoch: 051/100 | Batch 0027/0043 | Averaged Loss: 0.1745
+Epoch: 051/100 | Batch 0028/0043 | Averaged Loss: 0.1996
+Epoch: 051/100 | Batch 0029/0043 | Averaged Loss: 0.1959
+Epoch: 051/100 | Batch 0030/0043 | Averaged Loss: 0.1791
+Epoch: 051/100 | Batch 0031/0043 | Averaged Loss: 0.1743
+Epoch: 051/100 | Batch 0032/0043 | Averaged Loss: 0.1933
+Epoch: 051/100 | Batch 0033/0043 | Averaged Loss: 0.1439
+Epoch: 051/100 | Batch 0034/0043 | Averaged Loss: 0.1759
+Epoch: 051/100 | Batch 0035/0043 | Averaged Loss: 0.1772
+Epoch: 051/100 | Batch 0036/0043 | Averaged Loss: 0.1583
+Epoch: 051/100 | Batch 0037/0043 | Averaged Loss: 0.1924
+Epoch: 051/100 | Batch 0038/0043 | Averaged Loss: 0.1932
+Epoch: 051/100 | Batch 0039/0043 | Averaged Loss: 0.1773
+Epoch: 051/100 | Batch 0040/0043 | Averaged Loss: 0.2552
+Epoch: 051/100 | Batch 0041/0043 | Averaged Loss: 0.1591
+Epoch: 051/100 | Batch 0042/0043 | Averaged Loss: 0.1962
+Epoch: 051/100 | Train: 95.54% | Validation: 73.78%
+Time elapsed: 7.5744829177856445 min
+Epoch: 052/100 | Batch 0000/0043 | Averaged Loss: 0.1353
+Epoch: 052/100 | Batch 0001/0043 | Averaged Loss: 0.1929
+Epoch: 052/100 | Batch 0002/0043 | Averaged Loss: 0.1781
+Epoch: 052/100 | Batch 0003/0043 | Averaged Loss: 0.1170
+Epoch: 052/100 | Batch 0004/0043 | Averaged Loss: 0.1502
+Epoch: 052/100 | Batch 0005/0043 | Averaged Loss: 0.1448
+Epoch: 052/100 | Batch 0006/0043 | Averaged Loss: 0.1387
+Epoch: 052/100 | Batch 0007/0043 | Averaged Loss: 0.1370
+Epoch: 052/100 | Batch 0008/0043 | Averaged Loss: 0.1052
+Epoch: 052/100 | Batch 0009/0043 | Averaged Loss: 0.1118
+Epoch: 052/100 | Batch 0010/0043 | Averaged Loss: 0.1243
+Epoch: 052/100 | Batch 0011/0043 | Averaged Loss: 0.1535
+Epoch: 052/100 | Batch 0012/0043 | Averaged Loss: 0.1505
+Epoch: 052/100 | Batch 0013/0043 | Averaged Loss: 0.1448
+Epoch: 052/100 | Batch 0014/0043 | Averaged Loss: 0.1313
+Epoch: 052/100 | Batch 0015/0043 | Averaged Loss: 0.1272
+Epoch: 052/100 | Batch 0016/0043 | Averaged Loss: 0.1071
+Epoch: 052/100 | Batch 0017/0043 | Averaged Loss: 0.1433
+Epoch: 052/100 | Batch 0018/0043 | Averaged Loss: 0.1327
+Epoch: 052/100 | Batch 0019/0043 | Averaged Loss: 0.1646
+Epoch: 052/100 | Batch 0020/0043 | Averaged Loss: 0.1696
+Epoch: 052/100 | Batch 0021/0043 | Averaged Loss: 0.1809
+Epoch: 052/100 | Batch 0022/0043 | Averaged Loss: 0.1563
+Epoch: 052/100 | Batch 0023/0043 | Averaged Loss: 0.1451
+Epoch: 052/100 | Batch 0024/0043 | Averaged Loss: 0.1540
+Epoch: 052/100 | Batch 0025/0043 | Averaged Loss: 0.1824
+Epoch: 052/100 | Batch 0026/0043 | Averaged Loss: 0.1399
+Epoch: 052/100 | Batch 0027/0043 | Averaged Loss: 0.1889
+Epoch: 052/100 | Batch 0028/0043 | Averaged Loss: 0.1467
+Epoch: 052/100 | Batch 0029/0043 | Averaged Loss: 0.1555
+Epoch: 052/100 | Batch 0030/0043 | Averaged Loss: 0.1557
+Epoch: 052/100 | Batch 0031/0043 | Averaged Loss: 0.1709
+Epoch: 052/100 | Batch 0032/0043 | Averaged Loss: 0.1691
+Epoch: 052/100 | Batch 0033/0043 | Averaged Loss: 0.1677
+Epoch: 052/100 | Batch 0034/0043 | Averaged Loss: 0.1856
+Epoch: 052/100 | Batch 0035/0043 | Averaged Loss: 0.1769
+Epoch: 052/100 | Batch 0036/0043 | Averaged Loss: 0.1621
+Epoch: 052/100 | Batch 0037/0043 | Averaged Loss: 0.1781
+Epoch: 052/100 | Batch 0038/0043 | Averaged Loss: 0.1826
+Epoch: 052/100 | Batch 0039/0043 | Averaged Loss: 0.1418
+Epoch: 052/100 | Batch 0040/0043 | Averaged Loss: 0.1921
+Epoch: 052/100 | Batch 0041/0043 | Averaged Loss: 0.1974
+Epoch: 052/100 | Batch 0042/0043 | Averaged Loss: 0.1590
+Epoch: 052/100 | Train: 94.93% | Validation: 73.58%
+Time elapsed: 7.72252082824707 min
+Epoch: 053/100 | Batch 0000/0043 | Averaged Loss: 0.1200
+Epoch: 053/100 | Batch 0001/0043 | Averaged Loss: 0.1504
+Epoch: 053/100 | Batch 0002/0043 | Averaged Loss: 0.1516
+Epoch: 053/100 | Batch 0003/0043 | Averaged Loss: 0.1461
+Epoch: 053/100 | Batch 0004/0043 | Averaged Loss: 0.1113
+Epoch: 053/100 | Batch 0005/0043 | Averaged Loss: 0.1471
+Epoch: 053/100 | Batch 0006/0043 | Averaged Loss: 0.1429
+Epoch: 053/100 | Batch 0007/0043 | Averaged Loss: 0.1458
+Epoch: 053/100 | Batch 0008/0043 | Averaged Loss: 0.1563
+Epoch: 053/100 | Batch 0009/0043 | Averaged Loss: 0.1920
+Epoch: 053/100 | Batch 0010/0043 | Averaged Loss: 0.1169
+Epoch: 053/100 | Batch 0011/0043 | Averaged Loss: 0.1777
+Epoch: 053/100 | Batch 0012/0043 | Averaged Loss: 0.1727
+Epoch: 053/100 | Batch 0013/0043 | Averaged Loss: 0.1187
+Epoch: 053/100 | Batch 0014/0043 | Averaged Loss: 0.1746
+Epoch: 053/100 | Batch 0015/0043 | Averaged Loss: 0.1757
+Epoch: 053/100 | Batch 0016/0043 | Averaged Loss: 0.1617
+Epoch: 053/100 | Batch 0017/0043 | Averaged Loss: 0.1509
+Epoch: 053/100 | Batch 0018/0043 | Averaged Loss: 0.2141
+Epoch: 053/100 | Batch 0019/0043 | Averaged Loss: 0.1400
+Epoch: 053/100 | Batch 0020/0043 | Averaged Loss: 0.1367
+Epoch: 053/100 | Batch 0021/0043 | Averaged Loss: 0.1535
+Epoch: 053/100 | Batch 0022/0043 | Averaged Loss: 0.1677
+Epoch: 053/100 | Batch 0023/0043 | Averaged Loss: 0.1813
+Epoch: 053/100 | Batch 0024/0043 | Averaged Loss: 0.1237
+Epoch: 053/100 | Batch 0025/0043 | Averaged Loss: 0.1270
+Epoch: 053/100 | Batch 0026/0043 | Averaged Loss: 0.1487
+Epoch: 053/100 | Batch 0027/0043 | Averaged Loss: 0.1163
+Epoch: 053/100 | Batch 0028/0043 | Averaged Loss: 0.1505
+Epoch: 053/100 | Batch 0029/0043 | Averaged Loss: 0.1346
+Epoch: 053/100 | Batch 0030/0043 | Averaged Loss: 0.1429
+Epoch: 053/100 | Batch 0031/0043 | Averaged Loss: 0.1608
+Epoch: 053/100 | Batch 0032/0043 | Averaged Loss: 0.1778
+Epoch: 053/100 | Batch 0033/0043 | Averaged Loss: 0.1484
+Epoch: 053/100 | Batch 0034/0043 | Averaged Loss: 0.1618
+Epoch: 053/100 | Batch 0035/0043 | Averaged Loss: 0.1490
+Epoch: 053/100 | Batch 0036/0043 | Averaged Loss: 0.2045
+Epoch: 053/100 | Batch 0037/0043 | Averaged Loss: 0.1822
+Epoch: 053/100 | Batch 0038/0043 | Averaged Loss: 0.1208
+Epoch: 053/100 | Batch 0039/0043 | Averaged Loss: 0.1622
+Epoch: 053/100 | Batch 0040/0043 | Averaged Loss: 0.1553
+Epoch: 053/100 | Batch 0041/0043 | Averaged Loss: 0.1674
+Epoch: 053/100 | Batch 0042/0043 | Averaged Loss: 0.1854
+Epoch: 053/100 | Train: 96.36% | Validation: 75.66%
+Time elapsed: 7.870737075805664 min
+Epoch: 054/100 | Batch 0000/0043 | Averaged Loss: 0.1216
+Epoch: 054/100 | Batch 0001/0043 | Averaged Loss: 0.1299
+Epoch: 054/100 | Batch 0002/0043 | Averaged Loss: 0.1576
+Epoch: 054/100 | Batch 0003/0043 | Averaged Loss: 0.1412
+Epoch: 054/100 | Batch 0004/0043 | Averaged Loss: 0.1328
+Epoch: 054/100 | Batch 0005/0043 | Averaged Loss: 0.1700
+Epoch: 054/100 | Batch 0006/0043 | Averaged Loss: 0.1459
+Epoch: 054/100 | Batch 0007/0043 | Averaged Loss: 0.1413
+Epoch: 054/100 | Batch 0008/0043 | Averaged Loss: 0.1591
+Epoch: 054/100 | Batch 0009/0043 | Averaged Loss: 0.1243
+Epoch: 054/100 | Batch 0010/0043 | Averaged Loss: 0.1564
+Epoch: 054/100 | Batch 0011/0043 | Averaged Loss: 0.1302
+Epoch: 054/100 | Batch 0012/0043 | Averaged Loss: 0.1388
+Epoch: 054/100 | Batch 0013/0043 | Averaged Loss: 0.1363
+Epoch: 054/100 | Batch 0014/0043 | Averaged Loss: 0.1437
+Epoch: 054/100 | Batch 0015/0043 | Averaged Loss: 0.1312
+Epoch: 054/100 | Batch 0016/0043 | Averaged Loss: 0.1310
+Epoch: 054/100 | Batch 0017/0043 | Averaged Loss: 0.1546
+Epoch: 054/100 | Batch 0018/0043 | Averaged Loss: 0.1569
+Epoch: 054/100 | Batch 0019/0043 | Averaged Loss: 0.1295
+Epoch: 054/100 | Batch 0020/0043 | Averaged Loss: 0.1895
+Epoch: 054/100 | Batch 0021/0043 | Averaged Loss: 0.1772
+Epoch: 054/100 | Batch 0022/0043 | Averaged Loss: 0.1504
+Epoch: 054/100 | Batch 0023/0043 | Averaged Loss: 0.1761
+Epoch: 054/100 | Batch 0024/0043 | Averaged Loss: 0.1174
+Epoch: 054/100 | Batch 0025/0043 | Averaged Loss: 0.1526
+Epoch: 054/100 | Batch 0026/0043 | Averaged Loss: 0.1243
+Epoch: 054/100 | Batch 0027/0043 | Averaged Loss: 0.1917
+Epoch: 054/100 | Batch 0028/0043 | Averaged Loss: 0.1505
+Epoch: 054/100 | Batch 0029/0043 | Averaged Loss: 0.1713
+Epoch: 054/100 | Batch 0030/0043 | Averaged Loss: 0.1780
+Epoch: 054/100 | Batch 0031/0043 | Averaged Loss: 0.1337
+Epoch: 054/100 | Batch 0032/0043 | Averaged Loss: 0.1542
+Epoch: 054/100 | Batch 0033/0043 | Averaged Loss: 0.1371
+Epoch: 054/100 | Batch 0034/0043 | Averaged Loss: 0.1477
+Epoch: 054/100 | Batch 0035/0043 | Averaged Loss: 0.1715
+Epoch: 054/100 | Batch 0036/0043 | Averaged Loss: 0.1169
+Epoch: 054/100 | Batch 0037/0043 | Averaged Loss: 0.1566
+Epoch: 054/100 | Batch 0038/0043 | Averaged Loss: 0.1101
+Epoch: 054/100 | Batch 0039/0043 | Averaged Loss: 0.1204
+Epoch: 054/100 | Batch 0040/0043 | Averaged Loss: 0.1489
+Epoch: 054/100 | Batch 0041/0043 | Averaged Loss: 0.1530
+Epoch: 054/100 | Batch 0042/0043 | Averaged Loss: 0.1489
+Epoch: 054/100 | Train: 96.55% | Validation: 74.22%
+Time elapsed: 8.018817901611328 min
+Epoch: 055/100 | Batch 0000/0043 | Averaged Loss: 0.0935
+Epoch: 055/100 | Batch 0001/0043 | Averaged Loss: 0.1309
+Epoch: 055/100 | Batch 0002/0043 | Averaged Loss: 0.0974
+Epoch: 055/100 | Batch 0003/0043 | Averaged Loss: 0.1014
+Epoch: 055/100 | Batch 0004/0043 | Averaged Loss: 0.0988
+Epoch: 055/100 | Batch 0005/0043 | Averaged Loss: 0.1128
+Epoch: 055/100 | Batch 0006/0043 | Averaged Loss: 0.0954
+Epoch: 055/100 | Batch 0007/0043 | Averaged Loss: 0.1063
+Epoch: 055/100 | Batch 0008/0043 | Averaged Loss: 0.1570
+Epoch: 055/100 | Batch 0009/0043 | Averaged Loss: 0.1161
+Epoch: 055/100 | Batch 0010/0043 | Averaged Loss: 0.1230
+Epoch: 055/100 | Batch 0011/0043 | Averaged Loss: 0.1185
+Epoch: 055/100 | Batch 0012/0043 | Averaged Loss: 0.1399
+Epoch: 055/100 | Batch 0013/0043 | Averaged Loss: 0.1375
+Epoch: 055/100 | Batch 0014/0043 | Averaged Loss: 0.1132
+Epoch: 055/100 | Batch 0015/0043 | Averaged Loss: 0.1459
+Epoch: 055/100 | Batch 0016/0043 | Averaged Loss: 0.1136
+Epoch: 055/100 | Batch 0017/0043 | Averaged Loss: 0.1061
+Epoch: 055/100 | Batch 0018/0043 | Averaged Loss: 0.1150
+Epoch: 055/100 | Batch 0019/0043 | Averaged Loss: 0.1337
+Epoch: 055/100 | Batch 0020/0043 | Averaged Loss: 0.1502
+Epoch: 055/100 | Batch 0021/0043 | Averaged Loss: 0.1430
+Epoch: 055/100 | Batch 0022/0043 | Averaged Loss: 0.1395
+Epoch: 055/100 | Batch 0023/0043 | Averaged Loss: 0.1172
+Epoch: 055/100 | Batch 0024/0043 | Averaged Loss: 0.1774
+Epoch: 055/100 | Batch 0025/0043 | Averaged Loss: 0.1285
+Epoch: 055/100 | Batch 0026/0043 | Averaged Loss: 0.1167
+Epoch: 055/100 | Batch 0027/0043 | Averaged Loss: 0.1416
+Epoch: 055/100 | Batch 0028/0043 | Averaged Loss: 0.1068
+Epoch: 055/100 | Batch 0029/0043 | Averaged Loss: 0.1358
+Epoch: 055/100 | Batch 0030/0043 | Averaged Loss: 0.1305
+Epoch: 055/100 | Batch 0031/0043 | Averaged Loss: 0.1249
+Epoch: 055/100 | Batch 0032/0043 | Averaged Loss: 0.1295
+Epoch: 055/100 | Batch 0033/0043 | Averaged Loss: 0.1214
+Epoch: 055/100 | Batch 0034/0043 | Averaged Loss: 0.1510
+Epoch: 055/100 | Batch 0035/0043 | Averaged Loss: 0.1170
+Epoch: 055/100 | Batch 0036/0043 | Averaged Loss: 0.1595
+Epoch: 055/100 | Batch 0037/0043 | Averaged Loss: 0.1727
+Epoch: 055/100 | Batch 0038/0043 | Averaged Loss: 0.1611
+Epoch: 055/100 | Batch 0039/0043 | Averaged Loss: 0.1513
+Epoch: 055/100 | Batch 0040/0043 | Averaged Loss: 0.1975
+Epoch: 055/100 | Batch 0041/0043 | Averaged Loss: 0.1918
+Epoch: 055/100 | Batch 0042/0043 | Averaged Loss: 0.1358
+Epoch: 055/100 | Train: 95.59% | Validation: 73.95%
+Time elapsed: 8.167299270629883 min
+Epoch: 056/100 | Batch 0000/0043 | Averaged Loss: 0.1444
+Epoch: 056/100 | Batch 0001/0043 | Averaged Loss: 0.1134
+Epoch: 056/100 | Batch 0002/0043 | Averaged Loss: 0.1712
+Epoch: 056/100 | Batch 0003/0043 | Averaged Loss: 0.1219
+Epoch: 056/100 | Batch 0004/0043 | Averaged Loss: 0.1274
+Epoch: 056/100 | Batch 0005/0043 | Averaged Loss: 0.1590
+Epoch: 056/100 | Batch 0006/0043 | Averaged Loss: 0.1434
+Epoch: 056/100 | Batch 0007/0043 | Averaged Loss: 0.1430
+Epoch: 056/100 | Batch 0008/0043 | Averaged Loss: 0.1314
+Epoch: 056/100 | Batch 0009/0043 | Averaged Loss: 0.1111
+Epoch: 056/100 | Batch 0010/0043 | Averaged Loss: 0.1228
+Epoch: 056/100 | Batch 0011/0043 | Averaged Loss: 0.1268
+Epoch: 056/100 | Batch 0012/0043 | Averaged Loss: 0.1228
+Epoch: 056/100 | Batch 0013/0043 | Averaged Loss: 0.1657
+Epoch: 056/100 | Batch 0014/0043 | Averaged Loss: 0.1391
+Epoch: 056/100 | Batch 0015/0043 | Averaged Loss: 0.1595
+Epoch: 056/100 | Batch 0016/0043 | Averaged Loss: 0.1421
+Epoch: 056/100 | Batch 0017/0043 | Averaged Loss: 0.1871
+Epoch: 056/100 | Batch 0018/0043 | Averaged Loss: 0.1076
+Epoch: 056/100 | Batch 0019/0043 | Averaged Loss: 0.1252
+Epoch: 056/100 | Batch 0020/0043 | Averaged Loss: 0.1801
+Epoch: 056/100 | Batch 0021/0043 | Averaged Loss: 0.1644
+Epoch: 056/100 | Batch 0022/0043 | Averaged Loss: 0.1379
+Epoch: 056/100 | Batch 0023/0043 | Averaged Loss: 0.1515
+Epoch: 056/100 | Batch 0024/0043 | Averaged Loss: 0.1950
+Epoch: 056/100 | Batch 0025/0043 | Averaged Loss: 0.1441
+Epoch: 056/100 | Batch 0026/0043 | Averaged Loss: 0.1273
+Epoch: 056/100 | Batch 0027/0043 | Averaged Loss: 0.1296
+Epoch: 056/100 | Batch 0028/0043 | Averaged Loss: 0.1720
+Epoch: 056/100 | Batch 0029/0043 | Averaged Loss: 0.1306
+Epoch: 056/100 | Batch 0030/0043 | Averaged Loss: 0.1745
+Epoch: 056/100 | Batch 0031/0043 | Averaged Loss: 0.1473
+Epoch: 056/100 | Batch 0032/0043 | Averaged Loss: 0.1341
+Epoch: 056/100 | Batch 0033/0043 | Averaged Loss: 0.1511
+Epoch: 056/100 | Batch 0034/0043 | Averaged Loss: 0.1502
+Epoch: 056/100 | Batch 0035/0043 | Averaged Loss: 0.1476
+Epoch: 056/100 | Batch 0036/0043 | Averaged Loss: 0.1366
+Epoch: 056/100 | Batch 0037/0043 | Averaged Loss: 0.1291
+Epoch: 056/100 | Batch 0038/0043 | Averaged Loss: 0.1436
+Epoch: 056/100 | Batch 0039/0043 | Averaged Loss: 0.1381
+Epoch: 056/100 | Batch 0040/0043 | Averaged Loss: 0.1543
+Epoch: 056/100 | Batch 0041/0043 | Averaged Loss: 0.1708
+Epoch: 056/100 | Batch 0042/0043 | Averaged Loss: 0.1344
+Epoch: 056/100 | Train: 96.41% | Validation: 74.98%
+Time elapsed: 8.315506935119629 min
+Epoch: 057/100 | Batch 0000/0043 | Averaged Loss: 0.1327
+Epoch: 057/100 | Batch 0001/0043 | Averaged Loss: 0.0846
+Epoch: 057/100 | Batch 0002/0043 | Averaged Loss: 0.1166
+Epoch: 057/100 | Batch 0003/0043 | Averaged Loss: 0.1383
+Epoch: 057/100 | Batch 0004/0043 | Averaged Loss: 0.1145
+Epoch: 057/100 | Batch 0005/0043 | Averaged Loss: 0.1320
+Epoch: 057/100 | Batch 0006/0043 | Averaged Loss: 0.1626
+Epoch: 057/100 | Batch 0007/0043 | Averaged Loss: 0.0950
+Epoch: 057/100 | Batch 0008/0043 | Averaged Loss: 0.1073
+Epoch: 057/100 | Batch 0009/0043 | Averaged Loss: 0.1654
+Epoch: 057/100 | Batch 0010/0043 | Averaged Loss: 0.1305
+Epoch: 057/100 | Batch 0011/0043 | Averaged Loss: 0.1126
+Epoch: 057/100 | Batch 0012/0043 | Averaged Loss: 0.1463
+Epoch: 057/100 | Batch 0013/0043 | Averaged Loss: 0.1319
+Epoch: 057/100 | Batch 0014/0043 | Averaged Loss: 0.1267
+Epoch: 057/100 | Batch 0015/0043 | Averaged Loss: 0.1200
+Epoch: 057/100 | Batch 0016/0043 | Averaged Loss: 0.1495
+Epoch: 057/100 | Batch 0017/0043 | Averaged Loss: 0.1432
+Epoch: 057/100 | Batch 0018/0043 | Averaged Loss: 0.1302
+Epoch: 057/100 | Batch 0019/0043 | Averaged Loss: 0.1032
+Epoch: 057/100 | Batch 0020/0043 | Averaged Loss: 0.1194
+Epoch: 057/100 | Batch 0021/0043 | Averaged Loss: 0.1307
+Epoch: 057/100 | Batch 0022/0043 | Averaged Loss: 0.0956
+Epoch: 057/100 | Batch 0023/0043 | Averaged Loss: 0.1041
+Epoch: 057/100 | Batch 0024/0043 | Averaged Loss: 0.1028
+Epoch: 057/100 | Batch 0025/0043 | Averaged Loss: 0.1696
+Epoch: 057/100 | Batch 0026/0043 | Averaged Loss: 0.1470
+Epoch: 057/100 | Batch 0027/0043 | Averaged Loss: 0.1311
+Epoch: 057/100 | Batch 0028/0043 | Averaged Loss: 0.1108
+Epoch: 057/100 | Batch 0029/0043 | Averaged Loss: 0.1045
+Epoch: 057/100 | Batch 0030/0043 | Averaged Loss: 0.0992
+Epoch: 057/100 | Batch 0031/0043 | Averaged Loss: 0.1214
+Epoch: 057/100 | Batch 0032/0043 | Averaged Loss: 0.1420
+Epoch: 057/100 | Batch 0033/0043 | Averaged Loss: 0.1145
+Epoch: 057/100 | Batch 0034/0043 | Averaged Loss: 0.1512
+Epoch: 057/100 | Batch 0035/0043 | Averaged Loss: 0.1485
+Epoch: 057/100 | Batch 0036/0043 | Averaged Loss: 0.1386
+Epoch: 057/100 | Batch 0037/0043 | Averaged Loss: 0.1009
+Epoch: 057/100 | Batch 0038/0043 | Averaged Loss: 0.0943
+Epoch: 057/100 | Batch 0039/0043 | Averaged Loss: 0.1407
+Epoch: 057/100 | Batch 0040/0043 | Averaged Loss: 0.1044
+Epoch: 057/100 | Batch 0041/0043 | Averaged Loss: 0.1454
+Epoch: 057/100 | Batch 0042/0043 | Averaged Loss: 0.1405
+Epoch: 057/100 | Train: 96.67% | Validation: 74.07%
+Time elapsed: 8.46399974822998 min
+Epoch: 058/100 | Batch 0000/0043 | Averaged Loss: 0.1003
+Epoch: 058/100 | Batch 0001/0043 | Averaged Loss: 0.1526
+Epoch: 058/100 | Batch 0002/0043 | Averaged Loss: 0.1088
+Epoch: 058/100 | Batch 0003/0043 | Averaged Loss: 0.1161
+Epoch: 058/100 | Batch 0004/0043 | Averaged Loss: 0.1150
+Epoch: 058/100 | Batch 0005/0043 | Averaged Loss: 0.1363
+Epoch: 058/100 | Batch 0006/0043 | Averaged Loss: 0.1075
+Epoch: 058/100 | Batch 0007/0043 | Averaged Loss: 0.1029
+Epoch: 058/100 | Batch 0008/0043 | Averaged Loss: 0.1094
+Epoch: 058/100 | Batch 0009/0043 | Averaged Loss: 0.1667
+Epoch: 058/100 | Batch 0010/0043 | Averaged Loss: 0.1535
+Epoch: 058/100 | Batch 0011/0043 | Averaged Loss: 0.1459
+Epoch: 058/100 | Batch 0012/0043 | Averaged Loss: 0.0971
+Epoch: 058/100 | Batch 0013/0043 | Averaged Loss: 0.1129
+Epoch: 058/100 | Batch 0014/0043 | Averaged Loss: 0.1209
+Epoch: 058/100 | Batch 0015/0043 | Averaged Loss: 0.1040
+Epoch: 058/100 | Batch 0016/0043 | Averaged Loss: 0.1202
+Epoch: 058/100 | Batch 0017/0043 | Averaged Loss: 0.0930
+Epoch: 058/100 | Batch 0018/0043 | Averaged Loss: 0.1190
+Epoch: 058/100 | Batch 0019/0043 | Averaged Loss: 0.1263
+Epoch: 058/100 | Batch 0020/0043 | Averaged Loss: 0.0981
+Epoch: 058/100 | Batch 0021/0043 | Averaged Loss: 0.1859
+Epoch: 058/100 | Batch 0022/0043 | Averaged Loss: 0.1091
+Epoch: 058/100 | Batch 0023/0043 | Averaged Loss: 0.1101
+Epoch: 058/100 | Batch 0024/0043 | Averaged Loss: 0.1256
+Epoch: 058/100 | Batch 0025/0043 | Averaged Loss: 0.1347
+Epoch: 058/100 | Batch 0026/0043 | Averaged Loss: 0.1212
+Epoch: 058/100 | Batch 0027/0043 | Averaged Loss: 0.1031
+Epoch: 058/100 | Batch 0028/0043 | Averaged Loss: 0.1263
+Epoch: 058/100 | Batch 0029/0043 | Averaged Loss: 0.1418
+Epoch: 058/100 | Batch 0030/0043 | Averaged Loss: 0.1150
+Epoch: 058/100 | Batch 0031/0043 | Averaged Loss: 0.1310
+Epoch: 058/100 | Batch 0032/0043 | Averaged Loss: 0.1047
+Epoch: 058/100 | Batch 0033/0043 | Averaged Loss: 0.1684
+Epoch: 058/100 | Batch 0034/0043 | Averaged Loss: 0.1291
+Epoch: 058/100 | Batch 0035/0043 | Averaged Loss: 0.1685
+Epoch: 058/100 | Batch 0036/0043 | Averaged Loss: 0.1459
+Epoch: 058/100 | Batch 0037/0043 | Averaged Loss: 0.1406
+Epoch: 058/100 | Batch 0038/0043 | Averaged Loss: 0.1630
+Epoch: 058/100 | Batch 0039/0043 | Averaged Loss: 0.1466
+Epoch: 058/100 | Batch 0040/0043 | Averaged Loss: 0.1766
+Epoch: 058/100 | Batch 0041/0043 | Averaged Loss: 0.1408
+Epoch: 058/100 | Batch 0042/0043 | Averaged Loss: 0.1429
+Epoch: 058/100 | Train: 95.71% | Validation: 73.90%
+Time elapsed: 8.612544059753418 min
+Epoch: 059/100 | Batch 0000/0043 | Averaged Loss: 0.1448
+Epoch: 059/100 | Batch 0001/0043 | Averaged Loss: 0.1421
+Epoch: 059/100 | Batch 0002/0043 | Averaged Loss: 0.1201
+Epoch: 059/100 | Batch 0003/0043 | Averaged Loss: 0.1367
+Epoch: 059/100 | Batch 0004/0043 | Averaged Loss: 0.1059
+Epoch: 059/100 | Batch 0005/0043 | Averaged Loss: 0.1249
+Epoch: 059/100 | Batch 0006/0043 | Averaged Loss: 0.1653
+Epoch: 059/100 | Batch 0007/0043 | Averaged Loss: 0.1102
+Epoch: 059/100 | Batch 0008/0043 | Averaged Loss: 0.1275
+Epoch: 059/100 | Batch 0009/0043 | Averaged Loss: 0.1595
+Epoch: 059/100 | Batch 0010/0043 | Averaged Loss: 0.1321
+Epoch: 059/100 | Batch 0011/0043 | Averaged Loss: 0.1297
+Epoch: 059/100 | Batch 0012/0043 | Averaged Loss: 0.1214
+Epoch: 059/100 | Batch 0013/0043 | Averaged Loss: 0.1183
+Epoch: 059/100 | Batch 0014/0043 | Averaged Loss: 0.1150
+Epoch: 059/100 | Batch 0015/0043 | Averaged Loss: 0.1453
+Epoch: 059/100 | Batch 0016/0043 | Averaged Loss: 0.0805
+Epoch: 059/100 | Batch 0017/0043 | Averaged Loss: 0.1074
+Epoch: 059/100 | Batch 0018/0043 | Averaged Loss: 0.1345
+Epoch: 059/100 | Batch 0019/0043 | Averaged Loss: 0.1197
+Epoch: 059/100 | Batch 0020/0043 | Averaged Loss: 0.1413
+Epoch: 059/100 | Batch 0021/0043 | Averaged Loss: 0.1028
+Epoch: 059/100 | Batch 0022/0043 | Averaged Loss: 0.1378
+Epoch: 059/100 | Batch 0023/0043 | Averaged Loss: 0.1251
+Epoch: 059/100 | Batch 0024/0043 | Averaged Loss: 0.1190
+Epoch: 059/100 | Batch 0025/0043 | Averaged Loss: 0.1133
+Epoch: 059/100 | Batch 0026/0043 | Averaged Loss: 0.1561
+Epoch: 059/100 | Batch 0027/0043 | Averaged Loss: 0.1345
+Epoch: 059/100 | Batch 0028/0043 | Averaged Loss: 0.1115
+Epoch: 059/100 | Batch 0029/0043 | Averaged Loss: 0.1129
+Epoch: 059/100 | Batch 0030/0043 | Averaged Loss: 0.1170
+Epoch: 059/100 | Batch 0031/0043 | Averaged Loss: 0.1120
+Epoch: 059/100 | Batch 0032/0043 | Averaged Loss: 0.1451
+Epoch: 059/100 | Batch 0033/0043 | Averaged Loss: 0.1373
+Epoch: 059/100 | Batch 0034/0043 | Averaged Loss: 0.1155
+Epoch: 059/100 | Batch 0035/0043 | Averaged Loss: 0.1011
+Epoch: 059/100 | Batch 0036/0043 | Averaged Loss: 0.1100
+Epoch: 059/100 | Batch 0037/0043 | Averaged Loss: 0.1048
+Epoch: 059/100 | Batch 0038/0043 | Averaged Loss: 0.1590
+Epoch: 059/100 | Batch 0039/0043 | Averaged Loss: 0.1037
+Epoch: 059/100 | Batch 0040/0043 | Averaged Loss: 0.1594
+Epoch: 059/100 | Batch 0041/0043 | Averaged Loss: 0.0833
+Epoch: 059/100 | Batch 0042/0043 | Averaged Loss: 0.1860
+Epoch: 059/100 | Train: 96.03% | Validation: 74.12%
+Time elapsed: 8.76084041595459 min
+Epoch: 060/100 | Batch 0000/0043 | Averaged Loss: 0.1095
+Epoch: 060/100 | Batch 0001/0043 | Averaged Loss: 0.1301
+Epoch: 060/100 | Batch 0002/0043 | Averaged Loss: 0.1198
+Epoch: 060/100 | Batch 0003/0043 | Averaged Loss: 0.1255
+Epoch: 060/100 | Batch 0004/0043 | Averaged Loss: 0.0886
+Epoch: 060/100 | Batch 0005/0043 | Averaged Loss: 0.1110
+Epoch: 060/100 | Batch 0006/0043 | Averaged Loss: 0.0831
+Epoch: 060/100 | Batch 0007/0043 | Averaged Loss: 0.1069
+Epoch: 060/100 | Batch 0008/0043 | Averaged Loss: 0.1140
+Epoch: 060/100 | Batch 0009/0043 | Averaged Loss: 0.0770
+Epoch: 060/100 | Batch 0010/0043 | Averaged Loss: 0.1273
+Epoch: 060/100 | Batch 0011/0043 | Averaged Loss: 0.0789
+Epoch: 060/100 | Batch 0012/0043 | Averaged Loss: 0.0868
+Epoch: 060/100 | Batch 0013/0043 | Averaged Loss: 0.1083
+Epoch: 060/100 | Batch 0014/0043 | Averaged Loss: 0.1005
+Epoch: 060/100 | Batch 0015/0043 | Averaged Loss: 0.0655
+Epoch: 060/100 | Batch 0016/0043 | Averaged Loss: 0.1183
+Epoch: 060/100 | Batch 0017/0043 | Averaged Loss: 0.1152
+Epoch: 060/100 | Batch 0018/0043 | Averaged Loss: 0.1023
+Epoch: 060/100 | Batch 0019/0043 | Averaged Loss: 0.1149
+Epoch: 060/100 | Batch 0020/0043 | Averaged Loss: 0.0942
+Epoch: 060/100 | Batch 0021/0043 | Averaged Loss: 0.1105
+Epoch: 060/100 | Batch 0022/0043 | Averaged Loss: 0.1091
+Epoch: 060/100 | Batch 0023/0043 | Averaged Loss: 0.1313
+Epoch: 060/100 | Batch 0024/0043 | Averaged Loss: 0.0951
+Epoch: 060/100 | Batch 0025/0043 | Averaged Loss: 0.1051
+Epoch: 060/100 | Batch 0026/0043 | Averaged Loss: 0.1103
+Epoch: 060/100 | Batch 0027/0043 | Averaged Loss: 0.1313
+Epoch: 060/100 | Batch 0028/0043 | Averaged Loss: 0.1081
+Epoch: 060/100 | Batch 0029/0043 | Averaged Loss: 0.1231
+Epoch: 060/100 | Batch 0030/0043 | Averaged Loss: 0.1068
+Epoch: 060/100 | Batch 0031/0043 | Averaged Loss: 0.1067
+Epoch: 060/100 | Batch 0032/0043 | Averaged Loss: 0.1294
+Epoch: 060/100 | Batch 0033/0043 | Averaged Loss: 0.1244
+Epoch: 060/100 | Batch 0034/0043 | Averaged Loss: 0.0926
+Epoch: 060/100 | Batch 0035/0043 | Averaged Loss: 0.1051
+Epoch: 060/100 | Batch 0036/0043 | Averaged Loss: 0.1430
+Epoch: 060/100 | Batch 0037/0043 | Averaged Loss: 0.0990
+Epoch: 060/100 | Batch 0038/0043 | Averaged Loss: 0.1051
+Epoch: 060/100 | Batch 0039/0043 | Averaged Loss: 0.0889
+Epoch: 060/100 | Batch 0040/0043 | Averaged Loss: 0.1084
+Epoch: 060/100 | Batch 0041/0043 | Averaged Loss: 0.1026
+Epoch: 060/100 | Batch 0042/0043 | Averaged Loss: 0.1410
+Epoch: 060/100 | Train: 97.16% | Validation: 73.95%
+Time elapsed: 8.909676551818848 min
+Epoch: 061/100 | Batch 0000/0043 | Averaged Loss: 0.0886
+Epoch: 061/100 | Batch 0001/0043 | Averaged Loss: 0.0730
+Epoch: 061/100 | Batch 0002/0043 | Averaged Loss: 0.0869
+Epoch: 061/100 | Batch 0003/0043 | Averaged Loss: 0.1154
+Epoch: 061/100 | Batch 0004/0043 | Averaged Loss: 0.0621
+Epoch: 061/100 | Batch 0005/0043 | Averaged Loss: 0.0953
+Epoch: 061/100 | Batch 0006/0043 | Averaged Loss: 0.0885
+Epoch: 061/100 | Batch 0007/0043 | Averaged Loss: 0.0939
+Epoch: 061/100 | Batch 0008/0043 | Averaged Loss: 0.0710
+Epoch: 061/100 | Batch 0009/0043 | Averaged Loss: 0.1078
+Epoch: 061/100 | Batch 0010/0043 | Averaged Loss: 0.1065
+Epoch: 061/100 | Batch 0011/0043 | Averaged Loss: 0.0949
+Epoch: 061/100 | Batch 0012/0043 | Averaged Loss: 0.0888
+Epoch: 061/100 | Batch 0013/0043 | Averaged Loss: 0.0896
+Epoch: 061/100 | Batch 0014/0043 | Averaged Loss: 0.1096
+Epoch: 061/100 | Batch 0015/0043 | Averaged Loss: 0.1236
+Epoch: 061/100 | Batch 0016/0043 | Averaged Loss: 0.1112
+Epoch: 061/100 | Batch 0017/0043 | Averaged Loss: 0.0948
+Epoch: 061/100 | Batch 0018/0043 | Averaged Loss: 0.1075
+Epoch: 061/100 | Batch 0019/0043 | Averaged Loss: 0.1321
+Epoch: 061/100 | Batch 0020/0043 | Averaged Loss: 0.1252
+Epoch: 061/100 | Batch 0021/0043 | Averaged Loss: 0.1259
+Epoch: 061/100 | Batch 0022/0043 | Averaged Loss: 0.1034
+Epoch: 061/100 | Batch 0023/0043 | Averaged Loss: 0.1244
+Epoch: 061/100 | Batch 0024/0043 | Averaged Loss: 0.1344
+Epoch: 061/100 | Batch 0025/0043 | Averaged Loss: 0.1453
+Epoch: 061/100 | Batch 0026/0043 | Averaged Loss: 0.1359
+Epoch: 061/100 | Batch 0027/0043 | Averaged Loss: 0.0966
+Epoch: 061/100 | Batch 0028/0043 | Averaged Loss: 0.0788
+Epoch: 061/100 | Batch 0029/0043 | Averaged Loss: 0.1008
+Epoch: 061/100 | Batch 0030/0043 | Averaged Loss: 0.1159
+Epoch: 061/100 | Batch 0031/0043 | Averaged Loss: 0.1160
+Epoch: 061/100 | Batch 0032/0043 | Averaged Loss: 0.1215
+Epoch: 061/100 | Batch 0033/0043 | Averaged Loss: 0.1313
+Epoch: 061/100 | Batch 0034/0043 | Averaged Loss: 0.1123
+Epoch: 061/100 | Batch 0035/0043 | Averaged Loss: 0.1353
+Epoch: 061/100 | Batch 0036/0043 | Averaged Loss: 0.1032
+Epoch: 061/100 | Batch 0037/0043 | Averaged Loss: 0.1459
+Epoch: 061/100 | Batch 0038/0043 | Averaged Loss: 0.1611
+Epoch: 061/100 | Batch 0039/0043 | Averaged Loss: 0.1466
+Epoch: 061/100 | Batch 0040/0043 | Averaged Loss: 0.1088
+Epoch: 061/100 | Batch 0041/0043 | Averaged Loss: 0.1447
+Epoch: 061/100 | Batch 0042/0043 | Averaged Loss: 0.1162
+Epoch: 061/100 | Train: 97.09% | Validation: 75.22%
+Time elapsed: 9.05810546875 min
+Epoch: 062/100 | Batch 0000/0043 | Averaged Loss: 0.0823
+Epoch: 062/100 | Batch 0001/0043 | Averaged Loss: 0.0941
+Epoch: 062/100 | Batch 0002/0043 | Averaged Loss: 0.1036
+Epoch: 062/100 | Batch 0003/0043 | Averaged Loss: 0.1395
+Epoch: 062/100 | Batch 0004/0043 | Averaged Loss: 0.0878
+Epoch: 062/100 | Batch 0005/0043 | Averaged Loss: 0.0960
+Epoch: 062/100 | Batch 0006/0043 | Averaged Loss: 0.0835
+Epoch: 062/100 | Batch 0007/0043 | Averaged Loss: 0.1009
+Epoch: 062/100 | Batch 0008/0043 | Averaged Loss: 0.0765
+Epoch: 062/100 | Batch 0009/0043 | Averaged Loss: 0.1098
+Epoch: 062/100 | Batch 0010/0043 | Averaged Loss: 0.1127
+Epoch: 062/100 | Batch 0011/0043 | Averaged Loss: 0.1202
+Epoch: 062/100 | Batch 0012/0043 | Averaged Loss: 0.0985
+Epoch: 062/100 | Batch 0013/0043 | Averaged Loss: 0.0972
+Epoch: 062/100 | Batch 0014/0043 | Averaged Loss: 0.0844
+Epoch: 062/100 | Batch 0015/0043 | Averaged Loss: 0.1192
+Epoch: 062/100 | Batch 0016/0043 | Averaged Loss: 0.0874
+Epoch: 062/100 | Batch 0017/0043 | Averaged Loss: 0.0897
+Epoch: 062/100 | Batch 0018/0043 | Averaged Loss: 0.1329
+Epoch: 062/100 | Batch 0019/0043 | Averaged Loss: 0.0997
+Epoch: 062/100 | Batch 0020/0043 | Averaged Loss: 0.0843
+Epoch: 062/100 | Batch 0021/0043 | Averaged Loss: 0.1438
+Epoch: 062/100 | Batch 0022/0043 | Averaged Loss: 0.1036
+Epoch: 062/100 | Batch 0023/0043 | Averaged Loss: 0.0764
+Epoch: 062/100 | Batch 0024/0043 | Averaged Loss: 0.0820
+Epoch: 062/100 | Batch 0025/0043 | Averaged Loss: 0.1297
+Epoch: 062/100 | Batch 0026/0043 | Averaged Loss: 0.1123
+Epoch: 062/100 | Batch 0027/0043 | Averaged Loss: 0.1343
+Epoch: 062/100 | Batch 0028/0043 | Averaged Loss: 0.1073
+Epoch: 062/100 | Batch 0029/0043 | Averaged Loss: 0.1316
+Epoch: 062/100 | Batch 0030/0043 | Averaged Loss: 0.0931
+Epoch: 062/100 | Batch 0031/0043 | Averaged Loss: 0.1468
+Epoch: 062/100 | Batch 0032/0043 | Averaged Loss: 0.1032
+Epoch: 062/100 | Batch 0033/0043 | Averaged Loss: 0.1481
+Epoch: 062/100 | Batch 0034/0043 | Averaged Loss: 0.1268
+Epoch: 062/100 | Batch 0035/0043 | Averaged Loss: 0.1188
+Epoch: 062/100 | Batch 0036/0043 | Averaged Loss: 0.1326
+Epoch: 062/100 | Batch 0037/0043 | Averaged Loss: 0.1730
+Epoch: 062/100 | Batch 0038/0043 | Averaged Loss: 0.1011
+Epoch: 062/100 | Batch 0039/0043 | Averaged Loss: 0.1102
+Epoch: 062/100 | Batch 0040/0043 | Averaged Loss: 0.1086
+Epoch: 062/100 | Batch 0041/0043 | Averaged Loss: 0.1268
+Epoch: 062/100 | Batch 0042/0043 | Averaged Loss: 0.1468
+Epoch: 062/100 | Train: 96.90% | Validation: 74.49%
+Time elapsed: 9.20665454864502 min
+Epoch: 063/100 | Batch 0000/0043 | Averaged Loss: 0.0817
+Epoch: 063/100 | Batch 0001/0043 | Averaged Loss: 0.0812
+Epoch: 063/100 | Batch 0002/0043 | Averaged Loss: 0.0904
+Epoch: 063/100 | Batch 0003/0043 | Averaged Loss: 0.0901
+Epoch: 063/100 | Batch 0004/0043 | Averaged Loss: 0.1190
+Epoch: 063/100 | Batch 0005/0043 | Averaged Loss: 0.0875
+Epoch: 063/100 | Batch 0006/0043 | Averaged Loss: 0.0887
+Epoch: 063/100 | Batch 0007/0043 | Averaged Loss: 0.0822
+Epoch: 063/100 | Batch 0008/0043 | Averaged Loss: 0.0730
+Epoch: 063/100 | Batch 0009/0043 | Averaged Loss: 0.0870
+Epoch: 063/100 | Batch 0010/0043 | Averaged Loss: 0.1017
+Epoch: 063/100 | Batch 0011/0043 | Averaged Loss: 0.0845
+Epoch: 063/100 | Batch 0012/0043 | Averaged Loss: 0.1141
+Epoch: 063/100 | Batch 0013/0043 | Averaged Loss: 0.1164
+Epoch: 063/100 | Batch 0014/0043 | Averaged Loss: 0.0999
+Epoch: 063/100 | Batch 0015/0043 | Averaged Loss: 0.1164
+Epoch: 063/100 | Batch 0016/0043 | Averaged Loss: 0.1039
+Epoch: 063/100 | Batch 0017/0043 | Averaged Loss: 0.0899
+Epoch: 063/100 | Batch 0018/0043 | Averaged Loss: 0.1245
+Epoch: 063/100 | Batch 0019/0043 | Averaged Loss: 0.1408
+Epoch: 063/100 | Batch 0020/0043 | Averaged Loss: 0.1318
+Epoch: 063/100 | Batch 0021/0043 | Averaged Loss: 0.0856
+Epoch: 063/100 | Batch 0022/0043 | Averaged Loss: 0.1099
+Epoch: 063/100 | Batch 0023/0043 | Averaged Loss: 0.1215
+Epoch: 063/100 | Batch 0024/0043 | Averaged Loss: 0.1050
+Epoch: 063/100 | Batch 0025/0043 | Averaged Loss: 0.1006
+Epoch: 063/100 | Batch 0026/0043 | Averaged Loss: 0.1382
+Epoch: 063/100 | Batch 0027/0043 | Averaged Loss: 0.1187
+Epoch: 063/100 | Batch 0028/0043 | Averaged Loss: 0.1447
+Epoch: 063/100 | Batch 0029/0043 | Averaged Loss: 0.0894
+Epoch: 063/100 | Batch 0030/0043 | Averaged Loss: 0.1449
+Epoch: 063/100 | Batch 0031/0043 | Averaged Loss: 0.1263
+Epoch: 063/100 | Batch 0032/0043 | Averaged Loss: 0.0952
+Epoch: 063/100 | Batch 0033/0043 | Averaged Loss: 0.1160
+Epoch: 063/100 | Batch 0034/0043 | Averaged Loss: 0.1536
+Epoch: 063/100 | Batch 0035/0043 | Averaged Loss: 0.1158
+Epoch: 063/100 | Batch 0036/0043 | Averaged Loss: 0.1134
+Epoch: 063/100 | Batch 0037/0043 | Averaged Loss: 0.1144
+Epoch: 063/100 | Batch 0038/0043 | Averaged Loss: 0.1148
+Epoch: 063/100 | Batch 0039/0043 | Averaged Loss: 0.1048
+Epoch: 063/100 | Batch 0040/0043 | Averaged Loss: 0.1435
+Epoch: 063/100 | Batch 0041/0043 | Averaged Loss: 0.1305
+Epoch: 063/100 | Batch 0042/0043 | Averaged Loss: 0.1479
+Epoch: 063/100 | Train: 96.83% | Validation: 74.90%
+Time elapsed: 9.355243682861328 min
+Epoch: 064/100 | Batch 0000/0043 | Averaged Loss: 0.1231
+Epoch: 064/100 | Batch 0001/0043 | Averaged Loss: 0.0848
+Epoch: 064/100 | Batch 0002/0043 | Averaged Loss: 0.0739
+Epoch: 064/100 | Batch 0003/0043 | Averaged Loss: 0.1329
+Epoch: 064/100 | Batch 0004/0043 | Averaged Loss: 0.1307
+Epoch: 064/100 | Batch 0005/0043 | Averaged Loss: 0.0904
+Epoch: 064/100 | Batch 0006/0043 | Averaged Loss: 0.1116
+Epoch: 064/100 | Batch 0007/0043 | Averaged Loss: 0.1103
+Epoch: 064/100 | Batch 0008/0043 | Averaged Loss: 0.0936
+Epoch: 064/100 | Batch 0009/0043 | Averaged Loss: 0.0986
+Epoch: 064/100 | Batch 0010/0043 | Averaged Loss: 0.1024
+Epoch: 064/100 | Batch 0011/0043 | Averaged Loss: 0.0965
+Epoch: 064/100 | Batch 0012/0043 | Averaged Loss: 0.1095
+Epoch: 064/100 | Batch 0013/0043 | Averaged Loss: 0.0651
+Epoch: 064/100 | Batch 0014/0043 | Averaged Loss: 0.0906
+Epoch: 064/100 | Batch 0015/0043 | Averaged Loss: 0.1034
+Epoch: 064/100 | Batch 0016/0043 | Averaged Loss: 0.0810
+Epoch: 064/100 | Batch 0017/0043 | Averaged Loss: 0.1443
+Epoch: 064/100 | Batch 0018/0043 | Averaged Loss: 0.1157
+Epoch: 064/100 | Batch 0019/0043 | Averaged Loss: 0.0990
+Epoch: 064/100 | Batch 0020/0043 | Averaged Loss: 0.1040
+Epoch: 064/100 | Batch 0021/0043 | Averaged Loss: 0.1130
+Epoch: 064/100 | Batch 0022/0043 | Averaged Loss: 0.0930
+Epoch: 064/100 | Batch 0023/0043 | Averaged Loss: 0.1144
+Epoch: 064/100 | Batch 0024/0043 | Averaged Loss: 0.1640
+Epoch: 064/100 | Batch 0025/0043 | Averaged Loss: 0.1266
+Epoch: 064/100 | Batch 0026/0043 | Averaged Loss: 0.0775
+Epoch: 064/100 | Batch 0027/0043 | Averaged Loss: 0.1216
+Epoch: 064/100 | Batch 0028/0043 | Averaged Loss: 0.1130
+Epoch: 064/100 | Batch 0029/0043 | Averaged Loss: 0.1422
+Epoch: 064/100 | Batch 0030/0043 | Averaged Loss: 0.1438
+Epoch: 064/100 | Batch 0031/0043 | Averaged Loss: 0.1173
+Epoch: 064/100 | Batch 0032/0043 | Averaged Loss: 0.1390
+Epoch: 064/100 | Batch 0033/0043 | Averaged Loss: 0.0955
+Epoch: 064/100 | Batch 0034/0043 | Averaged Loss: 0.1332
+Epoch: 064/100 | Batch 0035/0043 | Averaged Loss: 0.1306
+Epoch: 064/100 | Batch 0036/0043 | Averaged Loss: 0.1544
+Epoch: 064/100 | Batch 0037/0043 | Averaged Loss: 0.1268
+Epoch: 064/100 | Batch 0038/0043 | Averaged Loss: 0.1263
+Epoch: 064/100 | Batch 0039/0043 | Averaged Loss: 0.1208
+Epoch: 064/100 | Batch 0040/0043 | Averaged Loss: 0.1222
+Epoch: 064/100 | Batch 0041/0043 | Averaged Loss: 0.1184
+Epoch: 064/100 | Batch 0042/0043 | Averaged Loss: 0.1106
+Epoch: 064/100 | Train: 96.73% | Validation: 75.12%
+Time elapsed: 9.504060745239258 min
+Epoch: 065/100 | Batch 0000/0043 | Averaged Loss: 0.1275
+Epoch: 065/100 | Batch 0001/0043 | Averaged Loss: 0.0779
+Epoch: 065/100 | Batch 0002/0043 | Averaged Loss: 0.0902
+Epoch: 065/100 | Batch 0003/0043 | Averaged Loss: 0.0900
+Epoch: 065/100 | Batch 0004/0043 | Averaged Loss: 0.1050
+Epoch: 065/100 | Batch 0005/0043 | Averaged Loss: 0.0816
+Epoch: 065/100 | Batch 0006/0043 | Averaged Loss: 0.0655
+Epoch: 065/100 | Batch 0007/0043 | Averaged Loss: 0.0815
+Epoch: 065/100 | Batch 0008/0043 | Averaged Loss: 0.1316
+Epoch: 065/100 | Batch 0009/0043 | Averaged Loss: 0.0855
+Epoch: 065/100 | Batch 0010/0043 | Averaged Loss: 0.0964
+Epoch: 065/100 | Batch 0011/0043 | Averaged Loss: 0.0859
+Epoch: 065/100 | Batch 0012/0043 | Averaged Loss: 0.0979
+Epoch: 065/100 | Batch 0013/0043 | Averaged Loss: 0.0960
+Epoch: 065/100 | Batch 0014/0043 | Averaged Loss: 0.0982
+Epoch: 065/100 | Batch 0015/0043 | Averaged Loss: 0.1125
+Epoch: 065/100 | Batch 0016/0043 | Averaged Loss: 0.1028
+Epoch: 065/100 | Batch 0017/0043 | Averaged Loss: 0.0588
+Epoch: 065/100 | Batch 0018/0043 | Averaged Loss: 0.0988
+Epoch: 065/100 | Batch 0019/0043 | Averaged Loss: 0.0818
+Epoch: 065/100 | Batch 0020/0043 | Averaged Loss: 0.1139
+Epoch: 065/100 | Batch 0021/0043 | Averaged Loss: 0.0984
+Epoch: 065/100 | Batch 0022/0043 | Averaged Loss: 0.0668
+Epoch: 065/100 | Batch 0023/0043 | Averaged Loss: 0.1496
+Epoch: 065/100 | Batch 0024/0043 | Averaged Loss: 0.0951
+Epoch: 065/100 | Batch 0025/0043 | Averaged Loss: 0.0885
+Epoch: 065/100 | Batch 0026/0043 | Averaged Loss: 0.0912
+Epoch: 065/100 | Batch 0027/0043 | Averaged Loss: 0.0739
+Epoch: 065/100 | Batch 0028/0043 | Averaged Loss: 0.1240
+Epoch: 065/100 | Batch 0029/0043 | Averaged Loss: 0.0899
+Epoch: 065/100 | Batch 0030/0043 | Averaged Loss: 0.0933
+Epoch: 065/100 | Batch 0031/0043 | Averaged Loss: 0.1143
+Epoch: 065/100 | Batch 0032/0043 | Averaged Loss: 0.1263
+Epoch: 065/100 | Batch 0033/0043 | Averaged Loss: 0.0922
+Epoch: 065/100 | Batch 0034/0043 | Averaged Loss: 0.0968
+Epoch: 065/100 | Batch 0035/0043 | Averaged Loss: 0.1436
+Epoch: 065/100 | Batch 0036/0043 | Averaged Loss: 0.1213
+Epoch: 065/100 | Batch 0037/0043 | Averaged Loss: 0.0901
+Epoch: 065/100 | Batch 0038/0043 | Averaged Loss: 0.1067
+Epoch: 065/100 | Batch 0039/0043 | Averaged Loss: 0.1322
+Epoch: 065/100 | Batch 0040/0043 | Averaged Loss: 0.0936
+Epoch: 065/100 | Batch 0041/0043 | Averaged Loss: 0.1181
+Epoch: 065/100 | Batch 0042/0043 | Averaged Loss: 0.0999
+Epoch: 065/100 | Train: 96.61% | Validation: 74.46%
+Time elapsed: 9.652099609375 min
+Epoch: 066/100 | Batch 0000/0043 | Averaged Loss: 0.0802
+Epoch: 066/100 | Batch 0001/0043 | Averaged Loss: 0.0879
+Epoch: 066/100 | Batch 0002/0043 | Averaged Loss: 0.0897
+Epoch: 066/100 | Batch 0003/0043 | Averaged Loss: 0.0853
+Epoch: 066/100 | Batch 0004/0043 | Averaged Loss: 0.0875
+Epoch: 066/100 | Batch 0005/0043 | Averaged Loss: 0.0782
+Epoch: 066/100 | Batch 0006/0043 | Averaged Loss: 0.0889
+Epoch: 066/100 | Batch 0007/0043 | Averaged Loss: 0.1041
+Epoch: 066/100 | Batch 0008/0043 | Averaged Loss: 0.0941
+Epoch: 066/100 | Batch 0009/0043 | Averaged Loss: 0.1024
+Epoch: 066/100 | Batch 0010/0043 | Averaged Loss: 0.0804
+Epoch: 066/100 | Batch 0011/0043 | Averaged Loss: 0.0852
+Epoch: 066/100 | Batch 0012/0043 | Averaged Loss: 0.1228
+Epoch: 066/100 | Batch 0013/0043 | Averaged Loss: 0.1381
+Epoch: 066/100 | Batch 0014/0043 | Averaged Loss: 0.1123
+Epoch: 066/100 | Batch 0015/0043 | Averaged Loss: 0.0891
+Epoch: 066/100 | Batch 0016/0043 | Averaged Loss: 0.0766
+Epoch: 066/100 | Batch 0017/0043 | Averaged Loss: 0.1151
+Epoch: 066/100 | Batch 0018/0043 | Averaged Loss: 0.0991
+Epoch: 066/100 | Batch 0019/0043 | Averaged Loss: 0.0968
+Epoch: 066/100 | Batch 0020/0043 | Averaged Loss: 0.1062
+Epoch: 066/100 | Batch 0021/0043 | Averaged Loss: 0.0893
+Epoch: 066/100 | Batch 0022/0043 | Averaged Loss: 0.0879
+Epoch: 066/100 | Batch 0023/0043 | Averaged Loss: 0.0883
+Epoch: 066/100 | Batch 0024/0043 | Averaged Loss: 0.0844
+Epoch: 066/100 | Batch 0025/0043 | Averaged Loss: 0.0963
+Epoch: 066/100 | Batch 0026/0043 | Averaged Loss: 0.1015
+Epoch: 066/100 | Batch 0027/0043 | Averaged Loss: 0.0880
+Epoch: 066/100 | Batch 0028/0043 | Averaged Loss: 0.1233
+Epoch: 066/100 | Batch 0029/0043 | Averaged Loss: 0.0968
+Epoch: 066/100 | Batch 0030/0043 | Averaged Loss: 0.0722
+Epoch: 066/100 | Batch 0031/0043 | Averaged Loss: 0.1090
+Epoch: 066/100 | Batch 0032/0043 | Averaged Loss: 0.1056
+Epoch: 066/100 | Batch 0033/0043 | Averaged Loss: 0.0980
+Epoch: 066/100 | Batch 0034/0043 | Averaged Loss: 0.1415
+Epoch: 066/100 | Batch 0035/0043 | Averaged Loss: 0.0830
+Epoch: 066/100 | Batch 0036/0043 | Averaged Loss: 0.0840
+Epoch: 066/100 | Batch 0037/0043 | Averaged Loss: 0.0914
+Epoch: 066/100 | Batch 0038/0043 | Averaged Loss: 0.1243
+Epoch: 066/100 | Batch 0039/0043 | Averaged Loss: 0.0835
+Epoch: 066/100 | Batch 0040/0043 | Averaged Loss: 0.0669
+Epoch: 066/100 | Batch 0041/0043 | Averaged Loss: 0.0880
+Epoch: 066/100 | Batch 0042/0043 | Averaged Loss: 0.0991
+Epoch: 066/100 | Train: 97.54% | Validation: 75.76%
+Time elapsed: 9.800562858581543 min
+Epoch: 067/100 | Batch 0000/0043 | Averaged Loss: 0.0739
+Epoch: 067/100 | Batch 0001/0043 | Averaged Loss: 0.0718
+Epoch: 067/100 | Batch 0002/0043 | Averaged Loss: 0.0951
+Epoch: 067/100 | Batch 0003/0043 | Averaged Loss: 0.0752
+Epoch: 067/100 | Batch 0004/0043 | Averaged Loss: 0.0705
+Epoch: 067/100 | Batch 0005/0043 | Averaged Loss: 0.0847
+Epoch: 067/100 | Batch 0006/0043 | Averaged Loss: 0.0755
+Epoch: 067/100 | Batch 0007/0043 | Averaged Loss: 0.0748
+Epoch: 067/100 | Batch 0008/0043 | Averaged Loss: 0.0805
+Epoch: 067/100 | Batch 0009/0043 | Averaged Loss: 0.0916
+Epoch: 067/100 | Batch 0010/0043 | Averaged Loss: 0.0924
+Epoch: 067/100 | Batch 0011/0043 | Averaged Loss: 0.0980
+Epoch: 067/100 | Batch 0012/0043 | Averaged Loss: 0.0815
+Epoch: 067/100 | Batch 0013/0043 | Averaged Loss: 0.0931
+Epoch: 067/100 | Batch 0014/0043 | Averaged Loss: 0.1148
+Epoch: 067/100 | Batch 0015/0043 | Averaged Loss: 0.0986
+Epoch: 067/100 | Batch 0016/0043 | Averaged Loss: 0.0866
+Epoch: 067/100 | Batch 0017/0043 | Averaged Loss: 0.0830
+Epoch: 067/100 | Batch 0018/0043 | Averaged Loss: 0.0799
+Epoch: 067/100 | Batch 0019/0043 | Averaged Loss: 0.1329
+Epoch: 067/100 | Batch 0020/0043 | Averaged Loss: 0.1231
+Epoch: 067/100 | Batch 0021/0043 | Averaged Loss: 0.0817
+Epoch: 067/100 | Batch 0022/0043 | Averaged Loss: 0.0976
+Epoch: 067/100 | Batch 0023/0043 | Averaged Loss: 0.1289
+Epoch: 067/100 | Batch 0024/0043 | Averaged Loss: 0.0858
+Epoch: 067/100 | Batch 0025/0043 | Averaged Loss: 0.0854
+Epoch: 067/100 | Batch 0026/0043 | Averaged Loss: 0.0836
+Epoch: 067/100 | Batch 0027/0043 | Averaged Loss: 0.1005
+Epoch: 067/100 | Batch 0028/0043 | Averaged Loss: 0.0757
+Epoch: 067/100 | Batch 0029/0043 | Averaged Loss: 0.0906
+Epoch: 067/100 | Batch 0030/0043 | Averaged Loss: 0.0981
+Epoch: 067/100 | Batch 0031/0043 | Averaged Loss: 0.1099
+Epoch: 067/100 | Batch 0032/0043 | Averaged Loss: 0.0899
+Epoch: 067/100 | Batch 0033/0043 | Averaged Loss: 0.0822
+Epoch: 067/100 | Batch 0034/0043 | Averaged Loss: 0.0720
+Epoch: 067/100 | Batch 0035/0043 | Averaged Loss: 0.0846
+Epoch: 067/100 | Batch 0036/0043 | Averaged Loss: 0.0790
+Epoch: 067/100 | Batch 0037/0043 | Averaged Loss: 0.0727
+Epoch: 067/100 | Batch 0038/0043 | Averaged Loss: 0.0877
+Epoch: 067/100 | Batch 0039/0043 | Averaged Loss: 0.0909
+Epoch: 067/100 | Batch 0040/0043 | Averaged Loss: 0.1018
+Epoch: 067/100 | Batch 0041/0043 | Averaged Loss: 0.1189
+Epoch: 067/100 | Batch 0042/0043 | Averaged Loss: 0.0794
+Epoch: 067/100 | Train: 97.43% | Validation: 74.12%
+Time elapsed: 9.948922157287598 min
+Epoch: 068/100 | Batch 0000/0043 | Averaged Loss: 0.0663
+Epoch: 068/100 | Batch 0001/0043 | Averaged Loss: 0.0576
+Epoch: 068/100 | Batch 0002/0043 | Averaged Loss: 0.0659
+Epoch: 068/100 | Batch 0003/0043 | Averaged Loss: 0.0591
+Epoch: 068/100 | Batch 0004/0043 | Averaged Loss: 0.0825
+Epoch: 068/100 | Batch 0005/0043 | Averaged Loss: 0.0529
+Epoch: 068/100 | Batch 0006/0043 | Averaged Loss: 0.0655
+Epoch: 068/100 | Batch 0007/0043 | Averaged Loss: 0.0708
+Epoch: 068/100 | Batch 0008/0043 | Averaged Loss: 0.0816
+Epoch: 068/100 | Batch 0009/0043 | Averaged Loss: 0.0912
+Epoch: 068/100 | Batch 0010/0043 | Averaged Loss: 0.0643
+Epoch: 068/100 | Batch 0011/0043 | Averaged Loss: 0.0534
+Epoch: 068/100 | Batch 0012/0043 | Averaged Loss: 0.0787
+Epoch: 068/100 | Batch 0013/0043 | Averaged Loss: 0.0712
+Epoch: 068/100 | Batch 0014/0043 | Averaged Loss: 0.0887
+Epoch: 068/100 | Batch 0015/0043 | Averaged Loss: 0.0911
+Epoch: 068/100 | Batch 0016/0043 | Averaged Loss: 0.0644
+Epoch: 068/100 | Batch 0017/0043 | Averaged Loss: 0.1001
+Epoch: 068/100 | Batch 0018/0043 | Averaged Loss: 0.0910
+Epoch: 068/100 | Batch 0019/0043 | Averaged Loss: 0.0696
+Epoch: 068/100 | Batch 0020/0043 | Averaged Loss: 0.0714
+Epoch: 068/100 | Batch 0021/0043 | Averaged Loss: 0.0582
+Epoch: 068/100 | Batch 0022/0043 | Averaged Loss: 0.0746
+Epoch: 068/100 | Batch 0023/0043 | Averaged Loss: 0.0845
+Epoch: 068/100 | Batch 0024/0043 | Averaged Loss: 0.0956
+Epoch: 068/100 | Batch 0025/0043 | Averaged Loss: 0.0692
+Epoch: 068/100 | Batch 0026/0043 | Averaged Loss: 0.0845
+Epoch: 068/100 | Batch 0027/0043 | Averaged Loss: 0.0692
+Epoch: 068/100 | Batch 0028/0043 | Averaged Loss: 0.0807
+Epoch: 068/100 | Batch 0029/0043 | Averaged Loss: 0.0805
+Epoch: 068/100 | Batch 0030/0043 | Averaged Loss: 0.1033
+Epoch: 068/100 | Batch 0031/0043 | Averaged Loss: 0.0939
+Epoch: 068/100 | Batch 0032/0043 | Averaged Loss: 0.0822
+Epoch: 068/100 | Batch 0033/0043 | Averaged Loss: 0.0767
+Epoch: 068/100 | Batch 0034/0043 | Averaged Loss: 0.0968
+Epoch: 068/100 | Batch 0035/0043 | Averaged Loss: 0.1112
+Epoch: 068/100 | Batch 0036/0043 | Averaged Loss: 0.1137
+Epoch: 068/100 | Batch 0037/0043 | Averaged Loss: 0.1048
+Epoch: 068/100 | Batch 0038/0043 | Averaged Loss: 0.1143
+Epoch: 068/100 | Batch 0039/0043 | Averaged Loss: 0.0847
+Epoch: 068/100 | Batch 0040/0043 | Averaged Loss: 0.0792
+Epoch: 068/100 | Batch 0041/0043 | Averaged Loss: 0.1202
+Epoch: 068/100 | Batch 0042/0043 | Averaged Loss: 0.1085
+Epoch: 068/100 | Train: 98.12% | Validation: 75.63%
+Time elapsed: 10.097262382507324 min
+Epoch: 069/100 | Batch 0000/0043 | Averaged Loss: 0.0635
+Epoch: 069/100 | Batch 0001/0043 | Averaged Loss: 0.0683
+Epoch: 069/100 | Batch 0002/0043 | Averaged Loss: 0.0717
+Epoch: 069/100 | Batch 0003/0043 | Averaged Loss: 0.0689
+Epoch: 069/100 | Batch 0004/0043 | Averaged Loss: 0.0830
+Epoch: 069/100 | Batch 0005/0043 | Averaged Loss: 0.0780
+Epoch: 069/100 | Batch 0006/0043 | Averaged Loss: 0.0604
+Epoch: 069/100 | Batch 0007/0043 | Averaged Loss: 0.0843
+Epoch: 069/100 | Batch 0008/0043 | Averaged Loss: 0.0735
+Epoch: 069/100 | Batch 0009/0043 | Averaged Loss: 0.0614
+Epoch: 069/100 | Batch 0010/0043 | Averaged Loss: 0.1030
+Epoch: 069/100 | Batch 0011/0043 | Averaged Loss: 0.0902
+Epoch: 069/100 | Batch 0012/0043 | Averaged Loss: 0.0781
+Epoch: 069/100 | Batch 0013/0043 | Averaged Loss: 0.0968
+Epoch: 069/100 | Batch 0014/0043 | Averaged Loss: 0.0783
+Epoch: 069/100 | Batch 0015/0043 | Averaged Loss: 0.0633
+Epoch: 069/100 | Batch 0016/0043 | Averaged Loss: 0.0620
+Epoch: 069/100 | Batch 0017/0043 | Averaged Loss: 0.0763
+Epoch: 069/100 | Batch 0018/0043 | Averaged Loss: 0.0992
+Epoch: 069/100 | Batch 0019/0043 | Averaged Loss: 0.0860
+Epoch: 069/100 | Batch 0020/0043 | Averaged Loss: 0.0608
+Epoch: 069/100 | Batch 0021/0043 | Averaged Loss: 0.0583
+Epoch: 069/100 | Batch 0022/0043 | Averaged Loss: 0.0823
+Epoch: 069/100 | Batch 0023/0043 | Averaged Loss: 0.1143
+Epoch: 069/100 | Batch 0024/0043 | Averaged Loss: 0.0677
+Epoch: 069/100 | Batch 0025/0043 | Averaged Loss: 0.0834
+Epoch: 069/100 | Batch 0026/0043 | Averaged Loss: 0.0984
+Epoch: 069/100 | Batch 0027/0043 | Averaged Loss: 0.0867
+Epoch: 069/100 | Batch 0028/0043 | Averaged Loss: 0.0657
+Epoch: 069/100 | Batch 0029/0043 | Averaged Loss: 0.1030
+Epoch: 069/100 | Batch 0030/0043 | Averaged Loss: 0.0783
+Epoch: 069/100 | Batch 0031/0043 | Averaged Loss: 0.1166
+Epoch: 069/100 | Batch 0032/0043 | Averaged Loss: 0.1218
+Epoch: 069/100 | Batch 0033/0043 | Averaged Loss: 0.0540
+Epoch: 069/100 | Batch 0034/0043 | Averaged Loss: 0.0961
+Epoch: 069/100 | Batch 0035/0043 | Averaged Loss: 0.0836
+Epoch: 069/100 | Batch 0036/0043 | Averaged Loss: 0.0888
+Epoch: 069/100 | Batch 0037/0043 | Averaged Loss: 0.0902
+Epoch: 069/100 | Batch 0038/0043 | Averaged Loss: 0.0785
+Epoch: 069/100 | Batch 0039/0043 | Averaged Loss: 0.0875
+Epoch: 069/100 | Batch 0040/0043 | Averaged Loss: 0.0800
+Epoch: 069/100 | Batch 0041/0043 | Averaged Loss: 0.0769
+Epoch: 069/100 | Batch 0042/0043 | Averaged Loss: 0.0683
+Epoch: 069/100 | Train: 97.99% | Validation: 74.54%
+Time elapsed: 10.24571704864502 min
+Epoch: 070/100 | Batch 0000/0043 | Averaged Loss: 0.0574
+Epoch: 070/100 | Batch 0001/0043 | Averaged Loss: 0.0542
+Epoch: 070/100 | Batch 0002/0043 | Averaged Loss: 0.0811
+Epoch: 070/100 | Batch 0003/0043 | Averaged Loss: 0.0660
+Epoch: 070/100 | Batch 0004/0043 | Averaged Loss: 0.0640
+Epoch: 070/100 | Batch 0005/0043 | Averaged Loss: 0.0571
+Epoch: 070/100 | Batch 0006/0043 | Averaged Loss: 0.0753
+Epoch: 070/100 | Batch 0007/0043 | Averaged Loss: 0.0654
+Epoch: 070/100 | Batch 0008/0043 | Averaged Loss: 0.0590
+Epoch: 070/100 | Batch 0009/0043 | Averaged Loss: 0.0399
+Epoch: 070/100 | Batch 0010/0043 | Averaged Loss: 0.0632
+Epoch: 070/100 | Batch 0011/0043 | Averaged Loss: 0.0828
+Epoch: 070/100 | Batch 0012/0043 | Averaged Loss: 0.0687
+Epoch: 070/100 | Batch 0013/0043 | Averaged Loss: 0.0765
+Epoch: 070/100 | Batch 0014/0043 | Averaged Loss: 0.0786
+Epoch: 070/100 | Batch 0015/0043 | Averaged Loss: 0.0634
+Epoch: 070/100 | Batch 0016/0043 | Averaged Loss: 0.0900
+Epoch: 070/100 | Batch 0017/0043 | Averaged Loss: 0.1102
+Epoch: 070/100 | Batch 0018/0043 | Averaged Loss: 0.0788
+Epoch: 070/100 | Batch 0019/0043 | Averaged Loss: 0.0705
+Epoch: 070/100 | Batch 0020/0043 | Averaged Loss: 0.0688
+Epoch: 070/100 | Batch 0021/0043 | Averaged Loss: 0.0832
+Epoch: 070/100 | Batch 0022/0043 | Averaged Loss: 0.0765
+Epoch: 070/100 | Batch 0023/0043 | Averaged Loss: 0.1029
+Epoch: 070/100 | Batch 0024/0043 | Averaged Loss: 0.0917
+Epoch: 070/100 | Batch 0025/0043 | Averaged Loss: 0.0933
+Epoch: 070/100 | Batch 0026/0043 | Averaged Loss: 0.0927
+Epoch: 070/100 | Batch 0027/0043 | Averaged Loss: 0.0767
+Epoch: 070/100 | Batch 0028/0043 | Averaged Loss: 0.1212
+Epoch: 070/100 | Batch 0029/0043 | Averaged Loss: 0.0799
+Epoch: 070/100 | Batch 0030/0043 | Averaged Loss: 0.0983
+Epoch: 070/100 | Batch 0031/0043 | Averaged Loss: 0.1351
+Epoch: 070/100 | Batch 0032/0043 | Averaged Loss: 0.0811
+Epoch: 070/100 | Batch 0033/0043 | Averaged Loss: 0.1034
+Epoch: 070/100 | Batch 0034/0043 | Averaged Loss: 0.0919
+Epoch: 070/100 | Batch 0035/0043 | Averaged Loss: 0.0899
+Epoch: 070/100 | Batch 0036/0043 | Averaged Loss: 0.1203
+Epoch: 070/100 | Batch 0037/0043 | Averaged Loss: 0.1327
+Epoch: 070/100 | Batch 0038/0043 | Averaged Loss: 0.0984
+Epoch: 070/100 | Batch 0039/0043 | Averaged Loss: 0.0685
+Epoch: 070/100 | Batch 0040/0043 | Averaged Loss: 0.0756
+Epoch: 070/100 | Batch 0041/0043 | Averaged Loss: 0.0820
+Epoch: 070/100 | Batch 0042/0043 | Averaged Loss: 0.1011
+Epoch: 070/100 | Train: 97.45% | Validation: 74.34%
+Time elapsed: 10.39417552947998 min
+Epoch: 071/100 | Batch 0000/0043 | Averaged Loss: 0.0639
+Epoch: 071/100 | Batch 0001/0043 | Averaged Loss: 0.0772
+Epoch: 071/100 | Batch 0002/0043 | Averaged Loss: 0.0624
+Epoch: 071/100 | Batch 0003/0043 | Averaged Loss: 0.0666
+Epoch: 071/100 | Batch 0004/0043 | Averaged Loss: 0.0682
+Epoch: 071/100 | Batch 0005/0043 | Averaged Loss: 0.0693
+Epoch: 071/100 | Batch 0006/0043 | Averaged Loss: 0.0657
+Epoch: 071/100 | Batch 0007/0043 | Averaged Loss: 0.1082
+Epoch: 071/100 | Batch 0008/0043 | Averaged Loss: 0.0950
+Epoch: 071/100 | Batch 0009/0043 | Averaged Loss: 0.0817
+Epoch: 071/100 | Batch 0010/0043 | Averaged Loss: 0.0839
+Epoch: 071/100 | Batch 0011/0043 | Averaged Loss: 0.0869
+Epoch: 071/100 | Batch 0012/0043 | Averaged Loss: 0.1031
+Epoch: 071/100 | Batch 0013/0043 | Averaged Loss: 0.0698
+Epoch: 071/100 | Batch 0014/0043 | Averaged Loss: 0.0824
+Epoch: 071/100 | Batch 0015/0043 | Averaged Loss: 0.0721
+Epoch: 071/100 | Batch 0016/0043 | Averaged Loss: 0.1026
+Epoch: 071/100 | Batch 0017/0043 | Averaged Loss: 0.0851
+Epoch: 071/100 | Batch 0018/0043 | Averaged Loss: 0.0902
+Epoch: 071/100 | Batch 0019/0043 | Averaged Loss: 0.0930
+Epoch: 071/100 | Batch 0020/0043 | Averaged Loss: 0.0921
+Epoch: 071/100 | Batch 0021/0043 | Averaged Loss: 0.1016
+Epoch: 071/100 | Batch 0022/0043 | Averaged Loss: 0.0663
+Epoch: 071/100 | Batch 0023/0043 | Averaged Loss: 0.0713
+Epoch: 071/100 | Batch 0024/0043 | Averaged Loss: 0.0861
+Epoch: 071/100 | Batch 0025/0043 | Averaged Loss: 0.0993
+Epoch: 071/100 | Batch 0026/0043 | Averaged Loss: 0.0972
+Epoch: 071/100 | Batch 0027/0043 | Averaged Loss: 0.0821
+Epoch: 071/100 | Batch 0028/0043 | Averaged Loss: 0.0997
+Epoch: 071/100 | Batch 0029/0043 | Averaged Loss: 0.0957
+Epoch: 071/100 | Batch 0030/0043 | Averaged Loss: 0.0882
+Epoch: 071/100 | Batch 0031/0043 | Averaged Loss: 0.0978
+Epoch: 071/100 | Batch 0032/0043 | Averaged Loss: 0.1065
+Epoch: 071/100 | Batch 0033/0043 | Averaged Loss: 0.1119
+Epoch: 071/100 | Batch 0034/0043 | Averaged Loss: 0.1088
+Epoch: 071/100 | Batch 0035/0043 | Averaged Loss: 0.0648
+Epoch: 071/100 | Batch 0036/0043 | Averaged Loss: 0.0639
+Epoch: 071/100 | Batch 0037/0043 | Averaged Loss: 0.0994
+Epoch: 071/100 | Batch 0038/0043 | Averaged Loss: 0.0942
+Epoch: 071/100 | Batch 0039/0043 | Averaged Loss: 0.0893
+Epoch: 071/100 | Batch 0040/0043 | Averaged Loss: 0.0904
+Epoch: 071/100 | Batch 0041/0043 | Averaged Loss: 0.0655
+Epoch: 071/100 | Batch 0042/0043 | Averaged Loss: 0.0705
+Epoch: 071/100 | Train: 97.69% | Validation: 75.15%
+Time elapsed: 10.542407035827637 min
+Epoch: 072/100 | Batch 0000/0043 | Averaged Loss: 0.0610
+Epoch: 072/100 | Batch 0001/0043 | Averaged Loss: 0.0589
+Epoch: 072/100 | Batch 0002/0043 | Averaged Loss: 0.0719
+Epoch: 072/100 | Batch 0003/0043 | Averaged Loss: 0.0624
+Epoch: 072/100 | Batch 0004/0043 | Averaged Loss: 0.0892
+Epoch: 072/100 | Batch 0005/0043 | Averaged Loss: 0.0575
+Epoch: 072/100 | Batch 0006/0043 | Averaged Loss: 0.0807
+Epoch: 072/100 | Batch 0007/0043 | Averaged Loss: 0.0454
+Epoch: 072/100 | Batch 0008/0043 | Averaged Loss: 0.0731
+Epoch: 072/100 | Batch 0009/0043 | Averaged Loss: 0.0942
+Epoch: 072/100 | Batch 0010/0043 | Averaged Loss: 0.0740
+Epoch: 072/100 | Batch 0011/0043 | Averaged Loss: 0.0991
+Epoch: 072/100 | Batch 0012/0043 | Averaged Loss: 0.0651
+Epoch: 072/100 | Batch 0013/0043 | Averaged Loss: 0.0569
+Epoch: 072/100 | Batch 0014/0043 | Averaged Loss: 0.0544
+Epoch: 072/100 | Batch 0015/0043 | Averaged Loss: 0.0610
+Epoch: 072/100 | Batch 0016/0043 | Averaged Loss: 0.0661
+Epoch: 072/100 | Batch 0017/0043 | Averaged Loss: 0.0605
+Epoch: 072/100 | Batch 0018/0043 | Averaged Loss: 0.0851
+Epoch: 072/100 | Batch 0019/0043 | Averaged Loss: 0.0694
+Epoch: 072/100 | Batch 0020/0043 | Averaged Loss: 0.0449
+Epoch: 072/100 | Batch 0021/0043 | Averaged Loss: 0.0593
+Epoch: 072/100 | Batch 0022/0043 | Averaged Loss: 0.0531
+Epoch: 072/100 | Batch 0023/0043 | Averaged Loss: 0.0605
+Epoch: 072/100 | Batch 0024/0043 | Averaged Loss: 0.0764
+Epoch: 072/100 | Batch 0025/0043 | Averaged Loss: 0.0608
+Epoch: 072/100 | Batch 0026/0043 | Averaged Loss: 0.0801
+Epoch: 072/100 | Batch 0027/0043 | Averaged Loss: 0.0863
+Epoch: 072/100 | Batch 0028/0043 | Averaged Loss: 0.0929
+Epoch: 072/100 | Batch 0029/0043 | Averaged Loss: 0.0631
+Epoch: 072/100 | Batch 0030/0043 | Averaged Loss: 0.1059
+Epoch: 072/100 | Batch 0031/0043 | Averaged Loss: 0.0717
+Epoch: 072/100 | Batch 0032/0043 | Averaged Loss: 0.1125
+Epoch: 072/100 | Batch 0033/0043 | Averaged Loss: 0.0889
+Epoch: 072/100 | Batch 0034/0043 | Averaged Loss: 0.0515
+Epoch: 072/100 | Batch 0035/0043 | Averaged Loss: 0.1000
+Epoch: 072/100 | Batch 0036/0043 | Averaged Loss: 0.1186
+Epoch: 072/100 | Batch 0037/0043 | Averaged Loss: 0.0974
+Epoch: 072/100 | Batch 0038/0043 | Averaged Loss: 0.1204
+Epoch: 072/100 | Batch 0039/0043 | Averaged Loss: 0.1332
+Epoch: 072/100 | Batch 0040/0043 | Averaged Loss: 0.0875
+Epoch: 072/100 | Batch 0041/0043 | Averaged Loss: 0.0815
+Epoch: 072/100 | Batch 0042/0043 | Averaged Loss: 0.0884
+Epoch: 072/100 | Train: 96.51% | Validation: 74.73%
+Time elapsed: 10.691036224365234 min
+Epoch: 073/100 | Batch 0000/0043 | Averaged Loss: 0.1045
+Epoch: 073/100 | Batch 0001/0043 | Averaged Loss: 0.1044
+Epoch: 073/100 | Batch 0002/0043 | Averaged Loss: 0.0782
+Epoch: 073/100 | Batch 0003/0043 | Averaged Loss: 0.0939
+Epoch: 073/100 | Batch 0004/0043 | Averaged Loss: 0.0867
+Epoch: 073/100 | Batch 0005/0043 | Averaged Loss: 0.0952
+Epoch: 073/100 | Batch 0006/0043 | Averaged Loss: 0.0933
+Epoch: 073/100 | Batch 0007/0043 | Averaged Loss: 0.0675
+Epoch: 073/100 | Batch 0008/0043 | Averaged Loss: 0.0906
+Epoch: 073/100 | Batch 0009/0043 | Averaged Loss: 0.1091
+Epoch: 073/100 | Batch 0010/0043 | Averaged Loss: 0.0918
+Epoch: 073/100 | Batch 0011/0043 | Averaged Loss: 0.0645
+Epoch: 073/100 | Batch 0012/0043 | Averaged Loss: 0.0450
+Epoch: 073/100 | Batch 0013/0043 | Averaged Loss: 0.0516
+Epoch: 073/100 | Batch 0014/0043 | Averaged Loss: 0.1366
+Epoch: 073/100 | Batch 0015/0043 | Averaged Loss: 0.0668
+Epoch: 073/100 | Batch 0016/0043 | Averaged Loss: 0.0727
+Epoch: 073/100 | Batch 0017/0043 | Averaged Loss: 0.0827
+Epoch: 073/100 | Batch 0018/0043 | Averaged Loss: 0.0691
+Epoch: 073/100 | Batch 0019/0043 | Averaged Loss: 0.0910
+Epoch: 073/100 | Batch 0020/0043 | Averaged Loss: 0.1041
+Epoch: 073/100 | Batch 0021/0043 | Averaged Loss: 0.0893
+Epoch: 073/100 | Batch 0022/0043 | Averaged Loss: 0.0480
+Epoch: 073/100 | Batch 0023/0043 | Averaged Loss: 0.0857
+Epoch: 073/100 | Batch 0024/0043 | Averaged Loss: 0.0818
+Epoch: 073/100 | Batch 0025/0043 | Averaged Loss: 0.1232
+Epoch: 073/100 | Batch 0026/0043 | Averaged Loss: 0.0825
+Epoch: 073/100 | Batch 0027/0043 | Averaged Loss: 0.1167
+Epoch: 073/100 | Batch 0028/0043 | Averaged Loss: 0.0933
+Epoch: 073/100 | Batch 0029/0043 | Averaged Loss: 0.0570
+Epoch: 073/100 | Batch 0030/0043 | Averaged Loss: 0.0886
+Epoch: 073/100 | Batch 0031/0043 | Averaged Loss: 0.0946
+Epoch: 073/100 | Batch 0032/0043 | Averaged Loss: 0.0844
+Epoch: 073/100 | Batch 0033/0043 | Averaged Loss: 0.0991
+Epoch: 073/100 | Batch 0034/0043 | Averaged Loss: 0.1187
+Epoch: 073/100 | Batch 0035/0043 | Averaged Loss: 0.1038
+Epoch: 073/100 | Batch 0036/0043 | Averaged Loss: 0.0638
+Epoch: 073/100 | Batch 0037/0043 | Averaged Loss: 0.1127
+Epoch: 073/100 | Batch 0038/0043 | Averaged Loss: 0.1037
+Epoch: 073/100 | Batch 0039/0043 | Averaged Loss: 0.0908
+Epoch: 073/100 | Batch 0040/0043 | Averaged Loss: 0.0628
+Epoch: 073/100 | Batch 0041/0043 | Averaged Loss: 0.1035
+Epoch: 073/100 | Batch 0042/0043 | Averaged Loss: 0.0685
+Epoch: 073/100 | Train: 98.11% | Validation: 75.05%
+Time elapsed: 10.839608192443848 min
+Epoch: 074/100 | Batch 0000/0043 | Averaged Loss: 0.0554
+Epoch: 074/100 | Batch 0001/0043 | Averaged Loss: 0.0821
+Epoch: 074/100 | Batch 0002/0043 | Averaged Loss: 0.0812
+Epoch: 074/100 | Batch 0003/0043 | Averaged Loss: 0.0582
+Epoch: 074/100 | Batch 0004/0043 | Averaged Loss: 0.0592
+Epoch: 074/100 | Batch 0005/0043 | Averaged Loss: 0.0699
+Epoch: 074/100 | Batch 0006/0043 | Averaged Loss: 0.0843
+Epoch: 074/100 | Batch 0007/0043 | Averaged Loss: 0.0867
+Epoch: 074/100 | Batch 0008/0043 | Averaged Loss: 0.0472
+Epoch: 074/100 | Batch 0009/0043 | Averaged Loss: 0.0676
+Epoch: 074/100 | Batch 0010/0043 | Averaged Loss: 0.0587
+Epoch: 074/100 | Batch 0011/0043 | Averaged Loss: 0.0690
+Epoch: 074/100 | Batch 0012/0043 | Averaged Loss: 0.0562
+Epoch: 074/100 | Batch 0013/0043 | Averaged Loss: 0.0619
+Epoch: 074/100 | Batch 0014/0043 | Averaged Loss: 0.0916
+Epoch: 074/100 | Batch 0015/0043 | Averaged Loss: 0.0744
+Epoch: 074/100 | Batch 0016/0043 | Averaged Loss: 0.0743
+Epoch: 074/100 | Batch 0017/0043 | Averaged Loss: 0.0509
+Epoch: 074/100 | Batch 0018/0043 | Averaged Loss: 0.0279
+Epoch: 074/100 | Batch 0019/0043 | Averaged Loss: 0.0633
+Epoch: 074/100 | Batch 0020/0043 | Averaged Loss: 0.0466
+Epoch: 074/100 | Batch 0021/0043 | Averaged Loss: 0.0467
+Epoch: 074/100 | Batch 0022/0043 | Averaged Loss: 0.0698
+Epoch: 074/100 | Batch 0023/0043 | Averaged Loss: 0.0472
+Epoch: 074/100 | Batch 0024/0043 | Averaged Loss: 0.0583
+Epoch: 074/100 | Batch 0025/0043 | Averaged Loss: 0.0739
+Epoch: 074/100 | Batch 0026/0043 | Averaged Loss: 0.0710
+Epoch: 074/100 | Batch 0027/0043 | Averaged Loss: 0.0682
+Epoch: 074/100 | Batch 0028/0043 | Averaged Loss: 0.0696
+Epoch: 074/100 | Batch 0029/0043 | Averaged Loss: 0.0475
+Epoch: 074/100 | Batch 0030/0043 | Averaged Loss: 0.0508
+Epoch: 074/100 | Batch 0031/0043 | Averaged Loss: 0.0805
+Epoch: 074/100 | Batch 0032/0043 | Averaged Loss: 0.0570
+Epoch: 074/100 | Batch 0033/0043 | Averaged Loss: 0.0802
+Epoch: 074/100 | Batch 0034/0043 | Averaged Loss: 0.0838
+Epoch: 074/100 | Batch 0035/0043 | Averaged Loss: 0.0580
+Epoch: 074/100 | Batch 0036/0043 | Averaged Loss: 0.0782
+Epoch: 074/100 | Batch 0037/0043 | Averaged Loss: 0.0746
+Epoch: 074/100 | Batch 0038/0043 | Averaged Loss: 0.0873
+Epoch: 074/100 | Batch 0039/0043 | Averaged Loss: 0.0734
+Epoch: 074/100 | Batch 0040/0043 | Averaged Loss: 0.0830
+Epoch: 074/100 | Batch 0041/0043 | Averaged Loss: 0.0902
+Epoch: 074/100 | Batch 0042/0043 | Averaged Loss: 0.0664
+Epoch: 074/100 | Train: 97.70% | Validation: 74.46%
+Time elapsed: 10.987943649291992 min
+Epoch: 075/100 | Batch 0000/0043 | Averaged Loss: 0.0522
+Epoch: 075/100 | Batch 0001/0043 | Averaged Loss: 0.0906
+Epoch: 075/100 | Batch 0002/0043 | Averaged Loss: 0.0711
+Epoch: 075/100 | Batch 0003/0043 | Averaged Loss: 0.0394
+Epoch: 075/100 | Batch 0004/0043 | Averaged Loss: 0.0945
+Epoch: 075/100 | Batch 0005/0043 | Averaged Loss: 0.0732
+Epoch: 075/100 | Batch 0006/0043 | Averaged Loss: 0.0503
+Epoch: 075/100 | Batch 0007/0043 | Averaged Loss: 0.0811
+Epoch: 075/100 | Batch 0008/0043 | Averaged Loss: 0.0974
+Epoch: 075/100 | Batch 0009/0043 | Averaged Loss: 0.0667
+Epoch: 075/100 | Batch 0010/0043 | Averaged Loss: 0.0549
+Epoch: 075/100 | Batch 0011/0043 | Averaged Loss: 0.0808
+Epoch: 075/100 | Batch 0012/0043 | Averaged Loss: 0.0645
+Epoch: 075/100 | Batch 0013/0043 | Averaged Loss: 0.1009
+Epoch: 075/100 | Batch 0014/0043 | Averaged Loss: 0.0886
+Epoch: 075/100 | Batch 0015/0043 | Averaged Loss: 0.0765
+Epoch: 075/100 | Batch 0016/0043 | Averaged Loss: 0.0866
+Epoch: 075/100 | Batch 0017/0043 | Averaged Loss: 0.0902
+Epoch: 075/100 | Batch 0018/0043 | Averaged Loss: 0.0890
+Epoch: 075/100 | Batch 0019/0043 | Averaged Loss: 0.1255
+Epoch: 075/100 | Batch 0020/0043 | Averaged Loss: 0.0608
+Epoch: 075/100 | Batch 0021/0043 | Averaged Loss: 0.0963
+Epoch: 075/100 | Batch 0022/0043 | Averaged Loss: 0.0589
+Epoch: 075/100 | Batch 0023/0043 | Averaged Loss: 0.0700
+Epoch: 075/100 | Batch 0024/0043 | Averaged Loss: 0.0637
+Epoch: 075/100 | Batch 0025/0043 | Averaged Loss: 0.1079
+Epoch: 075/100 | Batch 0026/0043 | Averaged Loss: 0.0800
+Epoch: 075/100 | Batch 0027/0043 | Averaged Loss: 0.0754
+Epoch: 075/100 | Batch 0028/0043 | Averaged Loss: 0.0942
+Epoch: 075/100 | Batch 0029/0043 | Averaged Loss: 0.0982
+Epoch: 075/100 | Batch 0030/0043 | Averaged Loss: 0.0933
+Epoch: 075/100 | Batch 0031/0043 | Averaged Loss: 0.0936
+Epoch: 075/100 | Batch 0032/0043 | Averaged Loss: 0.0977
+Epoch: 075/100 | Batch 0033/0043 | Averaged Loss: 0.0629
+Epoch: 075/100 | Batch 0034/0043 | Averaged Loss: 0.0754
+Epoch: 075/100 | Batch 0035/0043 | Averaged Loss: 0.1091
+Epoch: 075/100 | Batch 0036/0043 | Averaged Loss: 0.1167
+Epoch: 075/100 | Batch 0037/0043 | Averaged Loss: 0.0530
+Epoch: 075/100 | Batch 0038/0043 | Averaged Loss: 0.0659
+Epoch: 075/100 | Batch 0039/0043 | Averaged Loss: 0.0800
+Epoch: 075/100 | Batch 0040/0043 | Averaged Loss: 0.0818
+Epoch: 075/100 | Batch 0041/0043 | Averaged Loss: 0.0796
+Epoch: 075/100 | Batch 0042/0043 | Averaged Loss: 0.0711
+Epoch: 075/100 | Train: 98.16% | Validation: 76.39%
+Time elapsed: 11.136483192443848 min
+Epoch: 076/100 | Batch 0000/0043 | Averaged Loss: 0.0553
+Epoch: 076/100 | Batch 0001/0043 | Averaged Loss: 0.0655
+Epoch: 076/100 | Batch 0002/0043 | Averaged Loss: 0.0754
+Epoch: 076/100 | Batch 0003/0043 | Averaged Loss: 0.0538
+Epoch: 076/100 | Batch 0004/0043 | Averaged Loss: 0.0696
+Epoch: 076/100 | Batch 0005/0043 | Averaged Loss: 0.0681
+Epoch: 076/100 | Batch 0006/0043 | Averaged Loss: 0.0693
+Epoch: 076/100 | Batch 0007/0043 | Averaged Loss: 0.0566
+Epoch: 076/100 | Batch 0008/0043 | Averaged Loss: 0.0665
+Epoch: 076/100 | Batch 0009/0043 | Averaged Loss: 0.0637
+Epoch: 076/100 | Batch 0010/0043 | Averaged Loss: 0.0656
+Epoch: 076/100 | Batch 0011/0043 | Averaged Loss: 0.0753
+Epoch: 076/100 | Batch 0012/0043 | Averaged Loss: 0.0451
+Epoch: 076/100 | Batch 0013/0043 | Averaged Loss: 0.0727
+Epoch: 076/100 | Batch 0014/0043 | Averaged Loss: 0.0612
+Epoch: 076/100 | Batch 0015/0043 | Averaged Loss: 0.0479
+Epoch: 076/100 | Batch 0016/0043 | Averaged Loss: 0.0719
+Epoch: 076/100 | Batch 0017/0043 | Averaged Loss: 0.0562
+Epoch: 076/100 | Batch 0018/0043 | Averaged Loss: 0.0798
+Epoch: 076/100 | Batch 0019/0043 | Averaged Loss: 0.0536
+Epoch: 076/100 | Batch 0020/0043 | Averaged Loss: 0.0830
+Epoch: 076/100 | Batch 0021/0043 | Averaged Loss: 0.0618
+Epoch: 076/100 | Batch 0022/0043 | Averaged Loss: 0.0702
+Epoch: 076/100 | Batch 0023/0043 | Averaged Loss: 0.0705
+Epoch: 076/100 | Batch 0024/0043 | Averaged Loss: 0.0435
+Epoch: 076/100 | Batch 0025/0043 | Averaged Loss: 0.0885
+Epoch: 076/100 | Batch 0026/0043 | Averaged Loss: 0.0546
+Epoch: 076/100 | Batch 0027/0043 | Averaged Loss: 0.0432
+Epoch: 076/100 | Batch 0028/0043 | Averaged Loss: 0.0551
+Epoch: 076/100 | Batch 0029/0043 | Averaged Loss: 0.0892
+Epoch: 076/100 | Batch 0030/0043 | Averaged Loss: 0.0821
+Epoch: 076/100 | Batch 0031/0043 | Averaged Loss: 0.0571
+Epoch: 076/100 | Batch 0032/0043 | Averaged Loss: 0.0958
+Epoch: 076/100 | Batch 0033/0043 | Averaged Loss: 0.0673
+Epoch: 076/100 | Batch 0034/0043 | Averaged Loss: 0.0658
+Epoch: 076/100 | Batch 0035/0043 | Averaged Loss: 0.0750
+Epoch: 076/100 | Batch 0036/0043 | Averaged Loss: 0.0741
+Epoch: 076/100 | Batch 0037/0043 | Averaged Loss: 0.0701
+Epoch: 076/100 | Batch 0038/0043 | Averaged Loss: 0.1002
+Epoch: 076/100 | Batch 0039/0043 | Averaged Loss: 0.0516
+Epoch: 076/100 | Batch 0040/0043 | Averaged Loss: 0.0620
+Epoch: 076/100 | Batch 0041/0043 | Averaged Loss: 0.0588
+Epoch: 076/100 | Batch 0042/0043 | Averaged Loss: 0.1047
+Epoch: 076/100 | Train: 97.84% | Validation: 75.12%
+Time elapsed: 11.284931182861328 min
+Epoch: 077/100 | Batch 0000/0043 | Averaged Loss: 0.0503
+Epoch: 077/100 | Batch 0001/0043 | Averaged Loss: 0.0600
+Epoch: 077/100 | Batch 0002/0043 | Averaged Loss: 0.0475
+Epoch: 077/100 | Batch 0003/0043 | Averaged Loss: 0.0534
+Epoch: 077/100 | Batch 0004/0043 | Averaged Loss: 0.0901
+Epoch: 077/100 | Batch 0005/0043 | Averaged Loss: 0.0798
+Epoch: 077/100 | Batch 0006/0043 | Averaged Loss: 0.0669
+Epoch: 077/100 | Batch 0007/0043 | Averaged Loss: 0.0463
+Epoch: 077/100 | Batch 0008/0043 | Averaged Loss: 0.0522
+Epoch: 077/100 | Batch 0009/0043 | Averaged Loss: 0.0509
+Epoch: 077/100 | Batch 0010/0043 | Averaged Loss: 0.0799
+Epoch: 077/100 | Batch 0011/0043 | Averaged Loss: 0.0441
+Epoch: 077/100 | Batch 0012/0043 | Averaged Loss: 0.0594
+Epoch: 077/100 | Batch 0013/0043 | Averaged Loss: 0.0505
+Epoch: 077/100 | Batch 0014/0043 | Averaged Loss: 0.0436
+Epoch: 077/100 | Batch 0015/0043 | Averaged Loss: 0.0764
+Epoch: 077/100 | Batch 0016/0043 | Averaged Loss: 0.0528
+Epoch: 077/100 | Batch 0017/0043 | Averaged Loss: 0.0487
+Epoch: 077/100 | Batch 0018/0043 | Averaged Loss: 0.0407
+Epoch: 077/100 | Batch 0019/0043 | Averaged Loss: 0.0898
+Epoch: 077/100 | Batch 0020/0043 | Averaged Loss: 0.0581
+Epoch: 077/100 | Batch 0021/0043 | Averaged Loss: 0.0554
+Epoch: 077/100 | Batch 0022/0043 | Averaged Loss: 0.0480
+Epoch: 077/100 | Batch 0023/0043 | Averaged Loss: 0.0518
+Epoch: 077/100 | Batch 0024/0043 | Averaged Loss: 0.1083
+Epoch: 077/100 | Batch 0025/0043 | Averaged Loss: 0.0687
+Epoch: 077/100 | Batch 0026/0043 | Averaged Loss: 0.0530
+Epoch: 077/100 | Batch 0027/0043 | Averaged Loss: 0.0692
+Epoch: 077/100 | Batch 0028/0043 | Averaged Loss: 0.0676
+Epoch: 077/100 | Batch 0029/0043 | Averaged Loss: 0.0615
+Epoch: 077/100 | Batch 0030/0043 | Averaged Loss: 0.0871
+Epoch: 077/100 | Batch 0031/0043 | Averaged Loss: 0.0799
+Epoch: 077/100 | Batch 0032/0043 | Averaged Loss: 0.0712
+Epoch: 077/100 | Batch 0033/0043 | Averaged Loss: 0.0804
+Epoch: 077/100 | Batch 0034/0043 | Averaged Loss: 0.0764
+Epoch: 077/100 | Batch 0035/0043 | Averaged Loss: 0.0490
+Epoch: 077/100 | Batch 0036/0043 | Averaged Loss: 0.0649
+Epoch: 077/100 | Batch 0037/0043 | Averaged Loss: 0.0901
+Epoch: 077/100 | Batch 0038/0043 | Averaged Loss: 0.0682
+Epoch: 077/100 | Batch 0039/0043 | Averaged Loss: 0.0796
+Epoch: 077/100 | Batch 0040/0043 | Averaged Loss: 0.0931
+Epoch: 077/100 | Batch 0041/0043 | Averaged Loss: 0.1025
+Epoch: 077/100 | Batch 0042/0043 | Averaged Loss: 0.0680
+Epoch: 077/100 | Train: 98.25% | Validation: 75.63%
+Time elapsed: 11.433085441589355 min
+Epoch: 078/100 | Batch 0000/0043 | Averaged Loss: 0.0480
+Epoch: 078/100 | Batch 0001/0043 | Averaged Loss: 0.0716
+Epoch: 078/100 | Batch 0002/0043 | Averaged Loss: 0.0867
+Epoch: 078/100 | Batch 0003/0043 | Averaged Loss: 0.0511
+Epoch: 078/100 | Batch 0004/0043 | Averaged Loss: 0.0629
+Epoch: 078/100 | Batch 0005/0043 | Averaged Loss: 0.0755
+Epoch: 078/100 | Batch 0006/0043 | Averaged Loss: 0.0614
+Epoch: 078/100 | Batch 0007/0043 | Averaged Loss: 0.0828
+Epoch: 078/100 | Batch 0008/0043 | Averaged Loss: 0.0614
+Epoch: 078/100 | Batch 0009/0043 | Averaged Loss: 0.0633
+Epoch: 078/100 | Batch 0010/0043 | Averaged Loss: 0.0602
+Epoch: 078/100 | Batch 0011/0043 | Averaged Loss: 0.0615
+Epoch: 078/100 | Batch 0012/0043 | Averaged Loss: 0.0793
+Epoch: 078/100 | Batch 0013/0043 | Averaged Loss: 0.0684
+Epoch: 078/100 | Batch 0014/0043 | Averaged Loss: 0.0495
+Epoch: 078/100 | Batch 0015/0043 | Averaged Loss: 0.0625
+Epoch: 078/100 | Batch 0016/0043 | Averaged Loss: 0.0716
+Epoch: 078/100 | Batch 0017/0043 | Averaged Loss: 0.0541
+Epoch: 078/100 | Batch 0018/0043 | Averaged Loss: 0.0962
+Epoch: 078/100 | Batch 0019/0043 | Averaged Loss: 0.0683
+Epoch: 078/100 | Batch 0020/0043 | Averaged Loss: 0.0529
+Epoch: 078/100 | Batch 0021/0043 | Averaged Loss: 0.0945
+Epoch: 078/100 | Batch 0022/0043 | Averaged Loss: 0.0494
+Epoch: 078/100 | Batch 0023/0043 | Averaged Loss: 0.0830
+Epoch: 078/100 | Batch 0024/0043 | Averaged Loss: 0.0783
+Epoch: 078/100 | Batch 0025/0043 | Averaged Loss: 0.0571
+Epoch: 078/100 | Batch 0026/0043 | Averaged Loss: 0.0733
+Epoch: 078/100 | Batch 0027/0043 | Averaged Loss: 0.0738
+Epoch: 078/100 | Batch 0028/0043 | Averaged Loss: 0.0532
+Epoch: 078/100 | Batch 0029/0043 | Averaged Loss: 0.0759
+Epoch: 078/100 | Batch 0030/0043 | Averaged Loss: 0.0547
+Epoch: 078/100 | Batch 0031/0043 | Averaged Loss: 0.0306
+Epoch: 078/100 | Batch 0032/0043 | Averaged Loss: 0.0611
+Epoch: 078/100 | Batch 0033/0043 | Averaged Loss: 0.0845
+Epoch: 078/100 | Batch 0034/0043 | Averaged Loss: 0.0860
+Epoch: 078/100 | Batch 0035/0043 | Averaged Loss: 0.0637
+Epoch: 078/100 | Batch 0036/0043 | Averaged Loss: 0.0609
+Epoch: 078/100 | Batch 0037/0043 | Averaged Loss: 0.0747
+Epoch: 078/100 | Batch 0038/0043 | Averaged Loss: 0.0622
+Epoch: 078/100 | Batch 0039/0043 | Averaged Loss: 0.0985
+Epoch: 078/100 | Batch 0040/0043 | Averaged Loss: 0.0513
+Epoch: 078/100 | Batch 0041/0043 | Averaged Loss: 0.0343
+Epoch: 078/100 | Batch 0042/0043 | Averaged Loss: 0.0491
+Epoch: 078/100 | Train: 97.77% | Validation: 74.61%
+Time elapsed: 11.581162452697754 min
+Epoch: 079/100 | Batch 0000/0043 | Averaged Loss: 0.0767
+Epoch: 079/100 | Batch 0001/0043 | Averaged Loss: 0.0459
+Epoch: 079/100 | Batch 0002/0043 | Averaged Loss: 0.0774
+Epoch: 079/100 | Batch 0003/0043 | Averaged Loss: 0.0397
+Epoch: 079/100 | Batch 0004/0043 | Averaged Loss: 0.0535
+Epoch: 079/100 | Batch 0005/0043 | Averaged Loss: 0.0885
+Epoch: 079/100 | Batch 0006/0043 | Averaged Loss: 0.0459
+Epoch: 079/100 | Batch 0007/0043 | Averaged Loss: 0.0577
+Epoch: 079/100 | Batch 0008/0043 | Averaged Loss: 0.0485
+Epoch: 079/100 | Batch 0009/0043 | Averaged Loss: 0.0502
+Epoch: 079/100 | Batch 0010/0043 | Averaged Loss: 0.0673
+Epoch: 079/100 | Batch 0011/0043 | Averaged Loss: 0.0625
+Epoch: 079/100 | Batch 0012/0043 | Averaged Loss: 0.0606
+Epoch: 079/100 | Batch 0013/0043 | Averaged Loss: 0.0480
+Epoch: 079/100 | Batch 0014/0043 | Averaged Loss: 0.0527
+Epoch: 079/100 | Batch 0015/0043 | Averaged Loss: 0.0713
+Epoch: 079/100 | Batch 0016/0043 | Averaged Loss: 0.0453
+Epoch: 079/100 | Batch 0017/0043 | Averaged Loss: 0.0377
+Epoch: 079/100 | Batch 0018/0043 | Averaged Loss: 0.0641
+Epoch: 079/100 | Batch 0019/0043 | Averaged Loss: 0.0529
+Epoch: 079/100 | Batch 0020/0043 | Averaged Loss: 0.0477
+Epoch: 079/100 | Batch 0021/0043 | Averaged Loss: 0.0493
+Epoch: 079/100 | Batch 0022/0043 | Averaged Loss: 0.0714
+Epoch: 079/100 | Batch 0023/0043 | Averaged Loss: 0.0478
+Epoch: 079/100 | Batch 0024/0043 | Averaged Loss: 0.0490
+Epoch: 079/100 | Batch 0025/0043 | Averaged Loss: 0.0624
+Epoch: 079/100 | Batch 0026/0043 | Averaged Loss: 0.0393
+Epoch: 079/100 | Batch 0027/0043 | Averaged Loss: 0.0415
+Epoch: 079/100 | Batch 0028/0043 | Averaged Loss: 0.0535
+Epoch: 079/100 | Batch 0029/0043 | Averaged Loss: 0.0619
+Epoch: 079/100 | Batch 0030/0043 | Averaged Loss: 0.0628
+Epoch: 079/100 | Batch 0031/0043 | Averaged Loss: 0.0523
+Epoch: 079/100 | Batch 0032/0043 | Averaged Loss: 0.0722
+Epoch: 079/100 | Batch 0033/0043 | Averaged Loss: 0.0412
+Epoch: 079/100 | Batch 0034/0043 | Averaged Loss: 0.0410
+Epoch: 079/100 | Batch 0035/0043 | Averaged Loss: 0.0580
+Epoch: 079/100 | Batch 0036/0043 | Averaged Loss: 0.0723
+Epoch: 079/100 | Batch 0037/0043 | Averaged Loss: 0.0721
+Epoch: 079/100 | Batch 0038/0043 | Averaged Loss: 0.0572
+Epoch: 079/100 | Batch 0039/0043 | Averaged Loss: 0.0644
+Epoch: 079/100 | Batch 0040/0043 | Averaged Loss: 0.0987
+Epoch: 079/100 | Batch 0041/0043 | Averaged Loss: 0.0533
+Epoch: 079/100 | Batch 0042/0043 | Averaged Loss: 0.0807
+Epoch: 079/100 | Train: 98.57% | Validation: 75.83%
+Time elapsed: 11.729338645935059 min
+Epoch: 080/100 | Batch 0000/0043 | Averaged Loss: 0.0529
+Epoch: 080/100 | Batch 0001/0043 | Averaged Loss: 0.0540
+Epoch: 080/100 | Batch 0002/0043 | Averaged Loss: 0.0324
+Epoch: 080/100 | Batch 0003/0043 | Averaged Loss: 0.0749
+Epoch: 080/100 | Batch 0004/0043 | Averaged Loss: 0.0690
+Epoch: 080/100 | Batch 0005/0043 | Averaged Loss: 0.0512
+Epoch: 080/100 | Batch 0006/0043 | Averaged Loss: 0.0536
+Epoch: 080/100 | Batch 0007/0043 | Averaged Loss: 0.0714
+Epoch: 080/100 | Batch 0008/0043 | Averaged Loss: 0.0665
+Epoch: 080/100 | Batch 0009/0043 | Averaged Loss: 0.0646
+Epoch: 080/100 | Batch 0010/0043 | Averaged Loss: 0.0795
+Epoch: 080/100 | Batch 0011/0043 | Averaged Loss: 0.0437
+Epoch: 080/100 | Batch 0012/0043 | Averaged Loss: 0.0436
+Epoch: 080/100 | Batch 0013/0043 | Averaged Loss: 0.0528
+Epoch: 080/100 | Batch 0014/0043 | Averaged Loss: 0.0543
+Epoch: 080/100 | Batch 0015/0043 | Averaged Loss: 0.0469
+Epoch: 080/100 | Batch 0016/0043 | Averaged Loss: 0.0658
+Epoch: 080/100 | Batch 0017/0043 | Averaged Loss: 0.0731
+Epoch: 080/100 | Batch 0018/0043 | Averaged Loss: 0.0708
+Epoch: 080/100 | Batch 0019/0043 | Averaged Loss: 0.0654
+Epoch: 080/100 | Batch 0020/0043 | Averaged Loss: 0.0477
+Epoch: 080/100 | Batch 0021/0043 | Averaged Loss: 0.0806
+Epoch: 080/100 | Batch 0022/0043 | Averaged Loss: 0.0421
+Epoch: 080/100 | Batch 0023/0043 | Averaged Loss: 0.0487
+Epoch: 080/100 | Batch 0024/0043 | Averaged Loss: 0.0525
+Epoch: 080/100 | Batch 0025/0043 | Averaged Loss: 0.0481
+Epoch: 080/100 | Batch 0026/0043 | Averaged Loss: 0.0630
+Epoch: 080/100 | Batch 0027/0043 | Averaged Loss: 0.0508
+Epoch: 080/100 | Batch 0028/0043 | Averaged Loss: 0.0562
+Epoch: 080/100 | Batch 0029/0043 | Averaged Loss: 0.0435
+Epoch: 080/100 | Batch 0030/0043 | Averaged Loss: 0.0462
+Epoch: 080/100 | Batch 0031/0043 | Averaged Loss: 0.0742
+Epoch: 080/100 | Batch 0032/0043 | Averaged Loss: 0.0443
+Epoch: 080/100 | Batch 0033/0043 | Averaged Loss: 0.0543
+Epoch: 080/100 | Batch 0034/0043 | Averaged Loss: 0.0521
+Epoch: 080/100 | Batch 0035/0043 | Averaged Loss: 0.0505
+Epoch: 080/100 | Batch 0036/0043 | Averaged Loss: 0.0827
+Epoch: 080/100 | Batch 0037/0043 | Averaged Loss: 0.0556
+Epoch: 080/100 | Batch 0038/0043 | Averaged Loss: 0.0464
+Epoch: 080/100 | Batch 0039/0043 | Averaged Loss: 0.0421
+Epoch: 080/100 | Batch 0040/0043 | Averaged Loss: 0.0498
+Epoch: 080/100 | Batch 0041/0043 | Averaged Loss: 0.0443
+Epoch: 080/100 | Batch 0042/0043 | Averaged Loss: 0.0571
+Epoch: 080/100 | Train: 98.29% | Validation: 75.24%
+Time elapsed: 11.877532005310059 min
+Epoch: 081/100 | Batch 0000/0043 | Averaged Loss: 0.0713
+Epoch: 081/100 | Batch 0001/0043 | Averaged Loss: 0.0516
+Epoch: 081/100 | Batch 0002/0043 | Averaged Loss: 0.0398
+Epoch: 081/100 | Batch 0003/0043 | Averaged Loss: 0.0641
+Epoch: 081/100 | Batch 0004/0043 | Averaged Loss: 0.0550
+Epoch: 081/100 | Batch 0005/0043 | Averaged Loss: 0.0588
+Epoch: 081/100 | Batch 0006/0043 | Averaged Loss: 0.0560
+Epoch: 081/100 | Batch 0007/0043 | Averaged Loss: 0.0368
+Epoch: 081/100 | Batch 0008/0043 | Averaged Loss: 0.0559
+Epoch: 081/100 | Batch 0009/0043 | Averaged Loss: 0.1028
+Epoch: 081/100 | Batch 0010/0043 | Averaged Loss: 0.0288
+Epoch: 081/100 | Batch 0011/0043 | Averaged Loss: 0.0495
+Epoch: 081/100 | Batch 0012/0043 | Averaged Loss: 0.0417
+Epoch: 081/100 | Batch 0013/0043 | Averaged Loss: 0.0932
+Epoch: 081/100 | Batch 0014/0043 | Averaged Loss: 0.0700
+Epoch: 081/100 | Batch 0015/0043 | Averaged Loss: 0.0571
+Epoch: 081/100 | Batch 0016/0043 | Averaged Loss: 0.0753
+Epoch: 081/100 | Batch 0017/0043 | Averaged Loss: 0.0852
+Epoch: 081/100 | Batch 0018/0043 | Averaged Loss: 0.0581
+Epoch: 081/100 | Batch 0019/0043 | Averaged Loss: 0.0489
+Epoch: 081/100 | Batch 0020/0043 | Averaged Loss: 0.0600
+Epoch: 081/100 | Batch 0021/0043 | Averaged Loss: 0.0913
+Epoch: 081/100 | Batch 0022/0043 | Averaged Loss: 0.0642
+Epoch: 081/100 | Batch 0023/0043 | Averaged Loss: 0.0488
+Epoch: 081/100 | Batch 0024/0043 | Averaged Loss: 0.0643
+Epoch: 081/100 | Batch 0025/0043 | Averaged Loss: 0.0708
+Epoch: 081/100 | Batch 0026/0043 | Averaged Loss: 0.0721
+Epoch: 081/100 | Batch 0027/0043 | Averaged Loss: 0.0686
+Epoch: 081/100 | Batch 0028/0043 | Averaged Loss: 0.0788
+Epoch: 081/100 | Batch 0029/0043 | Averaged Loss: 0.0400
+Epoch: 081/100 | Batch 0030/0043 | Averaged Loss: 0.0545
+Epoch: 081/100 | Batch 0031/0043 | Averaged Loss: 0.0727
+Epoch: 081/100 | Batch 0032/0043 | Averaged Loss: 0.0407
+Epoch: 081/100 | Batch 0033/0043 | Averaged Loss: 0.0732
+Epoch: 081/100 | Batch 0034/0043 | Averaged Loss: 0.0600
+Epoch: 081/100 | Batch 0035/0043 | Averaged Loss: 0.0589
+Epoch: 081/100 | Batch 0036/0043 | Averaged Loss: 0.0416
+Epoch: 081/100 | Batch 0037/0043 | Averaged Loss: 0.0298
+Epoch: 081/100 | Batch 0038/0043 | Averaged Loss: 0.0419
+Epoch: 081/100 | Batch 0039/0043 | Averaged Loss: 0.0549
+Epoch: 081/100 | Batch 0040/0043 | Averaged Loss: 0.0558
+Epoch: 081/100 | Batch 0041/0043 | Averaged Loss: 0.0773
+Epoch: 081/100 | Batch 0042/0043 | Averaged Loss: 0.0553
+Epoch: 081/100 | Train: 98.58% | Validation: 74.85%
+Time elapsed: 12.026061058044434 min
+Epoch: 082/100 | Batch 0000/0043 | Averaged Loss: 0.0400
+Epoch: 082/100 | Batch 0001/0043 | Averaged Loss: 0.0554
+Epoch: 082/100 | Batch 0002/0043 | Averaged Loss: 0.0400
+Epoch: 082/100 | Batch 0003/0043 | Averaged Loss: 0.0580
+Epoch: 082/100 | Batch 0004/0043 | Averaged Loss: 0.0422
+Epoch: 082/100 | Batch 0005/0043 | Averaged Loss: 0.0794
+Epoch: 082/100 | Batch 0006/0043 | Averaged Loss: 0.0563
+Epoch: 082/100 | Batch 0007/0043 | Averaged Loss: 0.0389
+Epoch: 082/100 | Batch 0008/0043 | Averaged Loss: 0.0495
+Epoch: 082/100 | Batch 0009/0043 | Averaged Loss: 0.0424
+Epoch: 082/100 | Batch 0010/0043 | Averaged Loss: 0.0697
+Epoch: 082/100 | Batch 0011/0043 | Averaged Loss: 0.0515
+Epoch: 082/100 | Batch 0012/0043 | Averaged Loss: 0.0818
+Epoch: 082/100 | Batch 0013/0043 | Averaged Loss: 0.0749
+Epoch: 082/100 | Batch 0014/0043 | Averaged Loss: 0.0585
+Epoch: 082/100 | Batch 0015/0043 | Averaged Loss: 0.0400
+Epoch: 082/100 | Batch 0016/0043 | Averaged Loss: 0.0572
+Epoch: 082/100 | Batch 0017/0043 | Averaged Loss: 0.0602
+Epoch: 082/100 | Batch 0018/0043 | Averaged Loss: 0.0581
+Epoch: 082/100 | Batch 0019/0043 | Averaged Loss: 0.0646
+Epoch: 082/100 | Batch 0020/0043 | Averaged Loss: 0.0513
+Epoch: 082/100 | Batch 0021/0043 | Averaged Loss: 0.0600
+Epoch: 082/100 | Batch 0022/0043 | Averaged Loss: 0.0596
+Epoch: 082/100 | Batch 0023/0043 | Averaged Loss: 0.0595
+Epoch: 082/100 | Batch 0024/0043 | Averaged Loss: 0.0622
+Epoch: 082/100 | Batch 0025/0043 | Averaged Loss: 0.0967
+Epoch: 082/100 | Batch 0026/0043 | Averaged Loss: 0.0542
+Epoch: 082/100 | Batch 0027/0043 | Averaged Loss: 0.0390
+Epoch: 082/100 | Batch 0028/0043 | Averaged Loss: 0.0468
+Epoch: 082/100 | Batch 0029/0043 | Averaged Loss: 0.0767
+Epoch: 082/100 | Batch 0030/0043 | Averaged Loss: 0.0662
+Epoch: 082/100 | Batch 0031/0043 | Averaged Loss: 0.0547
+Epoch: 082/100 | Batch 0032/0043 | Averaged Loss: 0.0545
+Epoch: 082/100 | Batch 0033/0043 | Averaged Loss: 0.0526
+Epoch: 082/100 | Batch 0034/0043 | Averaged Loss: 0.0529
+Epoch: 082/100 | Batch 0035/0043 | Averaged Loss: 0.0462
+Epoch: 082/100 | Batch 0036/0043 | Averaged Loss: 0.0773
+Epoch: 082/100 | Batch 0037/0043 | Averaged Loss: 0.0842
+Epoch: 082/100 | Batch 0038/0043 | Averaged Loss: 0.0615
+Epoch: 082/100 | Batch 0039/0043 | Averaged Loss: 0.0353
+Epoch: 082/100 | Batch 0040/0043 | Averaged Loss: 0.0360
+Epoch: 082/100 | Batch 0041/0043 | Averaged Loss: 0.0517
+Epoch: 082/100 | Batch 0042/0043 | Averaged Loss: 0.0591
+Epoch: 082/100 | Train: 98.39% | Validation: 75.15%
+Time elapsed: 12.174510955810547 min
+Epoch: 083/100 | Batch 0000/0043 | Averaged Loss: 0.0659
+Epoch: 083/100 | Batch 0001/0043 | Averaged Loss: 0.0397
+Epoch: 083/100 | Batch 0002/0043 | Averaged Loss: 0.0443
+Epoch: 083/100 | Batch 0003/0043 | Averaged Loss: 0.0535
+Epoch: 083/100 | Batch 0004/0043 | Averaged Loss: 0.0455
+Epoch: 083/100 | Batch 0005/0043 | Averaged Loss: 0.0438
+Epoch: 083/100 | Batch 0006/0043 | Averaged Loss: 0.0568
+Epoch: 083/100 | Batch 0007/0043 | Averaged Loss: 0.0804
+Epoch: 083/100 | Batch 0008/0043 | Averaged Loss: 0.0412
+Epoch: 083/100 | Batch 0009/0043 | Averaged Loss: 0.0650
+Epoch: 083/100 | Batch 0010/0043 | Averaged Loss: 0.0482
+Epoch: 083/100 | Batch 0011/0043 | Averaged Loss: 0.0653
+Epoch: 083/100 | Batch 0012/0043 | Averaged Loss: 0.0508
+Epoch: 083/100 | Batch 0013/0043 | Averaged Loss: 0.0608
+Epoch: 083/100 | Batch 0014/0043 | Averaged Loss: 0.0341
+Epoch: 083/100 | Batch 0015/0043 | Averaged Loss: 0.0321
+Epoch: 083/100 | Batch 0016/0043 | Averaged Loss: 0.0600
+Epoch: 083/100 | Batch 0017/0043 | Averaged Loss: 0.0804
+Epoch: 083/100 | Batch 0018/0043 | Averaged Loss: 0.0554
+Epoch: 083/100 | Batch 0019/0043 | Averaged Loss: 0.0356
+Epoch: 083/100 | Batch 0020/0043 | Averaged Loss: 0.0521
+Epoch: 083/100 | Batch 0021/0043 | Averaged Loss: 0.0589
+Epoch: 083/100 | Batch 0022/0043 | Averaged Loss: 0.0336
+Epoch: 083/100 | Batch 0023/0043 | Averaged Loss: 0.0624
+Epoch: 083/100 | Batch 0024/0043 | Averaged Loss: 0.0336
+Epoch: 083/100 | Batch 0025/0043 | Averaged Loss: 0.0785
+Epoch: 083/100 | Batch 0026/0043 | Averaged Loss: 0.0821
+Epoch: 083/100 | Batch 0027/0043 | Averaged Loss: 0.0624
+Epoch: 083/100 | Batch 0028/0043 | Averaged Loss: 0.0764
+Epoch: 083/100 | Batch 0029/0043 | Averaged Loss: 0.0528
+Epoch: 083/100 | Batch 0030/0043 | Averaged Loss: 0.0616
+Epoch: 083/100 | Batch 0031/0043 | Averaged Loss: 0.0765
+Epoch: 083/100 | Batch 0032/0043 | Averaged Loss: 0.0786
+Epoch: 083/100 | Batch 0033/0043 | Averaged Loss: 0.0598
+Epoch: 083/100 | Batch 0034/0043 | Averaged Loss: 0.0601
+Epoch: 083/100 | Batch 0035/0043 | Averaged Loss: 0.0618
+Epoch: 083/100 | Batch 0036/0043 | Averaged Loss: 0.0887
+Epoch: 083/100 | Batch 0037/0043 | Averaged Loss: 0.0592
+Epoch: 083/100 | Batch 0038/0043 | Averaged Loss: 0.0725
+Epoch: 083/100 | Batch 0039/0043 | Averaged Loss: 0.1142
+Epoch: 083/100 | Batch 0040/0043 | Averaged Loss: 0.0798
+Epoch: 083/100 | Batch 0041/0043 | Averaged Loss: 0.0774
+Epoch: 083/100 | Batch 0042/0043 | Averaged Loss: 0.1120
+Epoch: 083/100 | Train: 98.05% | Validation: 74.54%
+Time elapsed: 12.323001861572266 min
+Epoch: 084/100 | Batch 0000/0043 | Averaged Loss: 0.0518
+Epoch: 084/100 | Batch 0001/0043 | Averaged Loss: 0.0584
+Epoch: 084/100 | Batch 0002/0043 | Averaged Loss: 0.0395
+Epoch: 084/100 | Batch 0003/0043 | Averaged Loss: 0.0726
+Epoch: 084/100 | Batch 0004/0043 | Averaged Loss: 0.0509
+Epoch: 084/100 | Batch 0005/0043 | Averaged Loss: 0.0585
+Epoch: 084/100 | Batch 0006/0043 | Averaged Loss: 0.0470
+Epoch: 084/100 | Batch 0007/0043 | Averaged Loss: 0.0429
+Epoch: 084/100 | Batch 0008/0043 | Averaged Loss: 0.0357
+Epoch: 084/100 | Batch 0009/0043 | Averaged Loss: 0.0480
+Epoch: 084/100 | Batch 0010/0043 | Averaged Loss: 0.0400
+Epoch: 084/100 | Batch 0011/0043 | Averaged Loss: 0.0498
+Epoch: 084/100 | Batch 0012/0043 | Averaged Loss: 0.0444
+Epoch: 084/100 | Batch 0013/0043 | Averaged Loss: 0.0337
+Epoch: 084/100 | Batch 0014/0043 | Averaged Loss: 0.0367
+Epoch: 084/100 | Batch 0015/0043 | Averaged Loss: 0.0417
+Epoch: 084/100 | Batch 0016/0043 | Averaged Loss: 0.0335
+Epoch: 084/100 | Batch 0017/0043 | Averaged Loss: 0.0327
+Epoch: 084/100 | Batch 0018/0043 | Averaged Loss: 0.0282
+Epoch: 084/100 | Batch 0019/0043 | Averaged Loss: 0.0413
+Epoch: 084/100 | Batch 0020/0043 | Averaged Loss: 0.0311
+Epoch: 084/100 | Batch 0021/0043 | Averaged Loss: 0.0487
+Epoch: 084/100 | Batch 0022/0043 | Averaged Loss: 0.0354
+Epoch: 084/100 | Batch 0023/0043 | Averaged Loss: 0.0480
+Epoch: 084/100 | Batch 0024/0043 | Averaged Loss: 0.0319
+Epoch: 084/100 | Batch 0025/0043 | Averaged Loss: 0.0500
+Epoch: 084/100 | Batch 0026/0043 | Averaged Loss: 0.0221
+Epoch: 084/100 | Batch 0027/0043 | Averaged Loss: 0.0490
+Epoch: 084/100 | Batch 0028/0043 | Averaged Loss: 0.0591
+Epoch: 084/100 | Batch 0029/0043 | Averaged Loss: 0.0222
+Epoch: 084/100 | Batch 0030/0043 | Averaged Loss: 0.0509
+Epoch: 084/100 | Batch 0031/0043 | Averaged Loss: 0.1235
+Epoch: 084/100 | Batch 0032/0043 | Averaged Loss: 0.0539
+Epoch: 084/100 | Batch 0033/0043 | Averaged Loss: 0.0280
+Epoch: 084/100 | Batch 0034/0043 | Averaged Loss: 0.0498
+Epoch: 084/100 | Batch 0035/0043 | Averaged Loss: 0.0814
+Epoch: 084/100 | Batch 0036/0043 | Averaged Loss: 0.0970
+Epoch: 084/100 | Batch 0037/0043 | Averaged Loss: 0.1253
+Epoch: 084/100 | Batch 0038/0043 | Averaged Loss: 0.0305
+Epoch: 084/100 | Batch 0039/0043 | Averaged Loss: 0.0411
+Epoch: 084/100 | Batch 0040/0043 | Averaged Loss: 0.0580
+Epoch: 084/100 | Batch 0041/0043 | Averaged Loss: 0.0601
+Epoch: 084/100 | Batch 0042/0043 | Averaged Loss: 0.0682
+Epoch: 084/100 | Train: 98.53% | Validation: 74.95%
+Time elapsed: 12.471353530883789 min
+Epoch: 085/100 | Batch 0000/0043 | Averaged Loss: 0.0506
+Epoch: 085/100 | Batch 0001/0043 | Averaged Loss: 0.0336
+Epoch: 085/100 | Batch 0002/0043 | Averaged Loss: 0.0438
+Epoch: 085/100 | Batch 0003/0043 | Averaged Loss: 0.0460
+Epoch: 085/100 | Batch 0004/0043 | Averaged Loss: 0.0517
+Epoch: 085/100 | Batch 0005/0043 | Averaged Loss: 0.0517
+Epoch: 085/100 | Batch 0006/0043 | Averaged Loss: 0.0495
+Epoch: 085/100 | Batch 0007/0043 | Averaged Loss: 0.0416
+Epoch: 085/100 | Batch 0008/0043 | Averaged Loss: 0.0446
+Epoch: 085/100 | Batch 0009/0043 | Averaged Loss: 0.0647
+Epoch: 085/100 | Batch 0010/0043 | Averaged Loss: 0.0459
+Epoch: 085/100 | Batch 0011/0043 | Averaged Loss: 0.0373
+Epoch: 085/100 | Batch 0012/0043 | Averaged Loss: 0.0662
+Epoch: 085/100 | Batch 0013/0043 | Averaged Loss: 0.0344
+Epoch: 085/100 | Batch 0014/0043 | Averaged Loss: 0.0303
+Epoch: 085/100 | Batch 0015/0043 | Averaged Loss: 0.0691
+Epoch: 085/100 | Batch 0016/0043 | Averaged Loss: 0.0528
+Epoch: 085/100 | Batch 0017/0043 | Averaged Loss: 0.0628
+Epoch: 085/100 | Batch 0018/0043 | Averaged Loss: 0.0406
+Epoch: 085/100 | Batch 0019/0043 | Averaged Loss: 0.0551
+Epoch: 085/100 | Batch 0020/0043 | Averaged Loss: 0.0525
+Epoch: 085/100 | Batch 0021/0043 | Averaged Loss: 0.0734
+Epoch: 085/100 | Batch 0022/0043 | Averaged Loss: 0.0457
+Epoch: 085/100 | Batch 0023/0043 | Averaged Loss: 0.0340
+Epoch: 085/100 | Batch 0024/0043 | Averaged Loss: 0.0344
+Epoch: 085/100 | Batch 0025/0043 | Averaged Loss: 0.0440
+Epoch: 085/100 | Batch 0026/0043 | Averaged Loss: 0.0384
+Epoch: 085/100 | Batch 0027/0043 | Averaged Loss: 0.0721
+Epoch: 085/100 | Batch 0028/0043 | Averaged Loss: 0.0512
+Epoch: 085/100 | Batch 0029/0043 | Averaged Loss: 0.0687
+Epoch: 085/100 | Batch 0030/0043 | Averaged Loss: 0.0393
+Epoch: 085/100 | Batch 0031/0043 | Averaged Loss: 0.0564
+Epoch: 085/100 | Batch 0032/0043 | Averaged Loss: 0.0433
+Epoch: 085/100 | Batch 0033/0043 | Averaged Loss: 0.0641
+Epoch: 085/100 | Batch 0034/0043 | Averaged Loss: 0.0534
+Epoch: 085/100 | Batch 0035/0043 | Averaged Loss: 0.0602
+Epoch: 085/100 | Batch 0036/0043 | Averaged Loss: 0.0994
+Epoch: 085/100 | Batch 0037/0043 | Averaged Loss: 0.0621
+Epoch: 085/100 | Batch 0038/0043 | Averaged Loss: 0.0704
+Epoch: 085/100 | Batch 0039/0043 | Averaged Loss: 0.0826
+Epoch: 085/100 | Batch 0040/0043 | Averaged Loss: 0.0436
+Epoch: 085/100 | Batch 0041/0043 | Averaged Loss: 0.0547
+Epoch: 085/100 | Batch 0042/0043 | Averaged Loss: 0.0526
+Epoch: 085/100 | Train: 98.18% | Validation: 74.71%
+Time elapsed: 12.619789123535156 min
+Epoch: 086/100 | Batch 0000/0043 | Averaged Loss: 0.0451
+Epoch: 086/100 | Batch 0001/0043 | Averaged Loss: 0.0564
+Epoch: 086/100 | Batch 0002/0043 | Averaged Loss: 0.0420
+Epoch: 086/100 | Batch 0003/0043 | Averaged Loss: 0.0340
+Epoch: 086/100 | Batch 0004/0043 | Averaged Loss: 0.0251
+Epoch: 086/100 | Batch 0005/0043 | Averaged Loss: 0.0293
+Epoch: 086/100 | Batch 0006/0043 | Averaged Loss: 0.0397
+Epoch: 086/100 | Batch 0007/0043 | Averaged Loss: 0.0426
+Epoch: 086/100 | Batch 0008/0043 | Averaged Loss: 0.0496
+Epoch: 086/100 | Batch 0009/0043 | Averaged Loss: 0.0454
+Epoch: 086/100 | Batch 0010/0043 | Averaged Loss: 0.0674
+Epoch: 086/100 | Batch 0011/0043 | Averaged Loss: 0.0331
+Epoch: 086/100 | Batch 0012/0043 | Averaged Loss: 0.0313
+Epoch: 086/100 | Batch 0013/0043 | Averaged Loss: 0.0468
+Epoch: 086/100 | Batch 0014/0043 | Averaged Loss: 0.0390
+Epoch: 086/100 | Batch 0015/0043 | Averaged Loss: 0.0649
+Epoch: 086/100 | Batch 0016/0043 | Averaged Loss: 0.0701
+Epoch: 086/100 | Batch 0017/0043 | Averaged Loss: 0.0353
+Epoch: 086/100 | Batch 0018/0043 | Averaged Loss: 0.0525
+Epoch: 086/100 | Batch 0019/0043 | Averaged Loss: 0.0542
+Epoch: 086/100 | Batch 0020/0043 | Averaged Loss: 0.0674
+Epoch: 086/100 | Batch 0021/0043 | Averaged Loss: 0.0742
+Epoch: 086/100 | Batch 0022/0043 | Averaged Loss: 0.0639
+Epoch: 086/100 | Batch 0023/0043 | Averaged Loss: 0.0687
+Epoch: 086/100 | Batch 0024/0043 | Averaged Loss: 0.0675
+Epoch: 086/100 | Batch 0025/0043 | Averaged Loss: 0.0332
+Epoch: 086/100 | Batch 0026/0043 | Averaged Loss: 0.0538
+Epoch: 086/100 | Batch 0027/0043 | Averaged Loss: 0.0423
+Epoch: 086/100 | Batch 0028/0043 | Averaged Loss: 0.0730
+Epoch: 086/100 | Batch 0029/0043 | Averaged Loss: 0.0511
+Epoch: 086/100 | Batch 0030/0043 | Averaged Loss: 0.0862
+Epoch: 086/100 | Batch 0031/0043 | Averaged Loss: 0.0453
+Epoch: 086/100 | Batch 0032/0043 | Averaged Loss: 0.0501
+Epoch: 086/100 | Batch 0033/0043 | Averaged Loss: 0.0596
+Epoch: 086/100 | Batch 0034/0043 | Averaged Loss: 0.0505
+Epoch: 086/100 | Batch 0035/0043 | Averaged Loss: 0.0398
+Epoch: 086/100 | Batch 0036/0043 | Averaged Loss: 0.0540
+Epoch: 086/100 | Batch 0037/0043 | Averaged Loss: 0.0595
+Epoch: 086/100 | Batch 0038/0043 | Averaged Loss: 0.0521
+Epoch: 086/100 | Batch 0039/0043 | Averaged Loss: 0.0712
+Epoch: 086/100 | Batch 0040/0043 | Averaged Loss: 0.0357
+Epoch: 086/100 | Batch 0041/0043 | Averaged Loss: 0.0526
+Epoch: 086/100 | Batch 0042/0043 | Averaged Loss: 0.0562
+Epoch: 086/100 | Train: 98.51% | Validation: 74.78%
+Time elapsed: 12.768245697021484 min
+Epoch: 087/100 | Batch 0000/0043 | Averaged Loss: 0.0616
+Epoch: 087/100 | Batch 0001/0043 | Averaged Loss: 0.0561
+Epoch: 087/100 | Batch 0002/0043 | Averaged Loss: 0.0523
+Epoch: 087/100 | Batch 0003/0043 | Averaged Loss: 0.0496
+Epoch: 087/100 | Batch 0004/0043 | Averaged Loss: 0.0378
+Epoch: 087/100 | Batch 0005/0043 | Averaged Loss: 0.0419
+Epoch: 087/100 | Batch 0006/0043 | Averaged Loss: 0.0288
+Epoch: 087/100 | Batch 0007/0043 | Averaged Loss: 0.0601
+Epoch: 087/100 | Batch 0008/0043 | Averaged Loss: 0.0299
+Epoch: 087/100 | Batch 0009/0043 | Averaged Loss: 0.0781
+Epoch: 087/100 | Batch 0010/0043 | Averaged Loss: 0.0422
+Epoch: 087/100 | Batch 0011/0043 | Averaged Loss: 0.0479
+Epoch: 087/100 | Batch 0012/0043 | Averaged Loss: 0.0676
+Epoch: 087/100 | Batch 0013/0043 | Averaged Loss: 0.0379
+Epoch: 087/100 | Batch 0014/0043 | Averaged Loss: 0.0687
+Epoch: 087/100 | Batch 0015/0043 | Averaged Loss: 0.0490
+Epoch: 087/100 | Batch 0016/0043 | Averaged Loss: 0.0374
+Epoch: 087/100 | Batch 0017/0043 | Averaged Loss: 0.0404
+Epoch: 087/100 | Batch 0018/0043 | Averaged Loss: 0.0441
+Epoch: 087/100 | Batch 0019/0043 | Averaged Loss: 0.0465
+Epoch: 087/100 | Batch 0020/0043 | Averaged Loss: 0.0414
+Epoch: 087/100 | Batch 0021/0043 | Averaged Loss: 0.0511
+Epoch: 087/100 | Batch 0022/0043 | Averaged Loss: 0.0335
+Epoch: 087/100 | Batch 0023/0043 | Averaged Loss: 0.0510
+Epoch: 087/100 | Batch 0024/0043 | Averaged Loss: 0.0550
+Epoch: 087/100 | Batch 0025/0043 | Averaged Loss: 0.0675
+Epoch: 087/100 | Batch 0026/0043 | Averaged Loss: 0.0516
+Epoch: 087/100 | Batch 0027/0043 | Averaged Loss: 0.0688
+Epoch: 087/100 | Batch 0028/0043 | Averaged Loss: 0.0468
+Epoch: 087/100 | Batch 0029/0043 | Averaged Loss: 0.0639
+Epoch: 087/100 | Batch 0030/0043 | Averaged Loss: 0.0535
+Epoch: 087/100 | Batch 0031/0043 | Averaged Loss: 0.0500
+Epoch: 087/100 | Batch 0032/0043 | Averaged Loss: 0.0671
+Epoch: 087/100 | Batch 0033/0043 | Averaged Loss: 0.0601
+Epoch: 087/100 | Batch 0034/0043 | Averaged Loss: 0.0626
+Epoch: 087/100 | Batch 0035/0043 | Averaged Loss: 0.0556
+Epoch: 087/100 | Batch 0036/0043 | Averaged Loss: 0.0337
+Epoch: 087/100 | Batch 0037/0043 | Averaged Loss: 0.0512
+Epoch: 087/100 | Batch 0038/0043 | Averaged Loss: 0.0566
+Epoch: 087/100 | Batch 0039/0043 | Averaged Loss: 0.0610
+Epoch: 087/100 | Batch 0040/0043 | Averaged Loss: 0.0676
+Epoch: 087/100 | Batch 0041/0043 | Averaged Loss: 0.0647
+Epoch: 087/100 | Batch 0042/0043 | Averaged Loss: 0.0472
+Epoch: 087/100 | Train: 98.64% | Validation: 75.59%
+Time elapsed: 12.916889190673828 min
+Epoch: 088/100 | Batch 0000/0043 | Averaged Loss: 0.0419
+Epoch: 088/100 | Batch 0001/0043 | Averaged Loss: 0.0309
+Epoch: 088/100 | Batch 0002/0043 | Averaged Loss: 0.0475
+Epoch: 088/100 | Batch 0003/0043 | Averaged Loss: 0.0625
+Epoch: 088/100 | Batch 0004/0043 | Averaged Loss: 0.0237
+Epoch: 088/100 | Batch 0005/0043 | Averaged Loss: 0.0476
+Epoch: 088/100 | Batch 0006/0043 | Averaged Loss: 0.0589
+Epoch: 088/100 | Batch 0007/0043 | Averaged Loss: 0.0490
+Epoch: 088/100 | Batch 0008/0043 | Averaged Loss: 0.0641
+Epoch: 088/100 | Batch 0009/0043 | Averaged Loss: 0.0388
+Epoch: 088/100 | Batch 0010/0043 | Averaged Loss: 0.0305
+Epoch: 088/100 | Batch 0011/0043 | Averaged Loss: 0.0427
+Epoch: 088/100 | Batch 0012/0043 | Averaged Loss: 0.0460
+Epoch: 088/100 | Batch 0013/0043 | Averaged Loss: 0.0465
+Epoch: 088/100 | Batch 0014/0043 | Averaged Loss: 0.0268
+Epoch: 088/100 | Batch 0015/0043 | Averaged Loss: 0.0437
+Epoch: 088/100 | Batch 0016/0043 | Averaged Loss: 0.0404
+Epoch: 088/100 | Batch 0017/0043 | Averaged Loss: 0.0601
+Epoch: 088/100 | Batch 0018/0043 | Averaged Loss: 0.0569
+Epoch: 088/100 | Batch 0019/0043 | Averaged Loss: 0.0354
+Epoch: 088/100 | Batch 0020/0043 | Averaged Loss: 0.0464
+Epoch: 088/100 | Batch 0021/0043 | Averaged Loss: 0.0420
+Epoch: 088/100 | Batch 0022/0043 | Averaged Loss: 0.0289
+Epoch: 088/100 | Batch 0023/0043 | Averaged Loss: 0.0494
+Epoch: 088/100 | Batch 0024/0043 | Averaged Loss: 0.0454
+Epoch: 088/100 | Batch 0025/0043 | Averaged Loss: 0.0442
+Epoch: 088/100 | Batch 0026/0043 | Averaged Loss: 0.0566
+Epoch: 088/100 | Batch 0027/0043 | Averaged Loss: 0.0419
+Epoch: 088/100 | Batch 0028/0043 | Averaged Loss: 0.0491
+Epoch: 088/100 | Batch 0029/0043 | Averaged Loss: 0.0453
+Epoch: 088/100 | Batch 0030/0043 | Averaged Loss: 0.0522
+Epoch: 088/100 | Batch 0031/0043 | Averaged Loss: 0.0177
+Epoch: 088/100 | Batch 0032/0043 | Averaged Loss: 0.0479
+Epoch: 088/100 | Batch 0033/0043 | Averaged Loss: 0.0378
+Epoch: 088/100 | Batch 0034/0043 | Averaged Loss: 0.0352
+Epoch: 088/100 | Batch 0035/0043 | Averaged Loss: 0.0225
+Epoch: 088/100 | Batch 0036/0043 | Averaged Loss: 0.0369
+Epoch: 088/100 | Batch 0037/0043 | Averaged Loss: 0.0432
+Epoch: 088/100 | Batch 0038/0043 | Averaged Loss: 0.0464
+Epoch: 088/100 | Batch 0039/0043 | Averaged Loss: 0.0527
+Epoch: 088/100 | Batch 0040/0043 | Averaged Loss: 0.0384
+Epoch: 088/100 | Batch 0041/0043 | Averaged Loss: 0.0383
+Epoch: 088/100 | Batch 0042/0043 | Averaged Loss: 0.0410
+Epoch: 088/100 | Train: 98.68% | Validation: 75.88%
+Time elapsed: 13.06528377532959 min
+Epoch: 089/100 | Batch 0000/0043 | Averaged Loss: 0.0407
+Epoch: 089/100 | Batch 0001/0043 | Averaged Loss: 0.0447
+Epoch: 089/100 | Batch 0002/0043 | Averaged Loss: 0.0398
+Epoch: 089/100 | Batch 0003/0043 | Averaged Loss: 0.0297
+Epoch: 089/100 | Batch 0004/0043 | Averaged Loss: 0.0516
+Epoch: 089/100 | Batch 0005/0043 | Averaged Loss: 0.0594
+Epoch: 089/100 | Batch 0006/0043 | Averaged Loss: 0.0401
+Epoch: 089/100 | Batch 0007/0043 | Averaged Loss: 0.0697
+Epoch: 089/100 | Batch 0008/0043 | Averaged Loss: 0.0577
+Epoch: 089/100 | Batch 0009/0043 | Averaged Loss: 0.0284
+Epoch: 089/100 | Batch 0010/0043 | Averaged Loss: 0.0653
+Epoch: 089/100 | Batch 0011/0043 | Averaged Loss: 0.0513
+Epoch: 089/100 | Batch 0012/0043 | Averaged Loss: 0.0510
+Epoch: 089/100 | Batch 0013/0043 | Averaged Loss: 0.0678
+Epoch: 089/100 | Batch 0014/0043 | Averaged Loss: 0.0332
+Epoch: 089/100 | Batch 0015/0043 | Averaged Loss: 0.0506
+Epoch: 089/100 | Batch 0016/0043 | Averaged Loss: 0.0592
+Epoch: 089/100 | Batch 0017/0043 | Averaged Loss: 0.0447
+Epoch: 089/100 | Batch 0018/0043 | Averaged Loss: 0.0535
+Epoch: 089/100 | Batch 0019/0043 | Averaged Loss: 0.0339
+Epoch: 089/100 | Batch 0020/0043 | Averaged Loss: 0.0561
+Epoch: 089/100 | Batch 0021/0043 | Averaged Loss: 0.0451
+Epoch: 089/100 | Batch 0022/0043 | Averaged Loss: 0.0694
+Epoch: 089/100 | Batch 0023/0043 | Averaged Loss: 0.0541
+Epoch: 089/100 | Batch 0024/0043 | Averaged Loss: 0.0448
+Epoch: 089/100 | Batch 0025/0043 | Averaged Loss: 0.0393
+Epoch: 089/100 | Batch 0026/0043 | Averaged Loss: 0.0424
+Epoch: 089/100 | Batch 0027/0043 | Averaged Loss: 0.0322
+Epoch: 089/100 | Batch 0028/0043 | Averaged Loss: 0.0526
+Epoch: 089/100 | Batch 0029/0043 | Averaged Loss: 0.0430
+Epoch: 089/100 | Batch 0030/0043 | Averaged Loss: 0.0408
+Epoch: 089/100 | Batch 0031/0043 | Averaged Loss: 0.0232
+Epoch: 089/100 | Batch 0032/0043 | Averaged Loss: 0.0517
+Epoch: 089/100 | Batch 0033/0043 | Averaged Loss: 0.0278
+Epoch: 089/100 | Batch 0034/0043 | Averaged Loss: 0.0437
+Epoch: 089/100 | Batch 0035/0043 | Averaged Loss: 0.0495
+Epoch: 089/100 | Batch 0036/0043 | Averaged Loss: 0.0443
+Epoch: 089/100 | Batch 0037/0043 | Averaged Loss: 0.0439
+Epoch: 089/100 | Batch 0038/0043 | Averaged Loss: 0.0555
+Epoch: 089/100 | Batch 0039/0043 | Averaged Loss: 0.0181
+Epoch: 089/100 | Batch 0040/0043 | Averaged Loss: 0.0599
+Epoch: 089/100 | Batch 0041/0043 | Averaged Loss: 0.0419
+Epoch: 089/100 | Batch 0042/0043 | Averaged Loss: 0.0445
+Epoch: 089/100 | Train: 98.71% | Validation: 75.85%
+Time elapsed: 13.213732719421387 min
+Epoch: 090/100 | Batch 0000/0043 | Averaged Loss: 0.0299
+Epoch: 090/100 | Batch 0001/0043 | Averaged Loss: 0.0400
+Epoch: 090/100 | Batch 0002/0043 | Averaged Loss: 0.0345
+Epoch: 090/100 | Batch 0003/0043 | Averaged Loss: 0.0409
+Epoch: 090/100 | Batch 0004/0043 | Averaged Loss: 0.0388
+Epoch: 090/100 | Batch 0005/0043 | Averaged Loss: 0.0280
+Epoch: 090/100 | Batch 0006/0043 | Averaged Loss: 0.0307
+Epoch: 090/100 | Batch 0007/0043 | Averaged Loss: 0.0335
+Epoch: 090/100 | Batch 0008/0043 | Averaged Loss: 0.0285
+Epoch: 090/100 | Batch 0009/0043 | Averaged Loss: 0.0554
+Epoch: 090/100 | Batch 0010/0043 | Averaged Loss: 0.0616
+Epoch: 090/100 | Batch 0011/0043 | Averaged Loss: 0.0257
+Epoch: 090/100 | Batch 0012/0043 | Averaged Loss: 0.0470
+Epoch: 090/100 | Batch 0013/0043 | Averaged Loss: 0.0335
+Epoch: 090/100 | Batch 0014/0043 | Averaged Loss: 0.0465
+Epoch: 090/100 | Batch 0015/0043 | Averaged Loss: 0.0250
+Epoch: 090/100 | Batch 0016/0043 | Averaged Loss: 0.0341
+Epoch: 090/100 | Batch 0017/0043 | Averaged Loss: 0.0344
+Epoch: 090/100 | Batch 0018/0043 | Averaged Loss: 0.0560
+Epoch: 090/100 | Batch 0019/0043 | Averaged Loss: 0.0558
+Epoch: 090/100 | Batch 0020/0043 | Averaged Loss: 0.0396
+Epoch: 090/100 | Batch 0021/0043 | Averaged Loss: 0.0570
+Epoch: 090/100 | Batch 0022/0043 | Averaged Loss: 0.0442
+Epoch: 090/100 | Batch 0023/0043 | Averaged Loss: 0.0494
+Epoch: 090/100 | Batch 0024/0043 | Averaged Loss: 0.0615
+Epoch: 090/100 | Batch 0025/0043 | Averaged Loss: 0.0501
+Epoch: 090/100 | Batch 0026/0043 | Averaged Loss: 0.0714
+Epoch: 090/100 | Batch 0027/0043 | Averaged Loss: 0.0524
+Epoch: 090/100 | Batch 0028/0043 | Averaged Loss: 0.0396
+Epoch: 090/100 | Batch 0029/0043 | Averaged Loss: 0.0317
+Epoch: 090/100 | Batch 0030/0043 | Averaged Loss: 0.0409
+Epoch: 090/100 | Batch 0031/0043 | Averaged Loss: 0.0462
+Epoch: 090/100 | Batch 0032/0043 | Averaged Loss: 0.0534
+Epoch: 090/100 | Batch 0033/0043 | Averaged Loss: 0.0711
+Epoch: 090/100 | Batch 0034/0043 | Averaged Loss: 0.0518
+Epoch: 090/100 | Batch 0035/0043 | Averaged Loss: 0.0723
+Epoch: 090/100 | Batch 0036/0043 | Averaged Loss: 0.0391
+Epoch: 090/100 | Batch 0037/0043 | Averaged Loss: 0.0844
+Epoch: 090/100 | Batch 0038/0043 | Averaged Loss: 0.0574
+Epoch: 090/100 | Batch 0039/0043 | Averaged Loss: 0.0625
+Epoch: 090/100 | Batch 0040/0043 | Averaged Loss: 0.0642
+Epoch: 090/100 | Batch 0041/0043 | Averaged Loss: 0.0510
+Epoch: 090/100 | Batch 0042/0043 | Averaged Loss: 0.0590
+Epoch: 090/100 | Train: 98.10% | Validation: 75.49%
+Time elapsed: 13.36210823059082 min
+Epoch: 091/100 | Batch 0000/0043 | Averaged Loss: 0.0556
+Epoch: 091/100 | Batch 0001/0043 | Averaged Loss: 0.0627
+Epoch: 091/100 | Batch 0002/0043 | Averaged Loss: 0.0473
+Epoch: 091/100 | Batch 0003/0043 | Averaged Loss: 0.0664
+Epoch: 091/100 | Batch 0004/0043 | Averaged Loss: 0.0598
+Epoch: 091/100 | Batch 0005/0043 | Averaged Loss: 0.0464
+Epoch: 091/100 | Batch 0006/0043 | Averaged Loss: 0.0316
+Epoch: 091/100 | Batch 0007/0043 | Averaged Loss: 0.0665
+Epoch: 091/100 | Batch 0008/0043 | Averaged Loss: 0.0336
+Epoch: 091/100 | Batch 0009/0043 | Averaged Loss: 0.0516
+Epoch: 091/100 | Batch 0010/0043 | Averaged Loss: 0.0577
+Epoch: 091/100 | Batch 0011/0043 | Averaged Loss: 0.0782
+Epoch: 091/100 | Batch 0012/0043 | Averaged Loss: 0.0679
+Epoch: 091/100 | Batch 0013/0043 | Averaged Loss: 0.0608
+Epoch: 091/100 | Batch 0014/0043 | Averaged Loss: 0.0442
+Epoch: 091/100 | Batch 0015/0043 | Averaged Loss: 0.0615
+Epoch: 091/100 | Batch 0016/0043 | Averaged Loss: 0.0842
+Epoch: 091/100 | Batch 0017/0043 | Averaged Loss: 0.0589
+Epoch: 091/100 | Batch 0018/0043 | Averaged Loss: 0.0431
+Epoch: 091/100 | Batch 0019/0043 | Averaged Loss: 0.0464
+Epoch: 091/100 | Batch 0020/0043 | Averaged Loss: 0.0475
+Epoch: 091/100 | Batch 0021/0043 | Averaged Loss: 0.0648
+Epoch: 091/100 | Batch 0022/0043 | Averaged Loss: 0.0443
+Epoch: 091/100 | Batch 0023/0043 | Averaged Loss: 0.0647
+Epoch: 091/100 | Batch 0024/0043 | Averaged Loss: 0.0559
+Epoch: 091/100 | Batch 0025/0043 | Averaged Loss: 0.0594
+Epoch: 091/100 | Batch 0026/0043 | Averaged Loss: 0.0424
+Epoch: 091/100 | Batch 0027/0043 | Averaged Loss: 0.0452
+Epoch: 091/100 | Batch 0028/0043 | Averaged Loss: 0.0711
+Epoch: 091/100 | Batch 0029/0043 | Averaged Loss: 0.0515
+Epoch: 091/100 | Batch 0030/0043 | Averaged Loss: 0.0351
+Epoch: 091/100 | Batch 0031/0043 | Averaged Loss: 0.0402
+Epoch: 091/100 | Batch 0032/0043 | Averaged Loss: 0.0360
+Epoch: 091/100 | Batch 0033/0043 | Averaged Loss: 0.0374
+Epoch: 091/100 | Batch 0034/0043 | Averaged Loss: 0.0295
+Epoch: 091/100 | Batch 0035/0043 | Averaged Loss: 0.0389
+Epoch: 091/100 | Batch 0036/0043 | Averaged Loss: 0.0562
+Epoch: 091/100 | Batch 0037/0043 | Averaged Loss: 0.0413
+Epoch: 091/100 | Batch 0038/0043 | Averaged Loss: 0.0322
+Epoch: 091/100 | Batch 0039/0043 | Averaged Loss: 0.0279
+Epoch: 091/100 | Batch 0040/0043 | Averaged Loss: 0.0417
+Epoch: 091/100 | Batch 0041/0043 | Averaged Loss: 0.0379
+Epoch: 091/100 | Batch 0042/0043 | Averaged Loss: 0.0542
+Epoch: 091/100 | Train: 99.09% | Validation: 76.00%
+Time elapsed: 13.510790824890137 min
+Epoch: 092/100 | Batch 0000/0043 | Averaged Loss: 0.0248
+Epoch: 092/100 | Batch 0001/0043 | Averaged Loss: 0.0428
+Epoch: 092/100 | Batch 0002/0043 | Averaged Loss: 0.0413
+Epoch: 092/100 | Batch 0003/0043 | Averaged Loss: 0.0531
+Epoch: 092/100 | Batch 0004/0043 | Averaged Loss: 0.0444
+Epoch: 092/100 | Batch 0005/0043 | Averaged Loss: 0.0427
+Epoch: 092/100 | Batch 0006/0043 | Averaged Loss: 0.0289
+Epoch: 092/100 | Batch 0007/0043 | Averaged Loss: 0.0701
+Epoch: 092/100 | Batch 0008/0043 | Averaged Loss: 0.0308
+Epoch: 092/100 | Batch 0009/0043 | Averaged Loss: 0.0552
+Epoch: 092/100 | Batch 0010/0043 | Averaged Loss: 0.0497
+Epoch: 092/100 | Batch 0011/0043 | Averaged Loss: 0.0525
+Epoch: 092/100 | Batch 0012/0043 | Averaged Loss: 0.0331
+Epoch: 092/100 | Batch 0013/0043 | Averaged Loss: 0.0410
+Epoch: 092/100 | Batch 0014/0043 | Averaged Loss: 0.0588
+Epoch: 092/100 | Batch 0015/0043 | Averaged Loss: 0.0608
+Epoch: 092/100 | Batch 0016/0043 | Averaged Loss: 0.0357
+Epoch: 092/100 | Batch 0017/0043 | Averaged Loss: 0.0407
+Epoch: 092/100 | Batch 0018/0043 | Averaged Loss: 0.0528
+Epoch: 092/100 | Batch 0019/0043 | Averaged Loss: 0.0460
+Epoch: 092/100 | Batch 0020/0043 | Averaged Loss: 0.0649
+Epoch: 092/100 | Batch 0021/0043 | Averaged Loss: 0.0827
+Epoch: 092/100 | Batch 0022/0043 | Averaged Loss: 0.0459
+Epoch: 092/100 | Batch 0023/0043 | Averaged Loss: 0.0376
+Epoch: 092/100 | Batch 0024/0043 | Averaged Loss: 0.0697
+Epoch: 092/100 | Batch 0025/0043 | Averaged Loss: 0.0525
+Epoch: 092/100 | Batch 0026/0043 | Averaged Loss: 0.0529
+Epoch: 092/100 | Batch 0027/0043 | Averaged Loss: 0.0840
+Epoch: 092/100 | Batch 0028/0043 | Averaged Loss: 0.0301
+Epoch: 092/100 | Batch 0029/0043 | Averaged Loss: 0.0691
+Epoch: 092/100 | Batch 0030/0043 | Averaged Loss: 0.0708
+Epoch: 092/100 | Batch 0031/0043 | Averaged Loss: 0.0384
+Epoch: 092/100 | Batch 0032/0043 | Averaged Loss: 0.0247
+Epoch: 092/100 | Batch 0033/0043 | Averaged Loss: 0.0422
+Epoch: 092/100 | Batch 0034/0043 | Averaged Loss: 0.0481
+Epoch: 092/100 | Batch 0035/0043 | Averaged Loss: 0.0778
+Epoch: 092/100 | Batch 0036/0043 | Averaged Loss: 0.0572
+Epoch: 092/100 | Batch 0037/0043 | Averaged Loss: 0.0315
+Epoch: 092/100 | Batch 0038/0043 | Averaged Loss: 0.0345
+Epoch: 092/100 | Batch 0039/0043 | Averaged Loss: 0.0408
+Epoch: 092/100 | Batch 0040/0043 | Averaged Loss: 0.0237
+Epoch: 092/100 | Batch 0041/0043 | Averaged Loss: 0.0455
+Epoch: 092/100 | Batch 0042/0043 | Averaged Loss: 0.0418
+Epoch: 092/100 | Train: 98.85% | Validation: 75.63%
+Time elapsed: 13.659354209899902 min
+Epoch: 093/100 | Batch 0000/0043 | Averaged Loss: 0.0365
+Epoch: 093/100 | Batch 0001/0043 | Averaged Loss: 0.0280
+Epoch: 093/100 | Batch 0002/0043 | Averaged Loss: 0.0676
+Epoch: 093/100 | Batch 0003/0043 | Averaged Loss: 0.0344
+Epoch: 093/100 | Batch 0004/0043 | Averaged Loss: 0.0434
+Epoch: 093/100 | Batch 0005/0043 | Averaged Loss: 0.0446
+Epoch: 093/100 | Batch 0006/0043 | Averaged Loss: 0.0303
+Epoch: 093/100 | Batch 0007/0043 | Averaged Loss: 0.0344
+Epoch: 093/100 | Batch 0008/0043 | Averaged Loss: 0.0445
+Epoch: 093/100 | Batch 0009/0043 | Averaged Loss: 0.0451
+Epoch: 093/100 | Batch 0010/0043 | Averaged Loss: 0.0797
+Epoch: 093/100 | Batch 0011/0043 | Averaged Loss: 0.0591
+Epoch: 093/100 | Batch 0012/0043 | Averaged Loss: 0.0704
+Epoch: 093/100 | Batch 0013/0043 | Averaged Loss: 0.0742
+Epoch: 093/100 | Batch 0014/0043 | Averaged Loss: 0.0414
+Epoch: 093/100 | Batch 0015/0043 | Averaged Loss: 0.0465
+Epoch: 093/100 | Batch 0016/0043 | Averaged Loss: 0.0342
+Epoch: 093/100 | Batch 0017/0043 | Averaged Loss: 0.0592
+Epoch: 093/100 | Batch 0018/0043 | Averaged Loss: 0.0517
+Epoch: 093/100 | Batch 0019/0043 | Averaged Loss: 0.0422
+Epoch: 093/100 | Batch 0020/0043 | Averaged Loss: 0.0297
+Epoch: 093/100 | Batch 0021/0043 | Averaged Loss: 0.0530
+Epoch: 093/100 | Batch 0022/0043 | Averaged Loss: 0.0415
+Epoch: 093/100 | Batch 0023/0043 | Averaged Loss: 0.0876
+Epoch: 093/100 | Batch 0024/0043 | Averaged Loss: 0.0497
+Epoch: 093/100 | Batch 0025/0043 | Averaged Loss: 0.0306
+Epoch: 093/100 | Batch 0026/0043 | Averaged Loss: 0.0386
+Epoch: 093/100 | Batch 0027/0043 | Averaged Loss: 0.0367
+Epoch: 093/100 | Batch 0028/0043 | Averaged Loss: 0.0505
+Epoch: 093/100 | Batch 0029/0043 | Averaged Loss: 0.0461
+Epoch: 093/100 | Batch 0030/0043 | Averaged Loss: 0.0654
+Epoch: 093/100 | Batch 0031/0043 | Averaged Loss: 0.0313
+Epoch: 093/100 | Batch 0032/0043 | Averaged Loss: 0.0289
+Epoch: 093/100 | Batch 0033/0043 | Averaged Loss: 0.0404
+Epoch: 093/100 | Batch 0034/0043 | Averaged Loss: 0.0417
+Epoch: 093/100 | Batch 0035/0043 | Averaged Loss: 0.0543
+Epoch: 093/100 | Batch 0036/0043 | Averaged Loss: 0.0318
+Epoch: 093/100 | Batch 0037/0043 | Averaged Loss: 0.0456
+Epoch: 093/100 | Batch 0038/0043 | Averaged Loss: 0.0465
+Epoch: 093/100 | Batch 0039/0043 | Averaged Loss: 0.0311
+Epoch: 093/100 | Batch 0040/0043 | Averaged Loss: 0.0519
+Epoch: 093/100 | Batch 0041/0043 | Averaged Loss: 0.0236
+Epoch: 093/100 | Batch 0042/0043 | Averaged Loss: 0.0543
+Epoch: 093/100 | Train: 98.87% | Validation: 75.68%
+Time elapsed: 13.80755615234375 min
+Epoch: 094/100 | Batch 0000/0043 | Averaged Loss: 0.0353
+Epoch: 094/100 | Batch 0001/0043 | Averaged Loss: 0.0276
+Epoch: 094/100 | Batch 0002/0043 | Averaged Loss: 0.0357
+Epoch: 094/100 | Batch 0003/0043 | Averaged Loss: 0.0282
+Epoch: 094/100 | Batch 0004/0043 | Averaged Loss: 0.0333
+Epoch: 094/100 | Batch 0005/0043 | Averaged Loss: 0.0126
+Epoch: 094/100 | Batch 0006/0043 | Averaged Loss: 0.0326
+Epoch: 094/100 | Batch 0007/0043 | Averaged Loss: 0.0369
+Epoch: 094/100 | Batch 0008/0043 | Averaged Loss: 0.0502
+Epoch: 094/100 | Batch 0009/0043 | Averaged Loss: 0.0410
+Epoch: 094/100 | Batch 0010/0043 | Averaged Loss: 0.0302
+Epoch: 094/100 | Batch 0011/0043 | Averaged Loss: 0.0470
+Epoch: 094/100 | Batch 0012/0043 | Averaged Loss: 0.0389
+Epoch: 094/100 | Batch 0013/0043 | Averaged Loss: 0.0763
+Epoch: 094/100 | Batch 0014/0043 | Averaged Loss: 0.0394
+Epoch: 094/100 | Batch 0015/0043 | Averaged Loss: 0.0440
+Epoch: 094/100 | Batch 0016/0043 | Averaged Loss: 0.0270
+Epoch: 094/100 | Batch 0017/0043 | Averaged Loss: 0.0342
+Epoch: 094/100 | Batch 0018/0043 | Averaged Loss: 0.0349
+Epoch: 094/100 | Batch 0019/0043 | Averaged Loss: 0.0510
+Epoch: 094/100 | Batch 0020/0043 | Averaged Loss: 0.0427
+Epoch: 094/100 | Batch 0021/0043 | Averaged Loss: 0.0606
+Epoch: 094/100 | Batch 0022/0043 | Averaged Loss: 0.0657
+Epoch: 094/100 | Batch 0023/0043 | Averaged Loss: 0.0439
+Epoch: 094/100 | Batch 0024/0043 | Averaged Loss: 0.0596
+Epoch: 094/100 | Batch 0025/0043 | Averaged Loss: 0.0370
+Epoch: 094/100 | Batch 0026/0043 | Averaged Loss: 0.0514
+Epoch: 094/100 | Batch 0027/0043 | Averaged Loss: 0.0665
+Epoch: 094/100 | Batch 0028/0043 | Averaged Loss: 0.0465
+Epoch: 094/100 | Batch 0029/0043 | Averaged Loss: 0.0716
+Epoch: 094/100 | Batch 0030/0043 | Averaged Loss: 0.0666
+Epoch: 094/100 | Batch 0031/0043 | Averaged Loss: 0.0684
+Epoch: 094/100 | Batch 0032/0043 | Averaged Loss: 0.0432
+Epoch: 094/100 | Batch 0033/0043 | Averaged Loss: 0.0401
+Epoch: 094/100 | Batch 0034/0043 | Averaged Loss: 0.0552
+Epoch: 094/100 | Batch 0035/0043 | Averaged Loss: 0.0469
+Epoch: 094/100 | Batch 0036/0043 | Averaged Loss: 0.0657
+Epoch: 094/100 | Batch 0037/0043 | Averaged Loss: 0.0548
+Epoch: 094/100 | Batch 0038/0043 | Averaged Loss: 0.0546
+Epoch: 094/100 | Batch 0039/0043 | Averaged Loss: 0.0831
+Epoch: 094/100 | Batch 0040/0043 | Averaged Loss: 0.0607
+Epoch: 094/100 | Batch 0041/0043 | Averaged Loss: 0.0434
+Epoch: 094/100 | Batch 0042/0043 | Averaged Loss: 0.0590
+Epoch: 094/100 | Train: 98.57% | Validation: 75.02%
+Time elapsed: 13.955704689025879 min
+Epoch: 095/100 | Batch 0000/0043 | Averaged Loss: 0.0701
+Epoch: 095/100 | Batch 0001/0043 | Averaged Loss: 0.0452
+Epoch: 095/100 | Batch 0002/0043 | Averaged Loss: 0.0571
+Epoch: 095/100 | Batch 0003/0043 | Averaged Loss: 0.0616
+Epoch: 095/100 | Batch 0004/0043 | Averaged Loss: 0.0659
+Epoch: 095/100 | Batch 0005/0043 | Averaged Loss: 0.0700
+Epoch: 095/100 | Batch 0006/0043 | Averaged Loss: 0.0750
+Epoch: 095/100 | Batch 0007/0043 | Averaged Loss: 0.0484
+Epoch: 095/100 | Batch 0008/0043 | Averaged Loss: 0.0495
+Epoch: 095/100 | Batch 0009/0043 | Averaged Loss: 0.0639
+Epoch: 095/100 | Batch 0010/0043 | Averaged Loss: 0.0581
+Epoch: 095/100 | Batch 0011/0043 | Averaged Loss: 0.0493
+Epoch: 095/100 | Batch 0012/0043 | Averaged Loss: 0.0467
+Epoch: 095/100 | Batch 0013/0043 | Averaged Loss: 0.0486
+Epoch: 095/100 | Batch 0014/0043 | Averaged Loss: 0.0566
+Epoch: 095/100 | Batch 0015/0043 | Averaged Loss: 0.0398
+Epoch: 095/100 | Batch 0016/0043 | Averaged Loss: 0.0386
+Epoch: 095/100 | Batch 0017/0043 | Averaged Loss: 0.0502
+Epoch: 095/100 | Batch 0018/0043 | Averaged Loss: 0.0389
+Epoch: 095/100 | Batch 0019/0043 | Averaged Loss: 0.0593
+Epoch: 095/100 | Batch 0020/0043 | Averaged Loss: 0.0360
+Epoch: 095/100 | Batch 0021/0043 | Averaged Loss: 0.0440
+Epoch: 095/100 | Batch 0022/0043 | Averaged Loss: 0.0458
+Epoch: 095/100 | Batch 0023/0043 | Averaged Loss: 0.0350
+Epoch: 095/100 | Batch 0024/0043 | Averaged Loss: 0.0368
+Epoch: 095/100 | Batch 0025/0043 | Averaged Loss: 0.0596
+Epoch: 095/100 | Batch 0026/0043 | Averaged Loss: 0.0315
+Epoch: 095/100 | Batch 0027/0043 | Averaged Loss: 0.0480
+Epoch: 095/100 | Batch 0028/0043 | Averaged Loss: 0.0591
+Epoch: 095/100 | Batch 0029/0043 | Averaged Loss: 0.0398
+Epoch: 095/100 | Batch 0030/0043 | Averaged Loss: 0.0498
+Epoch: 095/100 | Batch 0031/0043 | Averaged Loss: 0.0431
+Epoch: 095/100 | Batch 0032/0043 | Averaged Loss: 0.0632
+Epoch: 095/100 | Batch 0033/0043 | Averaged Loss: 0.0396
+Epoch: 095/100 | Batch 0034/0043 | Averaged Loss: 0.0304
+Epoch: 095/100 | Batch 0035/0043 | Averaged Loss: 0.0426
+Epoch: 095/100 | Batch 0036/0043 | Averaged Loss: 0.0588
+Epoch: 095/100 | Batch 0037/0043 | Averaged Loss: 0.0699
+Epoch: 095/100 | Batch 0038/0043 | Averaged Loss: 0.0383
+Epoch: 095/100 | Batch 0039/0043 | Averaged Loss: 0.0356
+Epoch: 095/100 | Batch 0040/0043 | Averaged Loss: 0.0413
+Epoch: 095/100 | Batch 0041/0043 | Averaged Loss: 0.0295
+Epoch: 095/100 | Batch 0042/0043 | Averaged Loss: 0.0583
+Epoch: 095/100 | Train: 98.91% | Validation: 75.46%
+Time elapsed: 14.104061126708984 min
+Epoch: 096/100 | Batch 0000/0043 | Averaged Loss: 0.0417
+Epoch: 096/100 | Batch 0001/0043 | Averaged Loss: 0.0331
+Epoch: 096/100 | Batch 0002/0043 | Averaged Loss: 0.0252
+Epoch: 096/100 | Batch 0003/0043 | Averaged Loss: 0.0261
+Epoch: 096/100 | Batch 0004/0043 | Averaged Loss: 0.0631
+Epoch: 096/100 | Batch 0005/0043 | Averaged Loss: 0.0347
+Epoch: 096/100 | Batch 0006/0043 | Averaged Loss: 0.0341
+Epoch: 096/100 | Batch 0007/0043 | Averaged Loss: 0.0433
+Epoch: 096/100 | Batch 0008/0043 | Averaged Loss: 0.0247
+Epoch: 096/100 | Batch 0009/0043 | Averaged Loss: 0.0611
+Epoch: 096/100 | Batch 0010/0043 | Averaged Loss: 0.0257
+Epoch: 096/100 | Batch 0011/0043 | Averaged Loss: 0.0440
+Epoch: 096/100 | Batch 0012/0043 | Averaged Loss: 0.0250
+Epoch: 096/100 | Batch 0013/0043 | Averaged Loss: 0.0626
+Epoch: 096/100 | Batch 0014/0043 | Averaged Loss: 0.0295
+Epoch: 096/100 | Batch 0015/0043 | Averaged Loss: 0.0426
+Epoch: 096/100 | Batch 0016/0043 | Averaged Loss: 0.0248
+Epoch: 096/100 | Batch 0017/0043 | Averaged Loss: 0.0246
+Epoch: 096/100 | Batch 0018/0043 | Averaged Loss: 0.0275
+Epoch: 096/100 | Batch 0019/0043 | Averaged Loss: 0.0359
+Epoch: 096/100 | Batch 0020/0043 | Averaged Loss: 0.0361
+Epoch: 096/100 | Batch 0021/0043 | Averaged Loss: 0.0314
+Epoch: 096/100 | Batch 0022/0043 | Averaged Loss: 0.0332
+Epoch: 096/100 | Batch 0023/0043 | Averaged Loss: 0.0216
+Epoch: 096/100 | Batch 0024/0043 | Averaged Loss: 0.0465
+Epoch: 096/100 | Batch 0025/0043 | Averaged Loss: 0.0366
+Epoch: 096/100 | Batch 0026/0043 | Averaged Loss: 0.0403
+Epoch: 096/100 | Batch 0027/0043 | Averaged Loss: 0.0456
+Epoch: 096/100 | Batch 0028/0043 | Averaged Loss: 0.0425
+Epoch: 096/100 | Batch 0029/0043 | Averaged Loss: 0.0343
+Epoch: 096/100 | Batch 0030/0043 | Averaged Loss: 0.0481
+Epoch: 096/100 | Batch 0031/0043 | Averaged Loss: 0.0412
+Epoch: 096/100 | Batch 0032/0043 | Averaged Loss: 0.0473
+Epoch: 096/100 | Batch 0033/0043 | Averaged Loss: 0.0501
+Epoch: 096/100 | Batch 0034/0043 | Averaged Loss: 0.0475
+Epoch: 096/100 | Batch 0035/0043 | Averaged Loss: 0.0326
+Epoch: 096/100 | Batch 0036/0043 | Averaged Loss: 0.0491
+Epoch: 096/100 | Batch 0037/0043 | Averaged Loss: 0.0350
+Epoch: 096/100 | Batch 0038/0043 | Averaged Loss: 0.0392
+Epoch: 096/100 | Batch 0039/0043 | Averaged Loss: 0.0378
+Epoch: 096/100 | Batch 0040/0043 | Averaged Loss: 0.0662
+Epoch: 096/100 | Batch 0041/0043 | Averaged Loss: 0.0588
+Epoch: 096/100 | Batch 0042/0043 | Averaged Loss: 0.0323
+Epoch: 096/100 | Train: 98.57% | Validation: 74.68%
+Time elapsed: 14.252117156982422 min
+Epoch: 097/100 | Batch 0000/0043 | Averaged Loss: 0.0418
+Epoch: 097/100 | Batch 0001/0043 | Averaged Loss: 0.0388
+Epoch: 097/100 | Batch 0002/0043 | Averaged Loss: 0.0700
+Epoch: 097/100 | Batch 0003/0043 | Averaged Loss: 0.0374
+Epoch: 097/100 | Batch 0004/0043 | Averaged Loss: 0.0389
+Epoch: 097/100 | Batch 0005/0043 | Averaged Loss: 0.0310
+Epoch: 097/100 | Batch 0006/0043 | Averaged Loss: 0.0605
+Epoch: 097/100 | Batch 0007/0043 | Averaged Loss: 0.0438
+Epoch: 097/100 | Batch 0008/0043 | Averaged Loss: 0.0315
+Epoch: 097/100 | Batch 0009/0043 | Averaged Loss: 0.0317
+Epoch: 097/100 | Batch 0010/0043 | Averaged Loss: 0.0383
+Epoch: 097/100 | Batch 0011/0043 | Averaged Loss: 0.0314
+Epoch: 097/100 | Batch 0012/0043 | Averaged Loss: 0.0255
+Epoch: 097/100 | Batch 0013/0043 | Averaged Loss: 0.0545
+Epoch: 097/100 | Batch 0014/0043 | Averaged Loss: 0.0524
+Epoch: 097/100 | Batch 0015/0043 | Averaged Loss: 0.0285
+Epoch: 097/100 | Batch 0016/0043 | Averaged Loss: 0.0482
+Epoch: 097/100 | Batch 0017/0043 | Averaged Loss: 0.0386
+Epoch: 097/100 | Batch 0018/0043 | Averaged Loss: 0.0463
+Epoch: 097/100 | Batch 0019/0043 | Averaged Loss: 0.0528
+Epoch: 097/100 | Batch 0020/0043 | Averaged Loss: 0.0222
+Epoch: 097/100 | Batch 0021/0043 | Averaged Loss: 0.0296
+Epoch: 097/100 | Batch 0022/0043 | Averaged Loss: 0.0305
+Epoch: 097/100 | Batch 0023/0043 | Averaged Loss: 0.0261
+Epoch: 097/100 | Batch 0024/0043 | Averaged Loss: 0.0665
+Epoch: 097/100 | Batch 0025/0043 | Averaged Loss: 0.0313
+Epoch: 097/100 | Batch 0026/0043 | Averaged Loss: 0.0566
+Epoch: 097/100 | Batch 0027/0043 | Averaged Loss: 0.0445
+Epoch: 097/100 | Batch 0028/0043 | Averaged Loss: 0.0469
+Epoch: 097/100 | Batch 0029/0043 | Averaged Loss: 0.0551
+Epoch: 097/100 | Batch 0030/0043 | Averaged Loss: 0.0281
+Epoch: 097/100 | Batch 0031/0043 | Averaged Loss: 0.0377
+Epoch: 097/100 | Batch 0032/0043 | Averaged Loss: 0.0272
+Epoch: 097/100 | Batch 0033/0043 | Averaged Loss: 0.0408
+Epoch: 097/100 | Batch 0034/0043 | Averaged Loss: 0.0395
+Epoch: 097/100 | Batch 0035/0043 | Averaged Loss: 0.0294
+Epoch: 097/100 | Batch 0036/0043 | Averaged Loss: 0.0337
+Epoch: 097/100 | Batch 0037/0043 | Averaged Loss: 0.0309
+Epoch: 097/100 | Batch 0038/0043 | Averaged Loss: 0.0319
+Epoch: 097/100 | Batch 0039/0043 | Averaged Loss: 0.0370
+Epoch: 097/100 | Batch 0040/0043 | Averaged Loss: 0.0571
+Epoch: 097/100 | Batch 0041/0043 | Averaged Loss: 0.0356
+Epoch: 097/100 | Batch 0042/0043 | Averaged Loss: 0.0305
+Epoch: 097/100 | Train: 99.15% | Validation: 76.61%
+Time elapsed: 14.400346755981445 min
+Epoch: 098/100 | Batch 0000/0043 | Averaged Loss: 0.0304
+Epoch: 098/100 | Batch 0001/0043 | Averaged Loss: 0.0406
+Epoch: 098/100 | Batch 0002/0043 | Averaged Loss: 0.0533
+Epoch: 098/100 | Batch 0003/0043 | Averaged Loss: 0.0418
+Epoch: 098/100 | Batch 0004/0043 | Averaged Loss: 0.0286
+Epoch: 098/100 | Batch 0005/0043 | Averaged Loss: 0.0217
+Epoch: 098/100 | Batch 0006/0043 | Averaged Loss: 0.0502
+Epoch: 098/100 | Batch 0007/0043 | Averaged Loss: 0.0305
+Epoch: 098/100 | Batch 0008/0043 | Averaged Loss: 0.0327
+Epoch: 098/100 | Batch 0009/0043 | Averaged Loss: 0.0302
+Epoch: 098/100 | Batch 0010/0043 | Averaged Loss: 0.0420
+Epoch: 098/100 | Batch 0011/0043 | Averaged Loss: 0.0378
+Epoch: 098/100 | Batch 0012/0043 | Averaged Loss: 0.0261
+Epoch: 098/100 | Batch 0013/0043 | Averaged Loss: 0.0239
+Epoch: 098/100 | Batch 0014/0043 | Averaged Loss: 0.0341
+Epoch: 098/100 | Batch 0015/0043 | Averaged Loss: 0.0343
+Epoch: 098/100 | Batch 0016/0043 | Averaged Loss: 0.0277
+Epoch: 098/100 | Batch 0017/0043 | Averaged Loss: 0.0346
+Epoch: 098/100 | Batch 0018/0043 | Averaged Loss: 0.0304
+Epoch: 098/100 | Batch 0019/0043 | Averaged Loss: 0.0293
+Epoch: 098/100 | Batch 0020/0043 | Averaged Loss: 0.0255
+Epoch: 098/100 | Batch 0021/0043 | Averaged Loss: 0.0434
+Epoch: 098/100 | Batch 0022/0043 | Averaged Loss: 0.0637
+Epoch: 098/100 | Batch 0023/0043 | Averaged Loss: 0.0192
+Epoch: 098/100 | Batch 0024/0043 | Averaged Loss: 0.0372
+Epoch: 098/100 | Batch 0025/0043 | Averaged Loss: 0.0460
+Epoch: 098/100 | Batch 0026/0043 | Averaged Loss: 0.0791
+Epoch: 098/100 | Batch 0027/0043 | Averaged Loss: 0.0550
+Epoch: 098/100 | Batch 0028/0043 | Averaged Loss: 0.0170
+Epoch: 098/100 | Batch 0029/0043 | Averaged Loss: 0.0362
+Epoch: 098/100 | Batch 0030/0043 | Averaged Loss: 0.0414
+Epoch: 098/100 | Batch 0031/0043 | Averaged Loss: 0.0513
+Epoch: 098/100 | Batch 0032/0043 | Averaged Loss: 0.0384
+Epoch: 098/100 | Batch 0033/0043 | Averaged Loss: 0.0393
+Epoch: 098/100 | Batch 0034/0043 | Averaged Loss: 0.0499
+Epoch: 098/100 | Batch 0035/0043 | Averaged Loss: 0.0446
+Epoch: 098/100 | Batch 0036/0043 | Averaged Loss: 0.0511
+Epoch: 098/100 | Batch 0037/0043 | Averaged Loss: 0.0373
+Epoch: 098/100 | Batch 0038/0043 | Averaged Loss: 0.0500
+Epoch: 098/100 | Batch 0039/0043 | Averaged Loss: 0.0411
+Epoch: 098/100 | Batch 0040/0043 | Averaged Loss: 0.0226
+Epoch: 098/100 | Batch 0041/0043 | Averaged Loss: 0.0570
+Epoch: 098/100 | Batch 0042/0043 | Averaged Loss: 0.0412
+Epoch: 098/100 | Train: 98.72% | Validation: 75.07%
+Time elapsed: 14.548616409301758 min
+Epoch: 099/100 | Batch 0000/0043 | Averaged Loss: 0.0227
+Epoch: 099/100 | Batch 0001/0043 | Averaged Loss: 0.0232
+Epoch: 099/100 | Batch 0002/0043 | Averaged Loss: 0.0371
+Epoch: 099/100 | Batch 0003/0043 | Averaged Loss: 0.0316
+Epoch: 099/100 | Batch 0004/0043 | Averaged Loss: 0.0289
+Epoch: 099/100 | Batch 0005/0043 | Averaged Loss: 0.0545
+Epoch: 099/100 | Batch 0006/0043 | Averaged Loss: 0.0657
+Epoch: 099/100 | Batch 0007/0043 | Averaged Loss: 0.0397
+Epoch: 099/100 | Batch 0008/0043 | Averaged Loss: 0.0520
+Epoch: 099/100 | Batch 0009/0043 | Averaged Loss: 0.0502
+Epoch: 099/100 | Batch 0010/0043 | Averaged Loss: 0.0468
+Epoch: 099/100 | Batch 0011/0043 | Averaged Loss: 0.0442
+Epoch: 099/100 | Batch 0012/0043 | Averaged Loss: 0.0455
+Epoch: 099/100 | Batch 0013/0043 | Averaged Loss: 0.0441
+Epoch: 099/100 | Batch 0014/0043 | Averaged Loss: 0.0351
+Epoch: 099/100 | Batch 0015/0043 | Averaged Loss: 0.0543
+Epoch: 099/100 | Batch 0016/0043 | Averaged Loss: 0.0356
+Epoch: 099/100 | Batch 0017/0043 | Averaged Loss: 0.0383
+Epoch: 099/100 | Batch 0018/0043 | Averaged Loss: 0.0489
+Epoch: 099/100 | Batch 0019/0043 | Averaged Loss: 0.0359
+Epoch: 099/100 | Batch 0020/0043 | Averaged Loss: 0.0368
+Epoch: 099/100 | Batch 0021/0043 | Averaged Loss: 0.0444
+Epoch: 099/100 | Batch 0022/0043 | Averaged Loss: 0.0549
+Epoch: 099/100 | Batch 0023/0043 | Averaged Loss: 0.0569
+Epoch: 099/100 | Batch 0024/0043 | Averaged Loss: 0.0436
+Epoch: 099/100 | Batch 0025/0043 | Averaged Loss: 0.0340
+Epoch: 099/100 | Batch 0026/0043 | Averaged Loss: 0.0407
+Epoch: 099/100 | Batch 0027/0043 | Averaged Loss: 0.0239
+Epoch: 099/100 | Batch 0028/0043 | Averaged Loss: 0.0475
+Epoch: 099/100 | Batch 0029/0043 | Averaged Loss: 0.0339
+Epoch: 099/100 | Batch 0030/0043 | Averaged Loss: 0.0329
+Epoch: 099/100 | Batch 0031/0043 | Averaged Loss: 0.0527
+Epoch: 099/100 | Batch 0032/0043 | Averaged Loss: 0.0269
+Epoch: 099/100 | Batch 0033/0043 | Averaged Loss: 0.0451
+Epoch: 099/100 | Batch 0034/0043 | Averaged Loss: 0.0567
+Epoch: 099/100 | Batch 0035/0043 | Averaged Loss: 0.0617
+Epoch: 099/100 | Batch 0036/0043 | Averaged Loss: 0.0395
+Epoch: 099/100 | Batch 0037/0043 | Averaged Loss: 0.0482
+Epoch: 099/100 | Batch 0038/0043 | Averaged Loss: 0.0361
+Epoch: 099/100 | Batch 0039/0043 | Averaged Loss: 0.0556
+Epoch: 099/100 | Batch 0040/0043 | Averaged Loss: 0.0457
+Epoch: 099/100 | Batch 0041/0043 | Averaged Loss: 0.0575
+Epoch: 099/100 | Batch 0042/0043 | Averaged Loss: 0.0290
+Epoch: 099/100 | Train: 98.91% | Validation: 76.12%
+Time elapsed: 14.696898460388184 min
+Epoch: 100/100 | Batch 0000/0043 | Averaged Loss: 0.0309
+Epoch: 100/100 | Batch 0001/0043 | Averaged Loss: 0.0304
+Epoch: 100/100 | Batch 0002/0043 | Averaged Loss: 0.0345
+Epoch: 100/100 | Batch 0003/0043 | Averaged Loss: 0.0408
+Epoch: 100/100 | Batch 0004/0043 | Averaged Loss: 0.0235
+Epoch: 100/100 | Batch 0005/0043 | Averaged Loss: 0.0326
+Epoch: 100/100 | Batch 0006/0043 | Averaged Loss: 0.0314
+Epoch: 100/100 | Batch 0007/0043 | Averaged Loss: 0.0360
+Epoch: 100/100 | Batch 0008/0043 | Averaged Loss: 0.0590
+Epoch: 100/100 | Batch 0009/0043 | Averaged Loss: 0.0337
+Epoch: 100/100 | Batch 0010/0043 | Averaged Loss: 0.0646
+Epoch: 100/100 | Batch 0011/0043 | Averaged Loss: 0.0494
+Epoch: 100/100 | Batch 0012/0043 | Averaged Loss: 0.0343
+Epoch: 100/100 | Batch 0013/0043 | Averaged Loss: 0.0524
+Epoch: 100/100 | Batch 0014/0043 | Averaged Loss: 0.0242
+Epoch: 100/100 | Batch 0015/0043 | Averaged Loss: 0.0247
+Epoch: 100/100 | Batch 0016/0043 | Averaged Loss: 0.0404
+Epoch: 100/100 | Batch 0017/0043 | Averaged Loss: 0.0624
+Epoch: 100/100 | Batch 0018/0043 | Averaged Loss: 0.0533
+Epoch: 100/100 | Batch 0019/0043 | Averaged Loss: 0.0223
+Epoch: 100/100 | Batch 0020/0043 | Averaged Loss: 0.0241
+Epoch: 100/100 | Batch 0021/0043 | Averaged Loss: 0.0307
+Epoch: 100/100 | Batch 0022/0043 | Averaged Loss: 0.0166
+Epoch: 100/100 | Batch 0023/0043 | Averaged Loss: 0.0432
+Epoch: 100/100 | Batch 0024/0043 | Averaged Loss: 0.0254
+Epoch: 100/100 | Batch 0025/0043 | Averaged Loss: 0.0305
+Epoch: 100/100 | Batch 0026/0043 | Averaged Loss: 0.0318
+Epoch: 100/100 | Batch 0027/0043 | Averaged Loss: 0.0432
+Epoch: 100/100 | Batch 0028/0043 | Averaged Loss: 0.0296
+Epoch: 100/100 | Batch 0029/0043 | Averaged Loss: 0.0387
+Epoch: 100/100 | Batch 0030/0043 | Averaged Loss: 0.0496
+Epoch: 100/100 | Batch 0031/0043 | Averaged Loss: 0.0294
+Epoch: 100/100 | Batch 0032/0043 | Averaged Loss: 0.0198
+Epoch: 100/100 | Batch 0033/0043 | Averaged Loss: 0.0336
+Epoch: 100/100 | Batch 0034/0043 | Averaged Loss: 0.0331
+Epoch: 100/100 | Batch 0035/0043 | Averaged Loss: 0.0333
+Epoch: 100/100 | Batch 0036/0043 | Averaged Loss: 0.0460
+Epoch: 100/100 | Batch 0037/0043 | Averaged Loss: 0.0326
+Epoch: 100/100 | Batch 0038/0043 | Averaged Loss: 0.0321
+Epoch: 100/100 | Batch 0039/0043 | Averaged Loss: 0.0392
+Epoch: 100/100 | Batch 0040/0043 | Averaged Loss: 0.0219
+Epoch: 100/100 | Batch 0041/0043 | Averaged Loss: 0.0280
+Epoch: 100/100 | Batch 0042/0043 | Averaged Loss: 0.0286
+Epoch: 100/100 | Train: 99.00% | Validation: 75.63%
+Time elapsed: 14.845315933227539 min
+Total Training Time: 14.845317840576172 min
+Test accuracy 74.85%
+
+============================= JOB FEEDBACK =============================
+
+NodeName=uc2n520
+Job ID: 25137704
+Cluster: uc2
+User/Group: xk5289/scc
+State: COMPLETED (exit code 0)
+Nodes: 1
+Cores per node: 40
+CPU Utilized: 00:59:40
+CPU Efficiency: 9.35% of 10:38:00 core-walltime
+Job Wall-clock time: 00:15:57
+Memory Utilized: 5.79 GB (estimated maximum)
+Memory Efficiency: 4.64% of 125.00 GB (125.00 GB/node)
diff --git a/5_dpnn/results/gpu_4/train_acc_4_gpu.pt b/5_dpnn/results/gpu_4/train_acc_4_gpu.pt
new file mode 100644
index 0000000000000000000000000000000000000000..282187f56a943bace1295fc6ce14916f9253387d
GIT binary patch
literal 1792
zcma)7Ye<t(82+YiiuI=ptrb?JE|uEF-3*hCt9HmK&C9l97sk>hW5em1*##-OX%|YO
ztCd)oWmZr@ra?Cpx)4-^zx<_ON@Y|Mlz)29&Y8HaMF-Bl_dDl(pZ7WM^PX=Ob9{mz
zBqa$^W4@3q1pH2SwZmCa;m|s&YU-4g&VW-<<MEn|5@BO-?2Fq{D`a4lI+@Z}KWh-9
zw2zs$I_|Y#q`W&Jee1}_DC4;!zd<|!KEL>i`T}e;NHEH-A62$JK7f(-VlY2Oxh;dZ
zHl-h<)o-|wKXPIyw`m-swN(S9<we^t!fr7I!f5^C-!n#)S1>BL`FKB=7!?*T0c$2k
zMSbI`4^IqXw83BC$-H$PBa8WsOLuwzqvBo1AnOw_nmQTP;-c4Vr9?_yN--)iLQuzj
zj7kfQ0IULI4Wbmm*e8P1NHHo~NYB>*M&fhkQ&wKYQk4DdhFa9Z){nWAG)C5DEvtq#
z?Q0uQgq3sr?AA#hZvD>BWz&CR*=_Z_n6|v9EW7Rgo~H5C9*k@s#u$0K>>w>XM)q_c
zL$+IKy@8?KLkXP$V~>rM&R)W4*z4x;mELBig?*?33bqQIA8{NamEUW5<-v$7{S1=6
zf~^tGkSyRhB&(lkVgHiON=HO5p!6ppsvzXNjx!`HsZ3+ArQ@X|3YsAcR+dq@AUoN5
zmJuVHE<q>s!Aj$1ZtR@{=h4H;W?moSRWOywl3^a+Xke8m-pi~Z$x$~e6shA>Ni!FP
z)bRcx?GT-N1dva0xyiLSrYSkk+lCzAjYO_=F?GnXP*mGD=>FLdeuIlkx<co=ZDy@S
z4qt^XP6MBQ-W%ir7ngMN1|aR#{5U-!z$s!cxs>FNh3Q3l53$OVC(o$dE``OMm{i!=
zc=dR^Akb-vdADW4im1E2byvXU^H;k3CSxYQ-0X_~^VjPR1iUVq`B2`lOae_xXY9;U
zK^K*Ihs(d+?W;B!)&HM2L`suPbH8h?vc^GPI%DUS3t3UQ74(y)-0t$bonH5DX8?FO
z+?6Jy{6DFv^cuZZXE10CE46y9O0CkVHLBpq&ep2+x*U~8V^HaGD9bG(zb^wA(rGb^
zDA}nm?{wYf(~YnZ>pL8YGF<~%7{Va*LnO+S;*l`Z%P9?mvB953Bn+1?5@h;m<zXQ2
z{{r%N9V}*vWI2yr6u)8$1ptC0^puO(`%W=?LR=!XT4^)FpxYL+IQW(zBnZ@>`ThYQ
CE;A(n
literal 0
HcmV?d00001
diff --git a/5_dpnn/results/gpu_4/valid_acc_4_gpu.pt b/5_dpnn/results/gpu_4/valid_acc_4_gpu.pt
new file mode 100644
index 0000000000000000000000000000000000000000..5154b64954cb70ad066d247e893ad33df42971c3
GIT binary patch
literal 1792
zcma)-PiPZC6vn6j(%Or;NE-!NsEw`Ic9Y#CC=|D$B-JQxQ>>s!nl@{RrZ!DuX~jbh
ziWe0RiU)07y?7`}Jt&BxTNK2LR&RpRf)qtBJ*aqeC-cpaKaF%?v%h@r&G+8!?Ck21
znp#2{8c5k_CXHk&o=PTS@sW|38XHZY?MlS6@y_(=R4C{m$MPdrePV*Ng|(Kud4sUl
z+MS~m)?_E3y%n&djz;W|=cNU>0#H6_Oxb%o$6#L!O3zk_DZRJo*`6Lg3fMnoG3CH7
zC<psYrW~5p7#yCp80g#54Encl;LiP<WrPd~aNJ@L-NZxleAY83LK7<t`d2k5Ef&BS
z0-Rg}m=-`X0ln7%(-NQuC8ALi%FQ8w%6*14C}T2Uk^5Zdp+`JfOn>Y|@+l6Y!>cr6
z-_<C;oTv!>kP#9&=a-<oNCPgIfSCorB?};u-#MY&kpWM6&YLADvnF6I4G=!xdFY;w
z&<#H8?GTiQTzMry;f4b)Clpaj7~ZE5dhNd@z<U$$go7v{sy|;r&I1h~Zun{r%3mJ(
zvjAlo$^ha0h&Fy&$Pr8Vh0mG?4D6JU^O`I5D^Qq^(Wyu38@~0l4qmPygdTduz3n0$
zWqW&gCTomm5=JHzY{%wiTl~Mjsbn^rGHB(C-Ctb}yCZs3uG~a?WtFFl%w%%>OepBt
zcwOF<(nxE6m>WG(ZP%qo<=SqttE_e>eWi6x8kwa1syi3Y+I_^5iBQnJ!B%9y;#Ylv
zfD+iF`c>H@`#g%AAD)2EtEyhFq6EBt5AAX+vfS5ZcSsLCQbn5;Z{EUnYo>vsQRzKg
z>QcA{t|Eue_VZBcB38WArto$)71<2hAB$2OEMKWf;nli}Og^q-vi>^g5r<<p=5DX)
p5H7Yse(<xqiv6EvUl{xDI_jx1M;}Ie(6An<%m1q-wS?xg+<&Z#ZGr#*
literal 0
HcmV?d00001
diff --git a/5_dpnn/results/serial/loss.pt b/5_dpnn/results/serial/loss.pt
new file mode 100644
index 0000000000000000000000000000000000000000..7baa46d18b5d75bf0cbb7f79e4e3775f27498a4e
GIT binary patch
literal 158356
zcmZVHb+{Gf8u#%{HwuD+pbU+mfRw}m20;)B>5_Orq)WPlp(F+A6owQ*LK>-|yF&zl
zVXrm4dN*wLexLc>!~4hko`3Go^}T1UnKdi6=UiUbY1b-S_PDs*x#Rx7{}hNT8aHg@
zm@(D*_Z{1}>ZqZ^ntxd+uE+oTA1Q0EALA;u5VH=;F7T(|7}7%V@QLK%cY!csjY1f@
z#4Ce;lf9C`r~B|>>0bHqgh*xVX0HrCqEZ=LP9aPvBM^ok@Gw+Ysf_C5VaPN_W!M*d
zx~CYb$b?b9i-hlwtAr7Iy)vj3qcZHCK<NKJK5>ODRl<-VUiqP_LS^Jjl`!xV4?pZ<
zRQi1?5e78z(60m!?i2Ryrx3agV}#D5RYIGPMEKt*k<j`sAxq*guZ*kbq1|~Y&V9lx
z@fk8fE2<DS?iC2@moma5K5+2E!ZP7hcZG27iigJQ1eC0W4iU2KuFZpcisMsc!t%c*
zLe;Duw*Behon<nu>Woteb-z#vdF%7(K2OJg1j45sB|^5cG7$G7heDXNTOj<jflvOZ
zxE3oF!ljiA$dYH2O67t@$U3r{f|6y>r;L!&RRFS{WGWDMA*%vpdw!GwSy+8S)-kex
zlJ(cz5~0^724u_qolNDcy&ihi;lX`5W>bN1v$P0g&ECT+<G1tR9&@m*OsHIy5rm*f
zDEdq$WDTi6-0$&>(6lX|?rXO?QOQ#IQ?HykqR`6hEfH39@XFda53}y7w5m2rBGj(s
zLHtE=OXJMx*}YP7yBvpeHC`jSXq@?82R_}icN{Dez8dY7(#-_7G|ueUQUNlKoRz68
zSSGm;<FfxEpk&<1OUPVyM5OX27a_CsTo2V^lDiV)CavJp#hABm_cxy^R9dxG34;fr
zG|v2{vP`AJucC{_nPm->#+e7&17h5YCSLhwf)wW-lX3J7A+vFkhlNRe<F2lE6@kpv
zcbE$?u0B(!G_NfX3J+CXh;ilV$EtDW@@tIBvR+s<&dg||xTSIC$IB%bVqA%N3?)OU
zB~$r!nOA;kuDYdh=CtmFj6%0%w}^3h=wsaYjh>;nrE%uoV*=rDcRt-~J~UFHGV>%O
zR4GTuh;~&`-t1dK$cPjZQ8GS_7l6#7vnAloe+y*7DnTTOR~ez?QH9XpnMkl#G9aUF
z6AyQnh$wGP2?CHY^dCO?6~4J$N};l~kW4sdQ+bm%OQlli7e3vWv#Y7%Ld;TTo8+Q#
z=9Y*GWOQ20r+dtccL<p$?+Pey?iwPH`AZp<N~=+Px@TXLPXIEiT~!EK-%|;nR2K<L
zDylBTxNge@7meTSuZhBM@0SuvM)gbuc(eZpMlkZoK*pv{3Xpy&kx|LBUj*LlYb68e
zEB7lvW}ff(<flk~HI48lyho)nVJjoN`A;D%d?NvG{@BI{<`xmiJny~B=?!ZMC~xvi
zlW8Rj49M`kR)IIZk7UB$&jlcJa8s}J|BFh-fN>I5Z+vq^AmiB`KHb|=cfN{}o@1W?
zy!qz?iAssBeDar*sYwiYllW8uGK6uW3o$NXtc>zz@L&PRH19DW{n-Nr$T+@1qT;JX
zc%4~Xq*87umGtAC6)IC}iNG6QF&TK(<S^mQiC-m@*XADzkUk}^OyzndiBRhd1Jb`*
zD*_q!Us8E9>tiafFKa4R=`$~QSUHXd_p>qJTN!xu<O2nG-D)$VvOPnl(q))RWyU>`
z5c%4JGf<+krLO{{9(k8f{>-lyc|Sz2F0~S<T*@y4=>v*Vc`fAg&@wN>>h)j$^UC$#
zBr0ted6@o%ijqENngYD;*;4{CYQ#}_)p@jtk{+`dq1SM)Tqr0}`DihfSJe}#q&?Wj
zH}1zcX}1ErZrNT2-b^W?QmGRb2njn0>8EBWRB|+ufb`!!qmrIHQbc*(f0zWMZ7R&D
zB$iQt^gkyPUOoAp2ltLeX9$EG4f%AhN9oFd*9*D`z^i-lBH^$u0k6ljU_e@(uT(1i
z3kblgP8Im%&y%*UkO;h9^!8>>DIuf0>OPxRug*RdP+oV=sQ_t%Zc2pX`MlCTPDXiE
z;j#*(rLU8LR~^O(gtQxkv}3IqN?M;}2}r$~O9o!`>&mBlVVnJ4DI5||UJolpNPA_8
zD6h1Z3`qO_icFZFgHQMVG|M3aX)_Nf!0S1N1iZ?AhfjXYYc@%yl65YXSM%pdREBL)
zfYfu%7?3vp1fTp-slTok32U-5!lI>ox~ol!2)w$nP^R+5`wEcu%|j2zKH<T=!s6LQ
z;Pu*mGT|%lw>Q1=D6cH}NWkh<uc-=<ns!bG(#|D$kc#oZpYT=G`|zhOC?!#uTS5TR
zwwI^!y4*4qC3Ts1@o9yx^59;4oooW|YX5o_NGr3|E6q<cl$7_6$b{zKt3cZ8;e7Ik
zCy)J10FthCQh=1aSs7tLT|#QrEh3fpj|Cv@!(I}Qy0D50q_llbC1pz~86~-aDgr4l
zPY{y-HU*Tle)T0F^-_KjNdK$43Zy0ZCE)eOr5=9tZg~3AassX1>nj3jZSPA!>h6v*
zkg|EPK;`}F65+7-$xMCS+AG3DzHx6|pIj;-8dL~RYRZIWF+xh!IVwt8p0N^;y7DOD
zRn^8k@aIW+a7_kMbLUfmw2uEsRKDrLr+ZAr?h2tp839PGv4)UR=N=)g^BNJW<P{PF
zQjU6mN>b+^rjnZVrh-+fT1f(u_jY4c-d<tSq+H%=>L!6!eFhVfSPu~;?eFChkh;5r
zhl<Z-l$2d*3`p)|6H?NCQBhv~I#32)f8L5pT9z~cCGDjs0x5SJ%Rp+|5-MT+L6MMg
zN(R!7zb8<6x|dOTTvGvFed$j3-I@PV1XBKaL`eJLwv3WGyc7dc-akip{d+wHCHdwx
zDk-~vS5eZsRTY5bZ3{&pC7bsVO-h@^s8n_cX(MwfD9OuPi$LmSnE`3z3&}uo-k<=a
z-nyUwDK%^H$*(yr*AAH=wHJYu4K<lp{#Ag~c3r5X7Wqj)Nx2>3lRx~`&ptxx@Qn;5
z`LD+Ukg~g@NcdosOz>3@fTTCGR4Rp{j7r|ARFePlK7Pp?7Ry*A_g^IfDOv-Su&xau
zG0x$^y^O~5sib~8Q9?=Tv!0M#Ww3&h(&Z^3DgL31lKf&AA?0R(2mU-sFXI@HVwa|p
zQhSv`rR6FHBtPpT1F6&d3qaDvj}^k+1|nh48bac?K?x;!$}vLXAKMg^q_hG8ko?0{
z58aDND2am>5t4t8S5T6_FCzmf<uCE+UTVrR6-a#1oB_!jdn;6Ge53+NYe&jJav|?M
zO@8BuD2eYK6M%$wOA}%j8z?AAO{a-Ka?c_Xkof*I1xRY?Q-PENLnR>bU@1a!o*N7$
zrOXsQ`4uK=rU0bYTPgv`Q_3oY^XmkHeVvfBv6zUGGHsy@B>wOXpZs}}`{WaWg!|_t
zAn8<DKKUxvDOsiRXtE3>KiNb`T3278QfVzANf|FtDK&&j^8VtCN;~gkoH*<4XSLEa
zS|u%7Pe?p)P{6?nSK7-!at-eZ6N+>eX|*g!tHgPKc;%fMUOCs6AL8DY11%K7|LS|?
z!_z8C!rV8E&^xaLq?X$*1Bv%4sX)^DIv$3WlTp&X^)4f6?f?ZP@$yavB>(z3Au*$Z
zOr>vEKHbY`Rh<DzJyR9J?o(6}$A7>BKPIW~XoVonWk6z`1AMxx$K_=p@r$k!;ZZI=
z`6~JJd@4!5e9TZ1U#=4g>xu}3i+2bKvE33%^2FgHkT~1>uqB2M^T3}l#R)K4eK<@4
zl0R=jCE=F$@Pwz=`Nlo_)SLn#$7~fy+`UXB+)E@RZ2U$>NuJY<N>X!$kl47Vf>qL=
z-j7w{dzmsyVo{$6B+NfRC8=c}1trlm_~d6#tmyqBC1vxz?@3Dz2w25Nu9OICwupq_
z{}dqM-9wD9=Tky*={GW!%S{y^aX~La>_RUYC1KlP6-cR4MFNt$ouraftsz56I`^Rn
zB)>mg0aDA<Qwa^dUz?<7GX<1{j3$KSg1rc-*>lQqcwJLls6a|`M+HbYQbh!kX7-V(
zsGGgAs=tSipD?TvNBl=f`uy$ZXSerhNgBGARtXb!C~@w&lFn}x2+vxpK;niFpYH16
zmkdZ5eM|uopY-L^eczu{lL+PCQGujN17$)P@AI5;Yblk)8E5#$ebjF?B|?oK7?9Mm
zutKGJnhYeaxk)8q%vU0nm+6G05)Zu9UGJ49k76>8NxENy5&BLcB!6+82lrKsXd?m1
zg&T=f5`zkmIB|+hkhY0HO0^InG51Iw+-ojx%Ru5EvlT+hey?=@iJ>IzAEN+?!2~|t
zlfLS%0x1_)DnMeVw;$|V(`A%|_Mb>V@|G=xB<qBV5{qldfW*vWeDWtu-1DIfB$$IG
zDxX?>x+}Yt03<Xks1PdOlz~|G00R>9uaba-#rsttdM^i`?!yc0V^p4XCPeOhETJTh
z@jeN$^G#JM%eTltQsQ?C5c}q+N;o)CAhel6B}woTqRR`3wEAtP3?yXhtrALm$0RI%
zq@ctm#snaFQ=9}O#0e^p*t8v=?lljH5)v*x6i^cQe{}|8r4A4hvnGisiHE$O)7Xjk
zs3e5{W>_WU8mR)Y?&U>7nR)`@!Zj60IC+Utd2>r3WPgVcy<1tNBC3Swk+KR(VwN6!
z@=Hzl>{}U#ef|}rQazhQ7+;@|m}iQB5*yr&0f{@t^65VO=n#?6Je~p3m2C-0vx-Y7
zvB#A}AVImwfW+E{0L1$36A78#S1(~z4;dxtzf2WK{QK=oF1=VliS;SRCqH}A;@1k5
zGZ7g`I9fsgq9vLzDvL%bROY5gREpN)lRrFW4iu@>x=e_Ey^sgKitIim1JRV_5~1HU
z5lHy#Z<$Kbk%WX2{S}nhgDV1|awS6a)<Xp)I_CzJq-+NWi62)aM8|C>B$O$r#<^!t
zeCIDhZ1F)47cLSaFV~B4xWXsv1t3=UAOjN0)g?qr=8{pOhwn;2!n|)86+0+WsX0QR
zGI_R2C4Lc=$i-|_VnxFetqSMzR@Z+~P@?tj$b=$wz49QJgpx33h)O8>t4R1rBP1>^
z%1~n4x_jkyKLsUW^+i72H|2ZpM<%L&EutjW{Ez_&1I81gi~Gw|Dy&fm;aVyX%M<oW
zb?;X+;r_?8ijIniI3{uWX9AG$u)G39SEMt-MDO9z`rg+w;(w2h2_Ii5(=m;{q!O9F
zTSZCOxJ(AZnS}%(R^ki;qDKlTK+>LZgxJ5UWt5mUflvODlYAisNGRHZ0kO{_e7bkT
z^u8bo8@>@y5<YzU#$8=X$0SVqO28_nuUCPD-6<lK^jVCs{F($L?i|mj`^?|mrxHtB
zETSYP4&{@tq91)uCAPP=hvwx}tYSCzdZ_V2rt)AQpZrk?P4{_aM;!$vHd<AIXrY`6
z5b3jzkZ`Jwgc6(ciUGl!QwY%?o)co?8wIP_*WM|j#cqgHI*ylsNb|-FNO+l7CRCY2
zC9?f8A-4G)0jtQRgEA2Nc!CNf4D>z#v3@E;iJqO#r+X8Q9Pr9N52?gH9wlQHJ~vPx
z6rL#p(F$E;Ahv5JAy{jZh!Va3K9xx3KtfbXR<H^ksVV`{BHm{;5?Dg3$gMsiR*~WN
zWgzL;=M0Et)2YOMttp^HBS#cM(X#>&`_=ojL|R4^lxU?mLL?9qQ6kzYuQbT7QW;fN
z0-_yP2|z;qF(Q>#<N0)N%dX-wkXY3xP+46?0%B!*D^z|hB>)LOmXLtR&ZP{9mYM6J
zk#`xfCf;9}=r=>XRrO3ls8Mq%4ksP!{Q-*RGGvs*`X?EXaNrJ=*hX7KiH+W&0?{lA
zA@Z=3h!R^^LjV#MKjxFaoN&9JsYLhhR8WEi7N|hzYBnm-dU*+v{|)s@#bRol`|#M4
zK@t!hSf5H_{d)pR<iW2J5K8b4j$FA+C9>|6j8$axYlTo|vkD~aJ|_Xue@Y63do4sD
zR?Ja=gyyAWAhfs`A;y@55<NVZPkvz$bCL?g>a-E53|PkqD;lUk@ZvK*-Ge(-BSbF`
z5vjx!0f_a;Plyc55K*GmeJasuDc&l_UKy*%z1<R_%ij!0Xq1IYbV({9mcOTfRb=fS
z3Sk)IlV3){^24q;2L42d+|Ne{cfP|nd=<TtK_yzKp1{iwmH6eC{GH2v9n%-cKy=g~
z2E-1x=hI!CefvSasY@u)Var4y@cRua;o?mcl<02_2$6-?8A{~#SOJJm2&+J><t#q=
zGsjk)CIn9fWh%=DOF-o4SBy|kQVG@H2te%fFd<qgUPXzN>cR-crg`Xgmk@p6-Nnex
zzo>-5mE}11?BQuiB4N`lDxv$gJzO{@VHI6ng#n?3?@$T7>_~`I*(YKZ&Rjr<c8W48
zU6%6c-ny)H1R&Z>kqCdEW<X?LRTT&q8!Q8{ag7xqV)y0Ky+7Ay2|yq?T>=tb^<qHu
zswx2Cc@w>|y^f3$nLW-cC5L$>+fap8X(PSWyuBXYjnOK!=qDl0J!z=lY!!&4J(Ymy
z+}R2cJ|&4j<YaC_^guO<%F-+<VdEBouvSwDtG*z_CjKO$1jo<h(|rqKX9OVd<PRzd
z?T<((k%@1=yH7hRD4|@NR9cl^!3ZbY$UwOB2@wcR{zU+yKUAj@I&)h=i3vY5AUL5A
zpYFReVLp{``aA(8eE5Gd5ZTz90kP>0A+)BRf)dF6y#xfWWiTN0Whp-0C)AEhgvuXz
zrA=`GB^uwH5jt;_fXEt$Pxpqm{euCq0q?6o^h75r(F>^xN~qA=+j}v)fD%3YOa>zT
zpQ%*lJSBuW=8>o@`+-mQ!pt=a5I*~cO7w#wGD^gsMFK)K8Z#i4$54Psi(Z7Vu~MW`
z;u4?y3B%iusz7u?eTA^Al0^7-tylic#i&fT1t8YZd*g!s7E~f_@KeNZLjTG95dO^3
z7u5*iEe}<cnE&l>%gX5@N?^|*S|y$+BA`U-Ugpz%rJ{H1A|)h=%Ka}?AiV0bLS=i1
z5UuQe9m0oS3RndTZjgb%cUAawUunKG3J_hpo)GS>^59--XgC8x&vq+>TT=uea_pd2
z)~EBp4-Tof7!a*-Pyhm*OG{M7mt;Vs<pP1w;gL$1JAwhu@Zb4#UtpPi3J|){j7qe@
zJcbfW+a>{#A|@g5Xci&1$Y!)Uyj=mp>zm6!P|G6#k#+6)bWhRwa|VRVUloDqz7Bl4
zt8K+4DtkIBK)Aao5FRfOf#}ZaDiFL}h!E<e$tdBcQ~2c16F%CP5U@r`D3O_;h(KuE
zUMd0O36((6+*BgT+o(7vzgObi=Ls11R3Loe6GEizDg`C<I#B|m$LH|ro^))2KxO1>
zg-~o80}|%OL?FC<gbaj)=?sW|a$6#d$|V5dyX`%EU06g3<?gHk5i824d#={iBp`Ia
z`-ux>I#g`cCu0>oww3|m2SGmhqnxTWB_P`B?PoQgte^y06%mMh__Iv-_7MZ3pHCHl
zSp0ggoR}%21TQxcfKb6d`E*ZlzNrXA%B09Zw9qrJjIF7lgvv}}fV0^9I)n-ik*LhB
zssQ2QABjNZzx)yqj6csOKSd<_VHF5{*-Zk%E8dj}=Lai5wB{{>{r3liNHy<y!fkr-
zL-_N=#Ci+}*Po>T!6A+L<g4hi0SXX&U=u>6Y7ip-mSb2szaJBU@Z$~&5E?sLpz`wr
z1_auUl&EB%PYCt>PDKgTo=ZsBwNye0JnW$Wp&4xjAl&{NDv>VU3k>dW!?21R(g}e9
zj|D2z&MAOh>4Xe8J8K9)z#gRn!F6Q_fwlbwl+Z};dLjj0$S8rJ_v;sQLR5nNUNEd8
zb-gzwykL@q5?UA)fpBaKA=vt%j1rl5Qw2gz41v&jg9HT5c;CwKO7Ghptki(f>fd4_
zVCC$h0>KWqWgsxbrV_aSEtSystGx2(4=fI!jejff$=|`i{`Lyth$#~G4ps@za!?7q
zQ<(>TaB%r+D&eBJJX~$(VU5b-++&=(DwWW}52ysX<x#M*1Itw))SwiV(DBs*igV*b
z83=uuT>@f*hZCGl=?o=&EjOS16yZwMB|_D9RDxgR6;MJSx0Hd%${8vUExDKhp|WSF
zgv<XUp+p-k<I}z7V{=p>@}!vvgxY(bmawvcp#(p;Dg%KPrKm*urSZTY9tu45P^gQH
z;w;FJfN;Eb>%wQl0!pZCDHRCzc>6IfUY^lv^eR6269zx@5E&3B(rRf9Myp0YQE|Rp
zBBMmAyhm^z7Z9jy+n^9u_f~<>%)4I6J4QwcUq8zyKUZ{!_r`_Rbf#74vr{Tok^A0{
zXXJf{p#)csQ3&sOpUhB?Rw_zp)VqB0Ck$VGPXZ$IG!Y1u9xDUUmQ57Eo^sX0%~m2x
z#P^W|1eSaMxDpxSy&PwT_ohUeJZ3m1mhUp3?zx`KkbuaEegc($yf-EAVOy0}bqyH^
zt(_wPPV#>Y2(Njf5^4zo;4C;tCD`g9LkU&tLWl%T3MkIqwhEy|9!8icQi*<WOF*$I
z4wHds=@dTsrP?#246r&n3J^TgkYKM~sG@|@1uB90hXs_NQC<a{s|OfhSxFLLo-fZQ
zf96o5nj+wI-6{ZfgDX^w8xsi5)}s<uVM`ML=lE*{2m}XunAlCAvdgam#)upYu-ENV
zfWYFM0uVfLoZyrj!cd%g-DM#7gHCW>zNex@icMyKb7`;)1X@KzAarMs3OIT4@X23J
zuv#t&2sHRj1e^=r3k+2JiwE~+{?v>R80P&H1kM~{SVdMX769i|4G+6hMHIWqMhS3+
z4p4z$_<KU2fcHtT>W&ex3f$>M2)FdUJ5KhPh*hxZL_Xa+pSvytBGazRKwzsb0@mX=
z1~><{D^$t_3C@IH8A|ZSe0;hu=QfjoXs*3fVjX-6O5o@|35Z-iNyRyy$f!L0hDx|p
zG9fS~Ki{|)_UB`faQPn<2q&IlfSu!00dTGq=hHp7PG=DaUb{}ky7rNR5=dw%0#3cB
zeDZ^x0<)>uua2oGMzOd5WL3@kv;-eCq?MiWE*%`4;@w`Oo%c&(P7N|VJajrq0JKv@
zs5mp0@!;N;E{&-KHScH8`Jjq|Rj^Qi;FSALL<yWtr4p|8w~S(y%u5Iiui#;42Zoi=
zqoM=^PVN;5x4%+>@cc)N%I2+9tZ*9<#hf)!2DDO1RGjasD=2~UQ~BiA9N2kK27)!t
z66_{N1(Znr;UW;MxJsh3?kfRswmqj3Ue;Acv5y_5VpTgu2#(C9#<{16oK9kZebO%j
z&L@RMAnfELILo#uD3P-tszBiBUlL%SSx*R_&gx;*4v|)MGzNq#{wD#UL(c>tu<$pp
z<Y-96sr-UU;Nfr<=U$i<7^ec}sWt+Yp!W_spWNkv9}{djkyerJizO60J&OVad)y#|
z_e_;g0`fx@aO%CL61=ckK?x12O(k@B92IlrUKuNULK>C81Meo-JL)s69O=2YD!GeF
zaL)z-tKi}*G7y+|or?462P%rQXEmSh3mkrf0rs)&5)f!SO9Z0*@5+GPD!)MG{p^JB
zjAS18!^5XqQ?UnY3X1tzdl9g{|Bw*b*i%9Yuc@T~rmwsJShH42KqR$@3fL19`Q*<W
zxY&#lDs@r;vu#a2`6`<IkwnFLOvRC!izwFY=Q5zzA1wf8!}=0nT+K}gwrwe+*wsg<
zfYZkNE(e(RqZj;BVOZ%8Hc50$fhz=SXe)*iDzjYy0-f7Zv0r<iOugSjfmZbfh=6s)
z=F`35DJg_-t<wsMnR_Y0&Uc5QIFDDzK%`-7D)y~JD#0P%mr>7n`#IdVO2EOPuG#6}
z;DM<kiaq&^3IvW{V}Oxg69DU@Ya$_Q3YFljl`@KxBB_AAxS$A_L%-vbU%Ywkx&nmu
z994mE%M^m$xV(ts>}Wy=X!|7;d(;02!PI*MXKNM(D{IPk0${(`&!`m1sRGXF%Y3?b
zY+^Ho(9Wk4D)bToGrfdDIQ5JH*4m98o=xJxefUZ5_6FanA)@Ft^2k8ohWBB!{%b8!
z8T(QJBH5P;fcee4A`mQfkl>tRGKzi1$ESOW5_JSZt$b7h$Kx4F;95_DGo_%6Vix|H
zN^s?`gvjanB38DuUjVG0ha^B-(p{#q(Z>ki-;w~mgI@#!3qpLlm(k0+8^Mn%$|zy0
zq(rN6@2Y_P?^p#eiiskX%zsrX&sHjcV_afDp!Oju`rcbS@Kf04E)fXM?yLaj(D?#j
zby%(f!76L`bg!rB1rZ4M8%ZUw=)Qmwsr5b~+$c>(aZaa5fE}C9fN=dcG7wn&Py(D@
zxm3U&|40Fx*AqSXeI6S0;~V$v(T@rC&PpPR)BPV6Fn-%20M^4#8I{2eWguXBzW`R1
zJSvJadjhSTA<ZNdGh9&toZb@{;Z2B&UctMEp**hztiq9qhs@S0mB%Bf=+9?KDCQ4g
z21JI`;FDi`AW&EUtO73>5USc$0qo!*6$pGdLI#X6w?!a$?Hr%{;m+Vq3ZYwb83;zI
z5<>MJ2`Ems!wj&yFXhuc>4Xz15L`N10z#FlQ3<Cl5>U*f$`WAp8Y>c(zm@?zPGx|T
z`wJBa))}t=p|oTXFfv}MfODq>mB5eX6cn@T9|B<CuBZ}z>_iA<X(yr>*)Ga}Ri?ZM
zn1^RefN?)BA@IbzLNlqgfR&X+6A9}lD1d*%6@t<G0T2AO+wZ(m0c+)l0^nqRE(7Mu
zqAC!7YBZnxU@bjP0IY6F3@}#|p<;e~k6^D|L9kj(kmB4&h3tMRV4iuZ0D-~7_~a||
zixnbZ$v-HBzK0lKj#wxI`jW>I5SUY(0d|Kggizt2gko+kz$bq~{g3+u`^+p6C0P76
z6<<tZD50GO6*GH#871=YjRI)%n=l}_DGQ(c39UiBB_L4gun3q%>hQ@|_SYK}Do<Wf
zF$*WCDAt>S3K03wBG|o8$td>fhZ11Ut;vAUu{H#2{xJo`{M)-`bNLP`_OBBdR@U(Z
znQ*q002r-p1_UA*gy4jAJh-=aT}%b+{r?Jp79P$3bHV`$(0?DtCqLL2_O(Ez-3FP8
z@I(R3Nq?(=RW*&^|9YK_VphmPCDcDfK{0Z8?~c<tQA7z0&;`KBxFZAhop-6&_xg%d
zeq6+WV1Xul^6POP_f!BQw2z9hc@?8F<b(j2_s>!Z?!2y`7_*rKn87>*WB2zmiuKz2
zm9T5&5>bNf$4P)$?;jbk5|;{qeZ~7>bdHu{C{EWO_~aKByz-d}nEl>}Kwzj!2oCMQ
zsGKV!5gM&j0cY=*1fxVwhGJAMBLL3o&tyVt?;{#K{3$~*O-lemldn^;^Y&J$bjZ&q
zzn;Lt_Jm+^nuKCM|5*ltnI}ZR>Qz?&^l*I@2%PCI0><Yd83;b|zHy=Hl^Ke8@*9a(
z=clSv^tB4$#D61G>E2r;%owf$&d-e)5Gr|8q*8n+73)=q2Y$`wdy)VcHz!Mg)!=~)
z*jX-!fN5-_V$^*~#s2G}ij}c_gaSAps#L7x=Q4_Y^m9J>DfHdR1nZMI48?x*Cm~p?
zql^+b_D}-s8QD}oKaiJC{%~tTM*%Q{3q`<bScpnsa(xNK3cXMWUoW8&=*9#TJAH}>
z82_6p1G;!q1w!Hw0SHdYA_0M_v-#vF4c&+_!0e#PfHUig1lXx#1uARnP|?3T%>#dw
zld?+&?0t(AK(F|eU{CC$qD1mdCfI!k^5EW|m3?KvZoWhXjP)}aA=-~%7VSu|e%LEw
zr9F>R0KMxPg7(Ru0+ns^RD#}(iaGQK!Rq7vbuhL~WH`qE=Kz_o@w^0BdYTG2A2B}p
z75b}}l>v*LR{&#AkO5|nkyOkh^?Be2o1yM1VBGE@09tYu8F2FE6alm6OaU;Oey33R
z>A4E%*Yk>iBjlF><9bO3=ppYj=Ip4VpadSzmT7gly;u5=RH>|fBLJrUlmW)O$r9nq
zxl{s^LL!R!byu&9Ss|eq<K~M%phZNcvg{fmw9mU9``x-!?05c_aEyI7LnYk%jZc0z
zf}dVvKp@{{1+eZ<qhcnn7EtU<8GQ1C?dcU{zzKMty1*~@Xcf3HnPFu&ktM)*`_IO9
zr6)3#TtCsuX>e6RvFdqWE35fK5yi|=i%<7$8ogNMaQ{sit@{rmnsH{HLYVMO0<_w_
z7-04~M@9eXV;=YuSSQm6zNufzD8`PV0$?6FBm!3djSR3qj!`jISX7MvT^4B7teFJp
z3Gb+Yb?8e*#rj(YwB!vU5cGL4djCc%yF+ch;n$-t8%EF`j+amZ<DXEmUwFqj$uAgI
zHrp%#jvST&Bl%B-P+^S<m|ND#fR&J);8bp-qBuQ|5bU2G5RAV^i&zETTS!HFGKFA$
zzFWb{e)%62y}=}g61db$0_<j!1;F^bh)kGMgAuw;;nV$eRJ|er)~rV=pl@nKu(w^3
zP|PFq`E=jjuDL}Zx~-cGI6L}CfYq}upZu8M*-sR}San$i^#7_WK)~;P?wsD$WfZg8
zRSB?mon?T2x;+)+zuy!T`|&}Ma66^~futUM@^gjl@-kozSf&C-&yNJa&R#$Q!kx;B
zK%j9!1{nVQ0ucHrlS=UDODekYuN3D#v$MM_!LB(#MKSLFOvUN;DHSL42P*c+J}l0C
zlwImS59)b>X=ccA?#lX~_bmwAUc>`0p?Pl<z`UHBigACYgknvZtO91;H3|?MHbw@F
zIXfl5{<}97eQRkEMQi^%pYFxC`j+7DDiX|Rp9x(2^IWb3!D#hRLNUtzECWve4GNVJ
zx2RavA}WgYY6BJh+e|9@VO7K`5Vt@A^h*EZlb<3`@S;FvPJNk5tHOM`XRq?E3OL1|
zNPvIFa0M{p`iOwOr2+%Yas6b#IG4^Rf4KHvF%{5nexLwG<R(Gea95zRc%usFUyY%n
zCA?NpjK>@J<j-SY?M1~uG#f$t>#B^EIZRgpM`+9d>;3)$p?zfu(0X*BVsvTE1Ajs*
zJzWLNpJvE_e(Gxm7=hmjTI9TdV*YOepYGj=+))U3QYFC2{agjKTow3qU-CFz0<671
z34l4`o=BzXdn%RqX9TNQ6@oru1VK9)RN~y{@mG3#ib>4{6n$=W2G|2qRVu4b@aevs
zni&Lrxu0OJO_H%P+x;a0!Lau$VKghKpxEE<qLnt~Ybr+GJ1SP@@DQzx;jI{oRX<$;
z%)7P>__Zw}p~V=29SQT`-l!bEDS+{E5(6whBRIQy6SP*ph_os#DTHCq33kIQ0*W1W
zWWw>^L_lxwE1&!n&c4+$5ZHA_0L-l>pL}Ih>m~!vmstqm$&*zSGny&_#-dF!U>$oV
z08TdVkH2{&AFZ^9M-;4#9w8Ahciol<ZxU1>Q16XGSp1FvIKAVk*ma*UDv8MiyU95L
z#ros43g|uaDnRhyV=4iy3_~%3dnCY`TTBL=R?SsF-|{1$?knv-RRF9x-XAi3>r{rK
zeYRf(;t$s$Xba}@z#nA}Op^&s`Z7SbGbJiZuM1Sv2o<})EfvL%#fgAcwvhr@8+9s1
zs5TX&+FS`M?bd$;D?W#SVyCxYfPV740+<czh=3iNCIe2syDAWF-ccm%yg<e5UQtFd
zf@>APN#7^|*1Ac2x_4us<CPwjsW{=+0#@d$vohcuouB~Lpgk&J{C<>A{&0K3#~#=&
z6(tbYTmqb2YeXs^H<kgrK~F~I{b~xJ*M33>)auOxf5On4?-*cSeEUggzEMF5j$0xD
z)_{wI!1F#Ll^y#OAQ+iXaDE@n1AjuRYgq}<&NgCHTJ=-`d-En4@R#c$0miOADxppb
zKHZ0J&r7ghmX=VAKl2gH^H~Z0ar0EHv_;zmz@IjeV8pLcP|UHH84ws^2!KD(l8SZd
zxk%-Mx-ww)u1(Ok`9u`0)fow}*8HIYW}zZ7;Jf>eS01EED0<Gy0-)_Eq!Rv&qvCtI
zL`Jbc+baQj`=KHbKl-LZWkP=$&=%ZO0rPw}3GmlF=9T^4M<PC4*jpX@n;*jOn9<t%
z7@IxEQPCf5SFqCOww3@RX{t<R!UYA;hQ@h#D9cogLR5@Nr3tz-PQ^->nFttv)Fx=d
zbPv5=%UJ1iW>7KG|02A6nwM7Qi@$_8T<VU|D&Q1-E(87ryA(i+pF%}1GMb9Mr<8z|
z_02CL;M;gmq4I4r88AD(D*{G`aT4HnekB;ICo>dt`qv7f;SCw^8y`u8Q?(diow_3c
z<~8r9$sV*_L<v@Dssh&MlO@0`^N$GlAI%l0RQOl{jPm70K<l@YV5E2-QRnbfuRP3D
z>6lxF1nBvWDS&Y?#;5z9c1sX|K*^#KU=BLQr@PwI!z)c{2`J${ofW|RZ7HAp7<1<!
zg4N=bf@0kCNr2uZBmjZ-mk2sr$pe3su7+g5fAx|G*y-NSg+02if?|~R6YR)s8AZRd
zR3IE4D+2bZ$r54sNEI-S=Mn%bG>l+Bc%KJ;(!kKo1bfDE1;tl>qYMNJ50C)=!EPd;
z7g#O>+K?a>W9<zd+>0;STOdT5D}ebwp9t7PmlDi`OA3l{x&f8IfQ=HBCdmR2|ILRA
zVEs^qpg*f5plCI3Nq{k+y9zj!yswp|3=~l;X*i$!dP0l(5zOz3t0;QXPXu%7B8H+J
z$;l@_SPOh00s5RM!R}K|L<ux$q5yi_34$>rw}fK#%1Onld`qPw)RzEb*DaCCoy#hr
z|5*{RQ}6M~Pip0_CKD?47XZ8A3o3T~Pel~RS4{x49yj^qk8&=(6ai!V9D&g0p#<n+
zS3dbMfoIPZz_?V(L%A;`lwkRaA`oJm1wg-jNCkYI<7L1a`3VDpIYv-15-vz6A#u6_
z_!nmI$<L*C$twfemZt>c%v^?IS5A}wf9DfajOo32;E(dZnk4~R?WGLxO?|8YW@S|Z
zv=*&Yz}KZapZrnUb4v!SXV0k^|DIzg`q!okSQp-x0i*Cu0nqagQ2~FUnIaHs^^XFW
zXD>;BQ!Rl3@zKu|z-Yf+2DGV12xgJ*6%_xDWD(FimzStiKQ04yaqlO^IGQA&_?E8a
z)4fq&f2&eC+Fk<ud3=2G)ysQ(7@!X;Edll~jby;=QeFVGIupFoV7Q24+?c5XddpM>
znE7G?m9kZ2z{pjPPkstp8l_M<mLUTAb%Eet_ewzVx2?_q>*PxnczLv`0>pncL?&e2
zOYq(MLPF7t&K3awn%*+tE3%ki_wFH}7~gjz*kiXVD1rAz^Xa~)6K_&6`dI>%io5va
zEA#2kRE#AtnM#pE5@5uB5dnRBTMt>^{&AVGmv7u>ZsW*+UF4<+I5~b-0bdQ9V07)R
zpjdUs^2s0NwE12I?9}ceU|#xH0IWTYRX`tpMxxUGAfN7~Ui5B@Ijx$4qTe_x5_Vq@
z0P}h=88BuJQm7mcQ1PeaU?`418=w3M{XNPlfN{qA{MbWY)5`37LB&eD{g44pS4#o>
zt-etKtM&^jM)Q0kikWAw1Q>IEXMpwbIu$V6l&7NIU!kDri*-KvWf*NYFhE<nSq1E^
zf69cabwxmrtPlXR^G1a*%e&#W+<~EJIgat^zOLi{69MyFH36`vZ6Ii8e^yYmg83NW
zzu!|JBu%1X^{hxG{`z!+(e*bu4$pkwyW!5>-+AEY($_5(03*J(1ek@2i-5kNiAqTM
zR{*Si2l#Yfa?MQyJ$pNX`Q*BUl{G$227(`)765(WWP(}ltb$_uCrN<w@a-|#3X3S_
z?8*Y+RlEW?>yI)(%jLaI!JMb4m`~614Zj}kgWeLL&#A}&<KLwOJ7JfKqF?DL6O`iu
zpxsauz*tkAU>=kN6#M#4D*8XudElq8zS}MW+SVEZ;I#7IQ+;?j9^9L`q7AL$=dBP?
ze3=I%z;AXY=#94$;zyqlu`*_!69E3FUj?+8y+puR*?{2hUY`ek3g7RNS4Ip_Q1pp^
z5VWI)gyNrZL<WqB#Rb5(c_9PhFYo5ly<@BY(lMsK(`_a=5uTin`SNcc<3}Fkw
zzvYNRW!$eapgDI1DifPgF}tiK7&{uNSm{Tei-1w4s|0AyA(haeDxdE4JojFgzB`XV
zW$hCcuyZcslOJQ{xxfH_$~Y15T|6ZNX3|C#@ZTED0KG%JOep-l1eimgh*ZwZWK?Pu
zQ>e80AD`}J{G<_#JqKheVeb>J|8-kLF*NTA&EQ{D{J;Idu(F@|`Q*>zyKqUOQeYy%
zD7{8O@u%-67;ST^DEjhFe7a|^`i%(qR@RiMd^(>2{>-%^p^2&h`t}7X@N!Q#f}Y3w
zK>E%-QLr*k_K*Po^x2H?n|H4Gqo2}B+k2B%R>^G&9kaZW2<U7+pYDaN{Zs*rOaI7#
z{`YJVaEx0Lm1ird_%>hRfj`PWvkwE}7bmNL@Aw2NTBF}Y6#bK;65#8fEC5EobQ#c>
zyyDY6`=WycD{do0v6?Ls0N?q0RE#ybL=-!-zW@XlH7A&T?(*P1PoS;{823vs!qMtf
z^xg$|;K#>QG)&mjx1F+Dl}>+0j4%{KrT0$qzR3#;Jg|WiS>0x`qNu{NWUWZoZOG
z^t#D>@<$m@6BwWg-${VJa2O%*K|>YAf3&UuINkhIjHiVO`sD->E8}5%8Sn>N@X4Rq
zdZ#r5jAmP?IG1vYDE8bs1hYi4gksfNtpJAFSOvnDe-x?gf5QNO`F@gz(FC*eLJzkl
zi&&Y-&8UQ%4pLFHDU}q!EO%W7^cQb`@VgxqQ378)7YJ?l@af)`5jj*qo7G7K{6~68
zfYEz}446wMsDS;jivk#vIx)b?Qb7Qm_J0u!eUXS_EHxECyZVnxrTGymM#@4e`pM@K
zR`w3>ABhd)5gntCtKhB5d+(Q3AWAEHrB8@+FWxEsFBP*%ei6mrY={h)T@O()PQPL(
zTJg8v<!QaC7+*h?u!`^LeG9a#FBBA?bxx(#>`qkl-iLVLC-v=36oL2xTdBmi%p#%a
zC0B|-{I2(@_~%`eQT&BZ34mSmh)BiEp#bKk@l^Dos*IxPOC-RzH@^VrrLT#A|3E!H
z-OJc-%Ybp&D`uX%3QGK-Bu1-uV=~~Yl0-1xyT+)DEh7QOy61fIC)6*OqGE4oucG*~
zW+;GOv<<;*K3zgFbJXRNKgtNaprY;hN<r~$)cE8p|D3ieVA>w6@AJqgzGo9@r4Lvw
zpqMjXDuAy|Z3Y-QeJUaKk_ec={R*HfzcIiWep~|l1K#74pHv&<{YIH(PpBwn>1PU|
zX8}I>F?MfD1{}frZ0MG(qG&k>GQep4p+Y621VI<$BoymfBR<{dsdZKajN&I0!2I_M
z8Ss6+kO6k##eDLE_4e^zsXkIf@lUQP0D8?re7eV!KPLm);txc?pEQ60`s`o4GWLL1
zSS>1g{52)ceYo-a1P`}-48=KKSpv*G6GcF;Z7P71bG=GsY=#6_zvQ7}l<33*e?tG8
zGXf!$Uj&Q_=Lt^hodSxvZ37ki)HNB!e$tXp{&4&4|C(ww@V>(OM}~xzv1*k{C^^N$
z=>{J5Y-U(#g*p>_B{~Zz=D(?Yx;N@fSONUeT#QPb_u=<%SgE3zS#J>x@uG;L@4qGi
zfqIEDU_9?80QNVN8K8fbNJYP!p9l9QD61qu7rrH!lXfyHt*gm^twi~BFKpsH52HU)
zQLLJkMZn0qP6qsg`!d456C&U{K3@jx?^{Y#n#T!%Qy`v-`Egx>p1%jdKmIQv4kxX;
zPz3B*+Y})DV=WmlOK#wkUzmBUHbFZaQBaI>9U0+Cnh3-PH%ow3a)1K(bChI&nHb@d
zKaZ7lC&AyTnuKEi_@N4zlX@{g*V>7Iul_m#ur7OFBL4zGMln~s{q=kIhJ<2#!T5CV
zVfEau__;=XE&%4=L4v*Xx{9Lhnav2te-;7b?WG#}V3k&rYJ14LP)70Vmqfsc8z=$#
zR~HDzrms{KdvRYG@b8K7>0VFsyCR@9_*(*A9-Sux{yBXZ;M@LM1hh%pRlr{<$^hR-
zoe1iM+Y*X@epP|6U>uca`W+QT|Er$>yr{BQ0(_OfQvv__t9<h7@jvM#0`VuFNq{-D
zy#Ux*tBb(Pa99R>1ApO@Kgws|ncaU=#y_%$h-R1DM@#MW5CP@o8c_uNe~e*(k@$}S
zs9ln%XbpcP_<wj$!phv#UjX#?8mNG8-@6iEZ~2L!56)CjtgF3bKr`}D36xzdq8M5J
z69E65b_6@Mn~Y+8drkzjnJEP0{k{Y<`c%cr+VY42!BLY0z!|cPPyXtStCdwkxk?hC
zH!s5{UumPJQHc*b1bt^y87r$q3xcoKLj}b*WFQq|dTtR#Yk!ST_j$T>k^v*59~FH;
zQwhcF)1Hdf`kaDdeUzpGKC^-Zm{uPV@P|t(fU$Ed6>VKv6-9fRRRYYi?HJ$>7gPX!
z*>M>#vfLK{D_&8k`09v&IjV{bST&j|gz?!JU~K$Z0L%*=2>QQURTRspNYMJ85K)|(
zov2tDIe6f&T|aPFB=|<kK;T7Q23SAaRGjLEC6w?HRROejs|$c}rh*KZ2lpy~wWFH^
zXr*rP>7L?p7Lm%;OETb0SttO;<k}1{i@j6;{ry!kp<Z4UFy|I$Kw!r-5pZU2Q2^`l
za23#3Kav3ZyFV4cxYAJtoKvqwK)bm?0?b=OWWon;e|_(DrDAt2A!4QPOP2tv{CR@;
z^Peh;Gt|cbW8*{t@LlRfu-@q<qG%=FA@~ZHU?^sVWdyzX|6~+>&Q=DP7Yj>-0+&?4
zSNyC37@ZHufcEEBKKadzA9$Vt#>qx1Vd_dg-TVCCY_Al5!ccsVy^A+4HI=BeDI^2d
zyB(=$?Z=BKX8rFKK#R0tfN%0#8Hm5MK>)Oe-w}M@?og=gJSzaf;>{GmSdhvmzaHnG
zWCld0{-6Nb?y`LHl~(dk23Y?!Bp9J7DvG}6R|T-`tqjo0nSAm`nSYroVAPl(0DAf)
zi7>J*730oo8O8olWq|(6w<=(1^AsuvzL5z7OB0NDj;JW6K8#O(F8}(QRMh9|M3lh&
zM+E<_2(7gIhXkzrB_An-i~0EE4>xknV^qvS0-*l7oS>z8zt7t0FQ~+qS)t$<z34gy
z`2X*Jce$`zq>^``0_cs-5VSv2WfcF<mn6WLb>G8~_f!=B1Mjbc(Y>R9qNQ#Vf%v<Q
zS9UC7D87byW#Hv!&k1_o%M8VLx`hJx3;rko#xM675FhVdkG|63!M$}~dJi{5Q$~5Y
z^alnQ+HWeLA0AC5{^}?RCBE7xRP=u9sA%)vqGK*vnR#Y;t6T%cINay-RebW3+7*{c
zfWCLE3K%PH%7FjQSphJQdf!SjbDo6ad;AZr{DW$XC@)KHQUG(%Z58mJyGTV}(Uu2(
zE^We3GGLVWO$6dwB}#zx!F?I<)vwC{y?h%12-F#^0$SZ`9)@^-LA0DY-?$%}&q@)j
ztRE{>Hdhn@zcHOpeoTD5Z49t-H<SszHYk99$bS+Lzp%SN<=;gjptb8D5q?ZzfIf1k
z2>7ymB>?`W4HYUIW=cT(z=0~D-Rvd+#>4_t{FU~|DC+e+4Dc=cTmj6lOG$wK_xpsG
zjd$|EFJAw*g94bVUQ&rab6!Q!)ygv9-+73N@l#ktF;bRMF@GE)p_tzPVPaz^PjM>#
z0omyo?dTQ-#~4@M6$zU*%7D430KtE(j*8+>^bw2#CA?La4;ZbMynPkLmP;slQg1%p
zx9Q<g8SouT5~&QYtN`%?&Pafs|7Q`<{$0i=f0X(7J{4=5LGbsASFqA|dB6O+)I~t?
zpL&-8FP~lElRw=4!=z#rw^S5=$DtyiOS`D}rf=rKz0Xq;8DLNME>*iVUqtccE~o-8
z&z56=*{G*PIJumP|Mm_6MPCx2q8~peqZp;zh=4V@2A}TT_$?O$w0SQTz}L<DxzGmu
zL@UF5$~W%8`>Ke5@6Rt)z+Y8X0B74zG9l+)4<jxSg5P~7Vr8!XgNmm9qo9QPtQP=(
zzvg_puf1d;5peR~qGH`$EurXJatMHFSu&OL4-~+-GKo+AaC5u&*@$00N=1oJy(I%$
z=6Mm&JLXXUtMu0bV6<K)1O77IB|y(ol#2e1ccb*QC$us)bYVEg9?^~v-+z!yrPyf^
z@Kq?u04v^62qlLQwCbl+6tiGZ0<0z_2wMIc0*e2b_p#AZCo?K5z29D=+#EVa{Vq+$
zO27471>)P)pkj`C=;4p03@cyx5(MpHc3Qm*wI^u1OA-7ZR*~Y|dl+BhQx)*ntHLKQ
zzDw5`pw${L1Nwqpe7f(~$A=Wc)oChVUp*iIzRACffR;9!U`*_#QYrs}Pxs+Y5h~`H
zcfInx_qsH_xQt_*gr)@lzHAJ|D98lBeDH$cJF-eev1{b!lRvXnG(^Qe>L)68pKrZY
zX`ONLv%l)B0A{)>0NUEKj7q_cRLrZj6%=1+3ZMK@TJ0NDtj=3S6!YQ(74Xd|CsWz9
zPNZ`BM-|W>6;J@<PF+TL`)|+LMZ6#L(86jGR@R_&5zzm-Cj;RQ2~^C(3kceWs}-!Y
z_^<hNUvl3H3ShE4e7Y}i!xRyC*=`WQcjyj7@pl?V#VA>npq<$wV`Y}z%O`)hKV`89
z7>DkuKzx=H4DeS!$|pa@H{&Y-FlXG90nHdI0>;PQZP5#R|2XIG<9!|UqPrL!v%3Ht
zY^`c7q4=bN0$_JPC<0cN12Ujr^D)4PHk1hUZ>qq{cC`gSADkisKK};_;J@!@K>Yd=
zBH{E78PKLj1j68=4A5tcmjPp0PKn@@A?R(!2`Em!*(zYy=*cI)3EGx*0$@%qCIOB<
zLjVG;Pb+}gts<ZNVB4G~0Q%5<9%`Q8fv?Qp2TOnzSu0X$@QRB6%q9gTuzZ&Ygndo<
z<c|sr&Y}W#)|&#L&uGY}`-HVFs)X;?NPsb^un6b_6Zv$nr_&Ozl$)udXy*qDgjUr>
zKz~+^ijlWILosW$Cs^4`6-6&`gNiR!ML;o9-+m4^jTcd@s)rb$WqTq6_JScQVd`QA
z7<0d%VxCzoQQ6v`PkyO-?3MsnP5+YtL(Z-Ox^ae2evCG}oC26~PzHQwFET*eeP01g
zDOmuF?$f9kwX#T5j-`9$kGKDDlIGny?dyUfj&bsBmH@r&d=-fQPgbZDdoB|G@4xxp
zdn6UB<46H3W2K*pUgR3VzhW4{Oxmr);Z-adCIkLi69vHNa*|+9nZSd4H)emS0*37u
z0CV#yKKaU$OEbW~?Cme)vV#(eHgq$swEs3y2~YlBz$!jtx3}8#CBf+$uVSTrT~!AB
z8%ha)S?;w0_ztNI@c(g52DHLg6e@=vi$FjqN<}*{Q$jJfe<>3_`b7owilZdJe}4m?
z?yDHzj8VDQlZsh3Q9-e{PLcpU;Qe$PaqAh1ul+0qh?i!nRP?%x%8JEQ;;R%Asg$WF
z0oE?>wwQN56sR2BL~t%-5mB^P_awj>Ie?13dIv)>zk4bH{+`|kAh4(f5ByaaC5Fp{
zO7AOx>9+*Hf907BXft+7RBn3T%J_mzMKRYrXMi*6eJV!5_Y@RuXge8*l=w;o{AJ@r
zz&Pq%n7+OdLou5^PzlA_i-40io}f1}2zIkka-4h3!6OS8;3PJn;><fvuqPCum35|$
z8s{FYcTAv_Q+d3E66o89;1{Mb6yxedKKa4U*Y9}c^S7Vyx%UOE{Ga$0z{vSZ0)ozN
z0nitQMZhn7M2P=22M_M;eg3TsSnm%a=#8HejPF)4th8-+CBPi&z3+jXW2jgOUlD>M
zD)B@36WZ$nDiEK!L;#H5gZXq<MKmhbp6Lw5x;um5yV+Mj@n0TJ2z+)+rm}km!OZus
zgz}Olsel%}Ap(K?cN8kdJ%ZV}qKsl59IOIX_b>V6m!TzGkO93-Z-KD02O}8X>(az7
zJn)0fIfe>2AB<ptndQFhVTc0wLjMx9+U*1sEo}?~%%jt(IQQ-<DB7h;65w=yC;;)L
zR?2`LpO*pFgFXUajQEUTj#{LmXxlOwpzp3O1IF{SDiF+8k^xrf1qxvI$WQ@)s5n7?
z^o4|C=4?a7AGcpb(K~#l0LHmJGGKnWS)`J~`$KOQ*{GuEJ61@*|KaJr<77JD|BtWe
zga{E`juJ!{qL-rw(R;KLU6h0fKI&0cbP-)7j^1mC&S7oeo!)15_uevnn;sH;?|EOx
z_pj^ucwP6*oT<LQUlO03go}jmsp$;GzgZ*%y6;p_0@gwa$eb{dkacB^h?4Vij!gKs
zKmdGIwo~z*8m^#tlSUK#$`uJE+uunAvNEnKfN!Vv7Wm43B%%bn-=|gJWul6bJ$4_J
zEMK&Ql9}E|1bnw!^U2?C|HFndkiEQy2xPhUD?s>@JPF9om_rCSugEAl^XgE^n%PlA
z$+{pgz<XhW00jHx$w1)Y4l2IFZahRDRDVtad^gsKgl_#9kn_3<pZrYryw4RNFlRpl
zysb7$K+eElMZmYI78RfOgn;5*a#;a#+6)m1Pqgdty_iD9EB`>q_HGn#Ci~~veDVhk
zb$`YH-|{jtmA0EC!t7Eakliju0(|D?gg{o7jFN3QAOe{!OR9wMLIKF0-(8{-x18Yn
z^dAu=`1Gy-@c*V+20~k^Q3*tM;elURaE+<}0n1<s@VC6nCtn5H%Q6tGRgnR{qT~7G
zD{t-kB9Q&OuK<K^4&;-seE0h^Am`sNsrX%qDvEbqG9%2=-eqt4T?Hk(mPp9i@=Qd@
z`7w@4uuyyB0?P`hgxb86amL^IB?E$&x(Ps_!~+Eg6@ROhyS)inOU|lT1%FSL0H4W2
z#d~*wg5q`bkbz*UsVa~=@~uMU;~G?Q${ToyJkvHWWFRMNi2wxUri`G=<TLUb9xJ5+
zA!iDea8vDfHMqC8NUJIH1;969JtHKIB^2cC5mCa?s!Hh7iO<MAug?$xuX7dy^2U58
z0HL~05y)CKnUIqnMextLM#zu$NKtqen@?)12AgCQ|Fv6u@`niZ)ZWVAxh=Hv`3@>r
zWi7fO0O5~|F(7BKc7?tc-9?m8$#V)2II@jTeusI}0}RM+Tvh?RT^dmF)}ADxWbQhn
z0zN|>1_b<Zgv{xc1Qh?7GX&pyK}88pJHmi$|0@OX|M`$gxJPFyfnkSKT75f%PkufA
zWA|ks_t!2el^>EdH0mv(1jZH>fSioKsDv+OQ}H#^K6U<X`xKn<kIkSnfzjIi_#B<+
zaHyaNohkTHD?Y#Wb_ZYXpjFU0UW~$pov6-$0Nz`EsX5!)QptNboe*C1jf^wd%p(AK
zU84zJ$2x`*TD(dI{6%XBKz9BpiAu^fM(C$qn7{1+mCC0-i9qJ8gM9Kw4-eLE9%wv-
zp=5PV5P<A{_awl7s{x;pSM}>UG7v7`Um(P2Uvpot7!@Tj`7FUd`L>Lbqi@ZC;FwAh
zq1Aa62rLK)fUm`8RI<M{ODLh|PZ{8wYvVI=sblYmR4yHqfIz7@8SrOzB?PN(=7Ap$
zEPqbLKk+9CC3pP?d`6yyeysxdl%6sWsIow%GGYavk;ht=LdAb$jD%8H<6kO)ZmEQv
z33UjecIUKJN+}^Ka`Vs+Yh@td?4<zSyU(cP&e^JA<YE!4(A)9~koWa;LTJN73B~`w
z%Yfkb?M1+Q;Ghin`#S|HpEwxc`{g7Rf0b2Ka^lv>QFzLOL_%Kw?+D@APXw&;PhMnz
ze@`rx+};Z%6tBG`A#ms;1tmP_{f|u40RhEZy(g^#Ne>DB_y0Qw*@ag`oXOFv5|G_t
zwgLnOmf@4XP`;8ss6gmuJOhIMnF^47f1V2Dj>)HzQ@<}2Z$=*xt6<3<0-@h$3<!L;
ziQw-(TtErQ{UpHGb1I*aH@9Cc1qke4Bm%w_qxs}3f9v`x5L%m0B{1fmKxO0&k+9Xo
z0DrS1BH%sVTL%0!CJ}reo={P;SB<3-o|hn@1kC@5fVa&*j7mW|pZw7S8xP5V_oRmq
zY<rTS1U@(>0lA$WeDcFNe|{?fzNiT@5SXLgPk2LnD*o7WB39w8uOuM!W~l=Bx9pdx
z#AqL5pEazlrVJ6V3M|xKg#7ja5hZ8x5Sg%BF9G44;tJqhaaaU=oj36rc{(!}F@kBS
z2n3^c@EN&ytM>KGYWSOkl6~nr5%9;@WFYwQj}nmgZ(j!ZqVrWCyzZt9<edIq1iXt@
zNI=&8rYexTFUV))u{ND!fcHEeEAns31Y#Rf3+-JZVCiSM1mBG55=zdtbVji4kO5!r
z4u$Z;90~B>Zmt5sxiKP@Ds`!3zStq4gzK)OlHL1XMrD#jCDd-2f)X5mm=I{3!B9d+
zD)GrL-=Cm;76P4h0!rT95;Cm@6jFJ-?2(9)_j8&I_!oRl@KwnaQG#8P6u|ex0Y;^U
zo6pEAbYPqec*oro0H4dO0{+~;B_O+c8A8^=MluTjr+*5NU2mEQ1dG*XKww`k!MpJ(
z5BwnlJ10s&{@1frz_)Z8mB5ZX9{3sWif#fBxVcCLa&#{Vfg_I@N}%e`1pm?M0*d!b
z?Q<9W{Gx)AePXo$_$!!1fHi$fCD^^3jFK%6Q~>{(RU!}=T}B{u(^1K}6vG3*u#ow>
z1O)sqL?CbFR)Y7WNC?(w%{P%}5qnR=i~SNx;A9S!>{f#WDyN(>5bBqq0Kw0Ci9jIk
z8I`OPt0WZPFR3CB3<e2>4<@K6{-NU)!g?>2++*7JAh@YLtwIy~33SG~nooX(!Nb}$
z`>$+NQ9@sKlmXw~*Mz`&w}=w-Rpyi5D$IVN5~!F@$ZL0zR^G?;s07pdD?IUs_%DS@
z;K~~+{*V0<R=M>a@yXBlcSaM!msZOtIXjk1Kz3Y+O5x~=0!q%NLlO{b`bq?RjZ3IN
z;JXt75Inv{27J#~N>s{4QwhFk$56bX<qF_SWQ5R%Ux+BVm9)>Q|DV%Tat=I`u?k&y
zDFNYh?JE;r_(DYqgrY>iKj#4<u>U$Cr_+BDR=F*YG9ZvxRV3V;LnX9T`@`0s85Xe0
zjb6hie|-XXu82Ue^-KnY^Zyfo(B*Cl5V%}l1;RN!_~f_BTX|Ckf@fw6Kz`ya1@Kor
zMF_`wC6xSwc_QGC%cl|;+(SVLAMi537Zzn8cmFRE5Zp3E1bhq2GAcEm%0QsQH3<k!
z`H;`Z^%UQz0=ZwG5rM#p=?n;79w-5YWd`uc59fEcFd)!%xeSCp8z%yJ+1hIuoLz!e
z`LACvtb)R3J|mBHb2b&f)IdTBRazng`R`_nK;URl0DQl#PzkeU$v~jZa0X;fEh7TH
z9>WzNyh(d=pAIf2p#&<u6aoKt+Iw2?NK|R{=Qu+CsKGq&3kyH~P6Tq@j|t(~GgTD-
zS6?z9X!u43ax%9OLhtTKDEZ?(gy8%C>@Ba(W&x|fTdxS@Z@VV}-d?{cK&aL}6)608
zH6h$wQcyzcb`X3wCh)+oFmTH(0{$=N$$+ou&nn>U<CK8x&UP8dwJi_;?~~Um5b8F7
z5G;9<2Y!d4ZkrV#+<pm_oT9}E!T7QS|MS0PF8;=KHBs?Ke?=wm=CD?-*{B5HeMrbQ
zv|xP4U)8*p1plRxv<g<6u3^S$B?`~0<xmMI7_o>?ek*Tg8!FjpMFbQUDk1{jXDt=L
z-_=SbyJ;pNusWaMEB=xn;rA2hx|)#HXQhmiUtyC11mnB&$<Gw5`$+)&eQroVpxRPC
zBR5xy$$)RELnYiu<dd)b2O2WKd%iTmzf7-CIn#<t_*F3;_+g)YIU#pktcViYHb((+
zq%<o0@5f*$p}r3UfVFZ<fd2#SK?6T|Y2_=CEMpZc|AgT0r+vSIUlvgD=DP%(3H05s
z0HF#`Rlqk%5j5m8ApEQLCGwjKRg_?#bOi{V*Gm4!Ur`Cqn;>Bo*nLk0!kJb+`7;ek
zzleap=4}}$%;_osp`vd^Akd{9qcSsBCRlqDf=aZ25~$fkCCm!QK<KLzgn%)bhsawn
zB2EPI?1!jiUuFV|ukJg7f6Cu7mGTD|Vdi292#<X(Qz^Ap0Rru2Fu<$tq5?VXo6A5T
zegc(%DM$$3kwjV*Z!H1-a=$5rmgNcI6Ke&OKvO;<*Bt!LF9Lx}9R(oNZIVPNr7|F}
zc%T6In)jEe*jw=#dCHEi3<&-3i3kL1JSF%Ko@Xe5iw9)D*XaFQncSD5_y_NlfS|Fd
z0(hHz%z!}s4kF;+cS<YCITA|Fom3GBKBzAMVe@MS1ez~Wfl%!NJ|j<X?-(k9W-cne
zS=x)4y`%&c|Ac79#lK%?0tygxos<Ay%qbZNtocbH92_A5!KO<!%-Eq)dAgVZ{{R0+
zMoJL@B~MqAPyP_$yl*5R_w-sB2)@(4ApQ*twH339Soy}E;4|_pYF|<b|4fz%Eq4(L
zSC>;z{B;_NKyHCg0>Uk|AK36WE2$LTsHtET$eBtdtm~_yJgJn;fWX)pd`2!czNiF*
zo&*SamD(sM!4pGeAaH%R0OY0)R)FxG&o%VZ5qt-hs8NwS4E4Az0=^IS%Ru1OaR!9y
z{z!N{bPx~xaNzfq0>J;z3<l&ZE0lroM|}jqd$+bic>at5{+eHjfbV8Kg-ZVvLcsNh
zjFNLYSpxj?HwZwW^EiS(dYeq8eq}x*FT%363J~honE{0ZTq=;)y{-&AS)l#e1Ri`Z
zpcEJfDL`<mTLi+>J{AZi^?dSs&Rv>IrC{y*cX`J>87uDq69a<VZ3+-R{QgU1?5?7O
zPH8W1zONpmGxh!;glbfmP;x9!sRY{YA@~|Mpq207J$@weI(Mrg1G!BW3qX##p8??>
z@d^-_`yHQ=YmPdh0-+8gWWe9^FrR$on|FZ$!Sx=2uy?XV2=@>H|LL8CaA})_lDn?C
z0z63?qXL17^$0mRXL;cF>~HiZ1M+4p7J-0n3Bli0`;iG(>mgy~f4+{-$n}_d$v~)c
zyaWWkO{d~}aZ)SyEDWpMZ^J53*hP|o@T?9J5IR+Y5T1BfK?!y{#DKuVp)wHOGF}CO
z_xyzXLElR#h3761!UZ!Gl)z=D3OruXUjz#4cqAY+ZJ!K;uPz`Ire0G~^8f$;C+YZ6
z0VVkIA3pgrEgYC21EDO33KYz)BU0HojSya!C7}3@+Z95)<qXKRlqUqwS5Q#$?w=BY
z{9+v?AkgqC0}9IKQVDG}t0=zsDKg;y@f!umYr2q6Tv$K_yd`T%KyJ4UGT?uts8lu#
z5rObR?M=zmzmZW2T2@j4-{U!a@;eNVEG7cMImZaO9~>a$cXX>*<qa7{$ep*4R>3Dj
z`6lv_m|R{0{4Wd)$U9e807AFAsX+c?g9P~29Fc*-nggihT{ns-!G-mx<TgrRD50@C
z3Awvp5W;~BzTpoVoN!(Q@>7E<ke8S*0HG&U`Q&GE>zz^w;SUH;yEc+g3ahu_lb;C{
z&Q*c%i?)Q^zYa=NPR(LKLAy!<kT)oeO8#ewJn)<QJ@pl!@U>n93JR8LrRmpH0-t>*
zU{(0#av8|m*n<I2#x_!b!d?$Vpx{AB07A|CDnMR^nKBU8HKbD5{sxu&R$eNh^sm(@
zJRMgP5%BNQK4XPTPB4_*Hy%FuHHQm(QVIPzj!LNEOBpNgn_DUn>fcfVe5XqYKsa`X
z3giymOvQJ6tAtWeWR?hoR&1gYe%z5t_=hesR^dKp6(HbmK_#!G_T~oryk=Mh&E090
zzdJ@kDQuH20Kt2cMIcmV8=sL^euVbXDwxxNp#&>#kbuzCqAHc?l^K=RIU<$$gQ<AW
zYhMt*Tt>jk|LP{Ka<{IPQNmq~gq*DQv<iRJRlv&kbUm#CUr(SCsIpJRD&WvDAhbw(
zf&tU_JVZWhccK-*pP9{o+%rRDAaB|vDq;UoD!B{(rV_a45ThbD&+qz30CN9W%z(h8
zo+6dT`(+?JMEe}(=RK0BtiLG$`F$s<z~eSqg#Z6<XhK(hldvjmJ%NxLZZD$*o-L*l
z`0GOvh5ud20+72`p%QpjPDSwt%vADHRw^iYP3Oo!ct9~e`LlRDbrd1|>lGCxe`jeC
z2oHM9XXKeKaxuX7Vz)}=db9+D%dHlG{JSG1!1uI`3Iw~46M(#`RjBy?xXA;*!*FU1
zD!HR4FqHhMKM-;+G$i;ttq`Mdg}?tN1L1ln2qDi4hLU?KO)LH1Q7O1qQN*g??+X%;
zo83<VLK|9BDHytj2YwlOa~jA%c#?J*;nIIGl!9V|1R(FgeH94w+bIH%zpSVLg%j^E
zpuiKuC%?mj3wISDH>(JhV3~dbO5x|vsDy{#R#2XF`AGr_ZyQwL@x<9OP>_CF1oDEd
zRN!%~rZQo8dlB%z{YnMGEtV-jZr^l~%GG3<;8?~dzp%%%+Y$=@%vDkHVsm65ykI$>
z{7hlYAPLBio+ndD)V_MTzr5mspULg=4WaP#P8B6@+i1e$XKe(O{BH*G$qzs29ae!*
zs}mwn@S5=%S<NaU0)<;Y7Jz~#jF5Xndr!kZh9s=QO^V7uUeU&U@|zco(*9&DXjEK9
z32&DKT6Js3fG53gDS*FtF%c-dzl{M;E6r1YC;8=Npzy>)KKVW8zZoV`*{*#(^9J8y
zRF=i7z~h5oQ+Yf-Bv5I&j1boMR#Ea}Cdxo~)p#m-af_$~2Y*N<H0etg6?xFGbV&hn
z&m|MWZ4dJhd8W=!84#$X{Q~6IERa!h)2gU|@6ct5O3RB1kXJN;0paVFCBXM#2a)jp
zpNaWL$vp6f2sEf5622KI13v#(eDam={lBm38&_GNlCYWq1*N7_$@!-o!QXBTtpYhE
z_>ssx$Nen>-cD<&cnb?u6kmRS8OTv(g1_h;5yk5XD?q5U_NnvDh^JM~^l1`SzMXrh
zWDU%r;*CEmVC7#rUj>4*!y=G<-b@I*d`R$qcU6gsTw&IbPpGiMmkcG85~l#Z%`f=m
zXZ+R^5)k-k5+UnmN14iZ4OAef`KJ;PEWd=1{ma)13VReI0zP9?KKUKyl#~U)J9(}Q
z_?G;j0GSup2tbx@9KrihhK!Q!oJa7@dw=GqcM4YCv2H@n_9Ou%$MZr4LWi~y0)IA@
zP`vJs8IakiP@(doCZGJmvfpTb66M4fiYPfpKU9FM;2w!kp#7xyS}+wQ=U^iSWd5%s
zm8>I&6cpc%`wYlzBC9~ggZToGeQ-II>{_#U;E(RDyHW%K%{me?r<fQ@wznb`pY=2k
z{BV|8P=M@n`>AAG6GRmDJdXicwO&a;PKQ-8kokNRAtP`=L181zgv{VK0*WtcFaxrV
z4pxAi4DBt*8f~OiW^@%9tL*D%6d-fNAU-3nVM13GNdKXo0A$vD$tPc>`@LGJ(?~+e
zES171U-`zACuAQVE}&#(Hc)`fuj{Ko_PgUE;lUmTWL`B$1ZNiwy7fHpJ7ixJYMAha
zjFP=yyVUGt?S0S8a7b9?tf?*nK39DPF#8B9nR)LdDswaVj6A_j9aJFmw|9iB2C+&M
zR!0*AAmh>J5|#N|_>4UFFNaadD!o}oVfA;5KxY5OeDX7yuh$T~j`1Q&&cKiYFlC7V
zWGuNU0a>a4h(PA1BLcwg9pjVVPtK}+49My`T>-M|bRc9dsH~!7n!gu;oZI&lAkA`!
z3ftO`O4f+>1oqi8H7at?S=E+HKw2YF1Tt&=rUGn!JpstNwweLiP4fBVx5_>}NCA95
zO_hPnVfPr|Y59)=uxB5rgtQ8L@>^x(E?_`r6RQklr;MVKF=>v3lJU~ZCqL{d5+zgl
zx*h}U`2)3bf46|bIvyf;Dp?dN7axj1`ss@VkN0;T_#LLJvj}X|TLC4f;}QwT9`iAu
z{0w_MRYTeb5|#A#ABht8RFupiPD1+B30m2=RK&^^qupWFv?{d9YVs+qGP64p(k`s!
zXZd5fqyHjgd+tjpnF~${K>DqrBH)O=K!w%b%uw8E%@rX1{eQ;9c6X6cJRg@Ju<I2V
z3S0PyiYt_(pk#mbMgr37+Ne0czsLi>o~)(ARUoIvIR-eVd{4+01_@Nu77~y?B8o~@
zyBUOxUz_m_zn`>UatT@c$1{}l!e<g-FS1_-GT+qTGxCC8Jfr|=Z?6$D7iqtl-p)0s
zWYt+A;f!bR6&YZkY^0KvV`EfCO`*ajnW%Wgxddm;-)ZI8HHMI8IU`3!9zA>SL;+wI
zicn$0M2X79IVzCVcNl@?Op{SOpSMtew6~H(cv^(d$kQn^oRD6$m5P#?TvP$riv1Gd
z$wEH);dJq-3}ml)E&y!Cbwb*cu_{XX<jNwz249z{<YzEK`NJZRUG=m8q#e&t31dVJ
z&XE#I)}BWqkUnz~12S_KN<dmq{4hrTO__}9e=2C6!_5fk3zo<ztj+%jnIEoWDBilu
z2rgGxMPW^T5`mmEJr%&~c_#s^LOm+>^*dCQ^peE{Amd<LLYiLt12D5jibAWE<@k)e
zRNITHK-PKf4qa<!NGNReS{d-{aA?JIQbEbS&{F`|M@AX&{JDhB$Q^!Ji%MGEtqdh&
zW*t8HDzlsRJDRrU6A8t&<Xb}4(!nZ9W?B*z_lIc=B|UhEinDfc6(#M8pNhLeO#vli
z;x|;%uiTbV*s#h9!8K7U-Yf~lUAvPCcyyCxAfxOO0btiIQ*qDDQ&2KiXfI~gi7_Hd
z_TnrRNE`46fz>G|qqu+BtN?6}K?1V+bQb{E%AiUxFg_!%MpT)I_-A9Z_Fk~_2{KBW
zy&Io=<(Y7n0rup665#Ani_gf-4;BhQ+UZJEvcwk(mF%Ag8Tac+R6f6`0-h#j0Z8Ay
zP6D$3_l5y!PNxF6Z`V}`$=wAYYsmlx*vod50e0%50%SCrA_8tPR{&U*b`p^G<Q)St
zUUa6C{pki7#Z|Sl3fS6oWI(zrk5B%*+`bzU;FzqvUm3clB8v0(whUk+wEJ<-KTj*$
zZ=Vv<#=lT;CS%Mc1@H_i&nLg<^iCaAfEBk(K*szxGL;o&R3QCQ8%CvSV?Oz<Sn(|)
z;i#wr_A2YO^21n$Vq3ml1zcrY5!}&zBoxQg2@G)eGZV6N>Z>T8=J9;;he)eDNd#Q~
zJf`Bx4G}W_3^A<IK3^{Z>}|3PxRwqR0DGQAr1E#P0yqpWMIi0eatUx(`#~nu(cX7g
z>JeJm&mK^*%82<$0Bl$KQE>>f7)s^`cND<&wu=g6Pg@}XX^#~VaNO!81L?;r5ZrSY
zDJa%S??0I<wQr?+yC`Ah9`l2CMqVSLr2Vm51+w;j&H#2Hj^G*{5K)}#w2zHvi=Bta
z+w?;h2DqOuQvk>2^D2<(8BcJw+s#l+|5XzKd%tB=vL=+3QQY@WsDNitZ3Rf**O&pO
zW3dE}IbT6ZYrR<oJdS5n?4m<J$(%QuPkv#ZPqr%n+nX!^c2jqPvs_6LCEeax1{{t0
zt3aB0oeZSsyH&!5t^(lRvY3kV>|KH@ud;-dr}hOIaIn29l|P0GfP2Pk5lF99o6pEA
zc5OEm_p8NJGFTmgXYyYPSKn?f<7@_`wcn-!?ul<jz*<|B0neNVj9|MX0$Il=sepUQ
zUIk!xD^tlh*+(lA;t1KleZ`MN9_#(zC!aOXp`v(7wUB|dcpIPmOh(dk32^r5r4Sxy
z-)?u!|H&xnE7lWS>%Sp*h7Ds_xeQJPa8~Fc1DR%l0qMt<5Zs#&swmEny$ay?u%rs4
zKNvu8Uuwpv>>5ag-TRl|if^i5W!uw^0iHq6M8MTo`_=NOUkfN{z0+h`_1Q+q$iAXb
z(T@~?v{mb<I2uirQS24JCAj9)W+=|m^+dq^(Rm4Q884~;Yutf~yU$dDBd~#Q_={=2
zxtEIb<XI|N1=`PPcJs;%XV`<!L?E+RNda(mh^3M?Ct5`@-r6q%&YWR<@_Y8&IV%B{
zy%$8l)i7BBY-`8!$<L%e)7}(o|Nn_7cJ@Y~RkPiE@-vwe%1D5t(Ju;MJ$y=}^6UV?
z)A&b*;+|ZSisj)71;x?!HwL)&6&C<|W(OH?ycx{^YrDr1;I7|Y1u`dUub87}Hw7ha
z{09=?8MBd!<7sb(V(;`y0@8Z@A^@(R;~C&yvX<byq<z%gX?IktZ14YWcDLyVI^+J}
z3g7Uj?5yGu0c%6I3V23pKSZ9+`veqcg@6cH?oUwxcg~kGke2taLS^(#2Dt9mPzn9)
z0^lk7kc#`#WQJmG>7-&Gc~d}fm+c}08DCvxfO*PyBH+2MU9)qVk5=xUvm~rC2G&sl
zTj%Knk9>$x@vW5s`%&##xGEN-k~X|E!z%56D<#1G(f<@G4Za}Qc7zBXrz&B^e)AAK
zWvcSPUpo8Zh8mXaqLp*=Qvs{=_x~?2_l9q&IKKTt#LAi#lmYuQ?UlFZzLikii=t^|
z3or%6c`}WPtKDuYmKnbhZ2wMYQMim7I)dxjr?hg8>>*&qI!#gm^S{?*z<GZQ73cjA
z6qJk`w?)9QU^^AZ&klx?W@{h-*30kT)5ld*6#I=N1xWXvXMi(W`|5dSI%t)V`4`_r
zzE)HIPyw@P6TvoOv54ZVT1f$1T{@_MJ@YofIj2BCNjubs0k&&BMIgP&3l(sBn^3Wp
z{ZgQ^!7Kyr#uo{h-i-vu=D~c!uP1%x2nBHVSV6^^ut}n_wI2gACXN&U$DVb3@>{vi
zl~Mp_=kH}8tEEE)($22slb>-HDW;WjTdCOEoffciTD~JVcCTV6j&pZpz<qLu3Z&hR
z69DV*R4S%BJ_*H{{9FXm>S|Z$N~|HESeiEBlRvtr)MF8FwW>kI{Y#3B;+)x51u|CV
z6FjYQ1QgG*oqY0}J0|R;V!88$gyMen9|IgmqXi)Sl`K(Nwv|d+zuj7SJDHI2zbk4~
z<PI&L*$D1$dJxhI7b{q$pN%JYx?K@aJlRV{z+ESrV6VNLklv}YfR*!7tOB?;)Di&u
z`+rB<QM!VJ;+b8U0cKrwD(;WIC)nElOe=fGQ&fy>j~IpLKJ}&wq)nHo*r%71P|UL?
zP)X|<OR%*rl(9;y5u}nnc#F0=++M-TRe1`Z{6(<Oi6J;Iju25CjXo#1ih3m!S8%rq
zn9q#jli$iU>H-yOwR{D|vH3cme8n1F764b8Lk8@1e$q;R?d4^C$I{A_@tue>&Lr(u
z%l7a459I9}I+Hf76`ipU$>L}ELzvy#ht0Zl1{K$}ED0-DNwW+%g_3;o!;WRM6u|Dj
z&j9O@_A=mjHJf0Ko+hEBf3=WcIyr!c$jv`@66{Sz%P96L?;q<~zJy}_=?7Z5JZD6d
zjIpn&*vwxkD7GHjFPr6m!)WF1aDZSbdql+<YqM+-u#eRfSc$$2#q4`U#Z~DF!TG?W
zVP6qBDsrh7ogx9IDkn7DSt6s@b$^S1wZ*3rm0cAW;FL#+fNAesDwdnE5{l=7_BL6k
zwNg+l+vkaZYtMQ<`GY!RtW<1{;ta)jU=tPh?y(Art;--4@Hkg8z}%_^!TPqfg5p^_
zM+RIw`>KRdWf|bSR8j=2@BbQt%oWRJ6wjVV3gD{h<C9;HXT<{sq*cll0Na+{RO}P~
zlThp*1^DELP3apLkd{|MBGms$1zg|%O2yv#nO1hSAvj*=XsekIG<@`*5*4{a`?=;+
zOu;20il=E2g5!BJ0mZFfsS<W|V1VsVLxIYe88TqMv4a7Q;(J9ZN3N=ba@t#Ij+Ye_
z^DhthjJy|y-U^}B2nJZ9&j?gL`;TDSS%;ydsmG{f*cK}&9wm`aem~X^XA6Kcx~mGf
zZ)Gch<K-}cU@|cvZU2`FVD{bPli%EN>oLK);HXUH!p|z;nl*`vZPsvtqs&AFE1j;c
z1X#~#uUPuOI|LN_l=2L49<|7TZAl#kus1x-Cx1|PgO-di;<60bZw!<GN7pV4u-z3k
zWVDv4l-wu+u6NqUFHN5*pg7q<KKVVHA8n8TTb0!^VEw8n!QHTlO6C3kWQJ!(14iYS
z@qF@|yP{VMfHnJP8E`ax%?O7KRIJuF0*Yx@69uq+wpJqestSO$gQ5VAxrbH2vTZN}
zY*S7MfH~j7XXKi<+!6uX>wYR={PvatSO!IjfV0eB0$?2;kN{K5n+$OGzNP}szkZYe
zQ}aXta1Cq40B5adGLW`mAHlNI#RGqI>m!Q*IJR$*0at_3D&X#@edb-?)}WQ;-a7#+
zN9g_c{qR_ZVsYP>X?1Rq2$+u<`Q#5_lTHah+KT5g;F|I|71!aTB8qWYErRu4o(7{&
z!OA+$$pG`#T3UH|UO=%==}W~{Wu$~+D<%to<5DRVFvs>00n?&x3YFs8Z;NfoXN*>#
zSVUU=utNnb!-mO#<E}#hj17Vc;G7ms#kS5Rqqv6DWq`G74Fzzj`}yS0-LX*n7P$9p
z6=}6fdw0_R%aZBLpvMAWZ)&CD{^_ntrLKnow(Hj=z*4uqLKtV(Q2IlH_3I@JE8E4Y
z+A3#)h+^GbP6jLkrb|>dUZG;vpP^zanZU5Jgf0qzd)^-eC!49F*xn6Q0B5PE46ry`
z^U0s7z4t?bQ1q+<I6s(6#n{x!P%KlT`Q(Sqb@r%$p=Vi%O6{TwU@2OX0j{*FGGJf7
zMFG+mzo3#)HJ^&(s&*@L$94Qj<eoj(XyyD?`{vrJc_pmu-A)RC$=zKAEaPr4z{<4O
z(2;zbR<?~3Wvq-PE>mIO3@11i^ki6>jx>^ijHA~?z_`pq#rD%5R2;{1BwB@I6~OWT
z-}gC%B8qiQD#5m5xrWu+*VAblAmNN-u%BQNE-5J10c{vyOBqP8=H8G|oIkdZ0Q>M}
z3Sb+(o&oOcmnx86Yo-V|XCF`i$D&a(V0|*10ZxBQ1u&O*qym;{l?c}Lg*-&w_hoAs
zVEm+-0=VjGFM@f=5+3*&YY**b*>$r8qw;SFmEaC5fNB3GDt6s2D%Js08CI5~+DFSV
zWD~8-_5xa&bNw=%Y0ydl(%=6T9=1!z8H%IGN?IA8{ZB-3&sr!0mR_|OVBJ`figkZ)
zf_?BP5i4i&Oa*Y)n@eyN>&*jyUdFt50WjUEqyU!L+UL+({E39(Xj`7o$P0CRkU-_m
z*DBzylR@x|`btJI?te_sPoK$9Y`c1ifYti`r(o<V6~*#kkpP&RZ>M6~W{^>A542CF
z`77<~U~RXW&e(@7<wyASxE_DO0JE;S0yuMO^BGxXOi}^M@VgRVbGrHDD|2vw2pHq)
z5$qrTA)z=I9$|nvdz%8-UvH-399UjNF}`dm1EzBZ2G~0W1i)71bD2<YhXk1P2~@08
zit!M+u(D4G#{7*8#a!l|2w4AJrxK2P2=+D~2q;E<T?H^5>P4_e4^>eT{;n(m&YgTj
z9>QAgq5?RtjgSCG%aS5sYNjI?2mYs`7{A}E0M;5?1;E_@Q-ZBpih|<a{7waQ@!AvA
z?GPmt^YsP-;o%+zIR2ijP<f{P#F-a-s#2L>j8?W-y@=wd|NeD0455`}!!5qy&%!?O
zqy%`XGL=fDPh~*AU<Vazxsd{jZD9uuPnydp_8%J&%$Bbu6i2i`FqhsTqBsxRsHCRN
zS5Q(L78L<Ql~1VXuh|K%Z-aajd3oc169Grf1uB&-7qlWT6;NzFtMSPX+qOHk(x5j(
zafL_AfO%AX39$Tql23lv^qnFD#<qhseD_vDG0Z%r0;arJDz?QvWfYrnxJanbR|QPh
z9}*0nD1xihd68EA21tOlr}hx`mwJX``ThNi`K3uju{G4o1m6Qb`4yVy7FPh<>{T-0
zX+K8*9FMOvz&7^;74uA=gkqhzK?H0av}bC`+RIRkxg)ie_gewQ)~U1#*wg-_V(Gk#
z;8^oB-|)+Dult`0q+EO@1NM8r6Wn(vizt>qS4x1vSd)sa`b{cnugA(*S%oYWu&ml7
z0;WdO6)Mlm%YZ$-A{9^hg(8Z%TN#4kLY|6Z2;@qD?PL>z<yC@!VlDC|158Kj$yEB6
z5CMy!ltd_ci2;U&#Y7<e;RXdTO*ukHtrf#iY|*g-;jaxED!3IYgPs%2;}=RO_UCUH
zV3{?JPyW(b`kx@!+O|?qj3!9}9OW*{RObIIQfbkIQMp=F1<XaO5Ny+r@WAiT)kXV5
zN_QYAqgX!Pr2^(VOaKfQS~J2`k&5o?ngqwm{UTQ8geQFRJG38~q5`Q4-x3VxXYdfY
z_%YiE&UH-`Du;GZv6LM|#cF${V&$CEg^IcHEeXXK>dGg-pM;9qdtp+YDvBlaQUPp_
zv`>QVSC52}dZnWXnEq=_h>O`Sqgd9qlK@-%G%Dt<UudulC)ivY`4Rr0sXr<zVEI)0
zgqvE&i72}1yBT0p4pFh*Y_6afTgLLqZ|>^SOa)AzJZFIY>?DG@<nJPi^YSYhu#|nu
zC%=_BKcE1{c|S>j<;qbRuv?djfH{4+z!CX3WlW=1s%S>Rs{oG0|A~NMxl0A?vz!uO
zJhz;RrCmiH_@kLhuA<_&_Lxyw7ON0S?2`a}7d@Z+u%)W@XNWbukBDNZuYF!@?Y|IE
zOebXpuoY}!fOBkE0?d^Lh=AocHx<*e|0yW;xRNqp`*@cKnEXFVfOTX=2G~9v&1d8?
zKB%q&sh1Wq!1&HC0_KHPCBW9_Fcr(Q8Y+tU2Oq)m$uovx89!PC49AueEIoS>j4!l5
zOw83fNH}B3tH1#3OgEqWGK~GShtLUCsMuR;KL)mze+W2ZejX&)XIzz0oT)i{MxKSf
zs|XnWt}6q!hgT#j3#u!CC8?zfSf_Or0QbqueDa%HbAMC-OTVfDV7n6~0>%m2N6Grl
zARhQ(eZdDZkk)7|1MFR|5-iplJn%E-&Vo$nHj-c+(?CLTynFxcY5SM9+P;KuBF|#^
ze+pnbdsPC=az7C;{7{u(fB#qQn1?MEu(G~u$p~lj2<CTs6~)wlfdF{EejozIL)92y
zt>BjkYBMUvmdgpY6WZ^Db3;9q&dj_j0EYK}MVR%5g$I7k?psp{wv(l06ys-)sF>uB
z7>c=9CxM_06#+}5bsBoj;ep?eP4KFO)!!?CuHa_@uuN@E#r|P89{6FyAREEDKFm<e
z`|nXPPQNaqm{xR>0q3whDu&}zRTNw0S`5%9O%wq8=7lO?+_shAaCa6_(uQ5&lRv1b
z$3`l4R+ORW?|TKn^lqw#v-K4!wHv8G>fhG{!1_3!iep2Df@043Q3h=D4yu5$?Jo*o
zYE@DKbQ78hfc5W6GN8}>jRB^8H5DLbTs)ur3XPv%QVB1=Cm0@$Qcz5j`=~%_ZAk&l
zU)2){&9#>$b<!Uaisfi`J|k~R@~>13Wh)6N&ic_JV0&{z1`LHp1+Y|qsRHKl6C}bT
z?YZk?#waMsp-cuiijENgL+S__(7*pcn)kl{I_&y{R+iuAs5oP6sJ-BpzAt%*Jmoo~
zv{mA5Dv6&kkygh_3xK}*Ek5~SyZlxL4D;@)fa8<ReDam0$36it|2b9#k~(jcsr-~9
z0J`VLM8N1fED&OM6HKe`Nhp>fA5$@e`w@({9uq7|Hj9cpmf^Lk0Q$j239e!{WfV*0
z=^|ji@}CNr?)|0!)}N{gK+3KaDxka9OayFE)dj$G!!HBL@8(i5m)=W7e_$WO%GR^B
z3Ygw%e}Wha=ZYxy(d8t-T=Ka9818(-0Nc14RP3L1mr*=d0tCaA?JA0HN4^4>{onD)
zpQ&lMA_4l_+Apo8_zDrlc(tJdm}d`?skF#b0sGJ~5@7xLEfve)vJ6FkaU`Goo~^b2
zQUKe*O)_Ap-drX8y_^B5W3KQSxr{RZD1g~>Ss<j}Wq|#(cJZc>-w7z@zf3Y<3k?wg
zQ=>mrz;OB@1I*Xw^U3eo__&x1m~U$Dm-Xpm1;yC;KLId3tS<sdU5m<8ewv^XMwAl)
z<F5MxVEAkZ157Qf3gDPKl}hTV`zlK6wKNH^?p`4SrvD9RfZ=r)Dux#ic;HuPE7+|9
z=EnCFKzDYa1Xw;(MZmDdr4WMKRKWI!_ThJS=*0uSAH)A_4A9*>pa9m1-%&9S-^K$!
zW8Sr20*o;)RKPTKfdIsB|49VQ$vqT8#hO&o_Q%L5w%@KZ!2GlJ8=kz$E~4nC8u^U8
z;7>g&VEXnmiE!6X#ZY;qh+;0GU8<$U=Q5RgleAT($vp7;v94tjVA^(xVCnKiM6pz^
zAQQ%Jmk85i_>4Sy!5;+6zh?y$)7t(DU~~6lfZ^CRKKWrwnR+r{ySPvQ^jFXE$yfR|
z+E0Nsx;3q=@#jUXj8B&n%!jr6G1u096D;4o=12I=%~xN^fMLoP5@7nXv`VO^{ZyLT
zKhjF$FZm{N@zv*v1XX*Xk`sn2DAw!4_~d5{)pI34xBY?uSiVVQfGO!?4N?CnD2CH{
zB4Av2PykFrH%NfKNCuygXZpcl5lGyiy_G4){!vj32X0D)9a#dQtASwjzEV*1Y^@9!
zi)|1H!f7f7VUB`gx_ygJ{#cgkOC`WDprlHr<P8xpd>5wx)*>1bvl6M8N_!Yqrq)*_
zAf-bh!SZ=0LUOGV3@e>HSpbYVjrfebV&Cs2ScV%^6yu3#3DB>v!2t8B6e{M`pU5bV
zLhS>P`Z$ADw(6Y}tPGZr0Hog9B>`r~NrLfhJ`emsjUKZKn97|{0Db#fS}CX|qu7s}
zAej6`1Qf&V+ajUbM}*X<;z|^r#hea&Mqa}Scc~cH=V@j3OARkO3S9grrSG>?4DbK?
ziPT+LRLq}smT|_AnIZy~bE*XBXEY|5ch3<~;{5>uunr!n0_KYM36`6aMHEA~rF=#%
z?BB2eBz1o%0;Z@91f#Ai!BRO{z$$e{hE~Rnq+*-1U&YGYK>HRXpB^WpSZaTz0J?5Z
z3C4zR8H(xrJPqn90mW$Rp_M1^KLDPPOsir)sDNqKE)lS7d@BI<<EyCX_w-XxjAdH$
z$)A^@Qf~pUw7jJP#xcDWz;J3D1MK#NeDcHg=VMjC(C!GqIJFNUS*$KbMa~#kz2!6V
ziYeNkrxy1iTE!%Nz_2nOdn^G-m1YWnq1hFJ<;|}=MDAzQ5-N_7t7Vkr$$J&RnBR;6
z&Su+WASK*M1gt;KqGCEZj8WNem`{F(=4szcfbpY5D%R;AiYWR4^Qf4zlNgHm+I|Hv
zEWJr3zJEg=`2D1gj-jGkJxD^aHv2~gVxE1Z0><U1Btnxv6+$h60s6|{i-2M0J_Rr~
z)V@2$mG=Y`TWK%B@WnSWis4E%1xRr%W`OytH8P-28YlpU5AO2Guh8_>ctTQ45s`|x
zO8|6dCdh!J@>mHll=)WxjQ+PWU}~{b0n9hQ*6_Nqgc7%Ns|r|;Pm>A7SMtdpR5$Ro
z3Yg39q>}L05r&dn|Fr<vx|bp(HOwO<`?kth8C$qSKxdmNP<g$YQF)Re0)~q#WFRT8
zrUb+;D^LN$<yBOydrrwH`rkhnfaKzzia^RyRR)sQtfgWYQbt8Fv=~Dr@r&95l~y_x
zus!;PitgSqLeiV^d=vRZ?HVKk)~*3U+#iWDim5|UKKYrHlV1pc&E8J|Ouo4?pd0QY
zq;6lKqNH|sM@YWEkzkrNgJ4jNLR91qjqhg2fW`4x1yYuERRH72yHspnL<uM<pEMAG
zgm<+SAW2G60c*wD5@0McRwitFtYP5^5yetlCjh#y7BE0RYq|uOqW`A?#-6K1ASS08
z1FW5W0$>=nn$O5Jn@5X)DXE1D81_0e)c%W6>AzhB^oPm`K;p(EDyE7zWt7zYcT~VQ
ze3t^4KR-Y)G&?4un99m3VCXuP0mkK11;DgzH$gx9t%8zR@Kyp0O`9_yB~|+qBz45E
z3W{m^A#L^deicP`;T@m+d717f6U;dw6~&sR{q`E$nk6cu;`ro;4T*PTAa&VH0Z1%Y
zi%RmPt1625@G}NjZZr@Gy^c{aEPo)P#1(19C%;4e&+!tFba$@|7(cv7FuXV<qB!`!
zJOLzKkA%oIH`D%HHr)9|K}lAhi?r%&=aZjF`8HMtEUx=1VfQ))=oY%E7_BK(Oq0%O
z=)Oma!p-0R@5Q9d`(8mYY}S6hEZ+BT)4HQ7R_3*N65&oa5pe9Br2vLL=@MXH&|9Rk
zlrbQ=pcNIJ=R6PmnvKC13Xr@^`>30r)s<=0a=Z$d>-Cg?6jy6L`K`?T4-pIx=ZGkl
zO-*IM)N!}~*#GRoCqHbN`h-eKvvev3a}2}EwC1`Bq?|9J0Q%fARP38n0mWRujtuA~
z83@T89!V(4pH>$E-PxrwU|4My0PEcigoGXY7>cFl92H3VY>EizLtkjPT~0zVU;RY|
z3@>I8^ozUmz@LR~WfLm8ic19)W8Z6hMqZ7S>M~%sUP}OCN3T}_{flt~qoFGg{ID_n
z{tLUlCquEuSp~pU`2fN2$0;88;p8Pt6~N&Cgkb9$Bv_}#6Le$hi&2roslGq?<cRNo
zN(Kxo+6q9*qM8z5nYTg(Or?+W$q&aIdLRRa3guM5B&`&Hq>|c$ni{|0fgjc<rOJTy
z%|Z!CYN;VHvA%#}IGig3$wf~}K&n%FFAOa!@xbrLSR|kldK_njJzokyYNzrtmHBN5
z$wTHd6w99{6(I41_D5~Xo%t$?A!m{ZnB7w)K-Wln7WN&lR1`zIe+Y?l=Lsm5_)0Qh
z`1p4Q=>Msq04d+}l!2t|4pdClZwe@e3Tya`Jd3nAkr4L}L0`tr17H3BH#h01K|s;p
zd8Pp7m~Ul3*J(eWkykb0cPh4Zdj%BZQSCF9R8;#zJYn-%zKI-eQ<|WASxTbPewYa8
zE`BcqhCA(4AZ3yw0<m2_qGH?;C!i!v7{!3p);A?U-?^wv<=BsW^2bWJFpyyWySjps
zx~z)|n42CJ0b`~45@7vX`$0=te^*A)J7x)hIpKQ^2Yysh3}?40gad~Nrp@zID&;Oy
zNp?MBD8`Zd1t7(KPX&@Et(O36UzHGFs=J8tC}W8NB)8}#0P!nIsDNQ)st6c=Yc2ui
zMK5F^)pnf$)_(02DvLkolfMY5p??IxG*?vseU}P+@|B_PPzI!IeM!aKdyjx(ic6uA
zl-`<P?~o~CmAtH#1SGY6qY_q~Q2>K}vj7<X8NjGGeieb_MHN-RbSPE<ba|hvfN9Y-
zDk*(d^T036I6?dB85$f>QIcZw1;9|pECQ+3v{ydm>0nyLeV55Mk()miWWca>DFaf5
z#VJ(2{F#cb!b({ySNP<&vi9#U0s4!7F~BgPtUzVNaS=%T)J7#Sx}=O^5w9}9-0PeO
zBwQY*0&&mx5cJ*6Jn)Aw4QtN;V<S~2ly5EoiA54cDowQ?;e@p*v@*VGB-5%-B?(CR
zs2ZP<J6xuHAoX9>W+;aH2L&KCSV9Gi*FI4IL*?QUklglPJ|j=LNfwouhgAsrFV`qo
zr4IJ<$<L%Vkww5<t3MT;Dl(LWhr1-eG&G%0emHsC4HbCw?P>uq9?=lDXFEeNjD8~k
zaSJxeKypRx;*-DKDWSwJ+RkU>4$C?ipj(xv0)|gZ%Yf;aT`L`!fRbWZK`?rLB<R)`
zXyx~TgcPx=8Wp+cgrN3mNpO5YCAI6%bSCx541R>4N$l1~0(6UqtAOE<AP}BZ<ddJ#
z1v`pB>Y{TpkT^;E03>t>DJYhSrBxs$Hb5o5W?6<3+v}|i7-XFU7&lenli#ype+5RE
z_kl#H*iisdi!G<}(DkW`V(st)pZr!S+uy%XyT2Dv61Ucofz<!CQ-P$RuN1((`wPOO
zYr6%M)bVdcz#zslKyR+8mH5dDO2RSitC!H_cUmQg#kDiebbf?iMryMr1Y?8w3QGL<
zCmE3Z%Af*?`YQx|krFaW!mcd}pew&c0QA-gd`6zvr~x9OdqqP0%$5R5a?q>*sUGbo
zH?_OWQ1m^t51T2rlZ=u)X)CQv+ed0;#Lp^LrUNr1KtJ_kKKX+tZ*3?8Dg8=|fc`I0
z01_8Z=aZi??HVlt$?yNgXMOJ&hN5$iQ2>Wq``s}A8pi{_mEpN|8HRJisTe!%AS6F%
zE8>i`>P&)h*JT+caad&)&~<q%0jX{42tdlaFIB*xwvnjp|BO%mSh20|%Ybf<_N%4q
z+*3kHYP^q6ekQe4lmHmgd<=*wRaOC#+Gszpk49_Xf@G5*V3oXOwF)G*VG^MK<S_$`
z=aLEW9S`upAL7xN1u~GlZG{3P9Mit|@h@)+D5l*PX{8%nO-3;)yEIIDK`>3~DA8(s
zq6!!++N)~$u(-DRJWIhUY5B(tFv&g@NNsgZ0gT@;0noXI^2x8z(rusu=wE+LNIi6!
zp`?7cKmbgu*Q-G4=;s39YS*1$a88#{5|(a~0mGlisTkCwJn)Awb-F79x`QQEz;HcL
z0L;zw1igQuilRTS{g|6aYo7%Di?Osy>RwmD8N=nK3`m{OL;w>1b5luJpxv{vPLzU`
z?)^W|H=k+G1HT?a^(+}kdA3}k;#y5GuN)_!#HI#B!uZw-kg~l%1>$=DCIHEGx==}e
z8Y82mENH|he~6^%t3@Dnn|3{hthEY?e(tZd(rp>XP~!j4seqyDJc4C1E-dnI${2os
zC1Dx&XbUa*pYVhfe+>bvn2jAoAfagj15!VEL`7#FCZWW)m`q5RUnrpHPTTq97n^v$
zg#_5<-jacb6Rr}B#eO3s?U=-{ipkYJCi)-e2q-Dz|51R1_9I2Y>Z??eudO2(JN(VC
zicb#9fFboZ6@#w4ijp#^EuZ{)Vn=H~d&bGdMHGvaLU?q&n}8Biw3rOU>gN%X=W1_M
zjN>l>tK?rss6gVVy)uw6cd!Jc<kw+9+@Ap|kIGtjh&=kna|)0$Y#Eirmpug(-Pt}2
z&}T#wbn_2OC`nzaDTMv?R3Np(02wfUA&5ZIvg!;-NgqWpP8lhonA`P|0b}jIRUpwj
zLIli%CQ*s6x`dGQ`&L>dz9`L)@M}&z=+R(0K`TS`77Q!H%q|4o;hwaL+q_1?N==x<
zC%?JrMHLY+{CA&<p<!KylJe*m6-bExRRSLA#_`E-b-U401yG0H7J-x!bEqUR?a@;Y
z-Bhtko_K>0n^8ibGJBQ+B+gn%FnoH22Y%0ST|0|FVnvSxn5yh$R4%?0fh2n=J|h?Z
z!6^xd|MI8|#B>iaAYoNL6?4OE5yjYWE}#5XF{?fhf%u<S5|Zl-6i^H!CQvb*-lL)<
z%zmr@y1Kv0faOe85isbRt3cxK*#e-ibyoorHq{gX!!U`8>Hc*p2}ysdSS1EW$$+6@
zGZA<+u``u~gijfYspANOzIS&C#WFEN1QOb8RDp-93Ir;**E1mLk0&A(+n*9(um@$p
zUUd#ZU$h(#{M9gT8La@8+}aE<Y|4`W{W9&{F`6a|w34G$AoZu}B4A!uno7L&QyIlj
zG*<+4C3X=EC9kO{3Gqj$#BJ$KCGql7DyjMvN>t=}j290O;^%xLqZn+h1R!>SR|JfW
z&MQFtg?NIg`fWnuyUik2iT2GZ5VxlY6@9520!s41{}jOBZ7l=wOP;7e>=5mP^eFhF
zh+^tgT?SIUDJr0^rhTmvtu_%QPB&iy5}O`YfuyKpKKav0S*Crzblv?_3~Mq8i5Wfx
zXHr`=)5`gdDvHq(F93#>A4-6w$Qu<%95P7;^ch3><PV`AwNnD(3U>;?gM;e`Ci4&l
z#rX0HpZsw0q5~oj_w{i?(*Dr|!y)ZsoKUhjKf({|l0y;@YfBM;)aYa?N%0kU;Aaf4
zn-X;6mI){i<2p)!?#^2gNWMQ=CRE8(fTUMn@yTzV6kSh4tGk3$r7FWJF@L;PGO|RJ
z*q+%kka}v103>|eKmn5Lk0t0bo)Th{n=!27U#1i8UAxEwf6xcbvLwK8y$6*fLp>2C
zVenB2h!+nCz{3vO&EwboBvR=zS_0zIb%Y0J%kjYPIezyel~8G}02nKGSAc}ulO-U2
z!9OC9RALIBkz0+OCQ~W%vjFHkjj1G$+oqyes(d5>DKjj5@>}T+HCBP-{$~^*e*SA2
zh<pBrKqa}K2pCT0OMq_I6M;&pG!0#v$SAry`xubg=R*~Ux#gyE`|%$<@CS{pdV~;T
zQ$&=+<r@_srNkmCNx!ZVP-4Qrser+@R3T*kEfYreC+I`%M3lJCn`pSYilHRjc&7lV
zMIQ=)aZ<WU=$R!0DMPeNO+HhDR;hL767K)FpC94ZoILy}m4qQp1Qb(JmIx#iTfhLL
zJc~-KrLuyOtaep_gijhXz%Vpb0aCl#_~iGTnxI`q(y%rHN=j@`6-Zj2r~t9OhRIYu
zoFM?Z#o9+JCMTCx$&0NbR*$mx5|U<jQc)6rO%?!S!_E>Ab4~jIq+TDYprq7)BmxP8
zKNEnsHGU1wL=CTAh*-tEXv!zQ_=NCB3h*c_i9p=lI6nC*{z*qF$z`9*D8`)?1;Dgm
z7oYr0N_3cte%D(n@$Oy{R!Poi1u)E<DiMAur2>hI-Y9^6-Aq0s_w#==-FLi>^&iLa
zW0RFl8ka=&rpzv(Y?)c{B|G6qWV?inBof&!hqKqc_xRrP-1iwAR7S>?owAB3>-RnH
z&*iVr^YQvz*L~gloL@hw<1ZY@T`^1r3QNAP0R@%E%7jdp0_2XaO30t|p@b4iU9018
za}mT1=1-o$2$Gvhbnp@tr7-7|0))3UBph9kC4x9BZ{IAH&|iN`oR~d{O2H-l0Ulpi
zQo}I?_7Y->vqHC$7?4+9|I#by`<a4Lczu+Z;+Ud05*ToNut@@9v8!T=RpFekHQ>1A
zsRZQiZ6Kytom~8c%JF6$7>cLVRT+r&xuyVtyKgZdui;}EDCqLK1{{s(HyHPwWhh0L
z+Q>lovtAldQ1BN60{OKGMfG+OVyE?Qn^*&*jAL?_>2I6-lBFe-q7H*-<sX(t#do8C
zVHKM&T}<&PyxlL!01w_4Q<Q+?rUc|YJ0b)5KkCn6w0uNC@qgS=17ern<3R4{SyT#s
zIi;dRcIZEV3Wnd2Q3@aU8Lh79-y5-u@6swa{uQlwth9z>3O9D<K*8EADiAYVRe;<x
z9T^Z=7?6Qu12%CWzr+j$2>%*TfugLR#T1XL;K^|XD6FnO33&}iGL*cfuS!5c?<ke{
zoNWq9_|(4=5ZO^zFKh2IlvpUnfoT7iG~nm~CZ@PnEGJn33jVpwfuh?{F~utWZ66iL
z8)nde!oMaFj@nwwC<WdzRB{JRQBdL&|DtkiLt84rxeGZ~;SJBJ9G$hDq4-~RQwfFY
zi6AZ-oz#X(tWQ4;CD^Dbm4Z)B$S8S3^p8VysQw$;iBF!=DtKoMfnRvaaq#iUOT-kn
zeEj@NDiGbjhyl5)IuT-B-%wEEnOh{_1e?u(ywV-0M8bZK;wzY^0O30?Udn)`mq1go
z5!WhgZsLS*xdQk~tRfWl*i1OFHA%xNx7kuL#aWSu0~ip0mLLPMu1i(IwgF;_WAdxt
zl7Zuaj~EaeGnz{1ts+9$zDR7u!SO*gBp{Tsh5@<V%P9cA(oF(_!}KS!An^c|!rAXC
zScUB_4jfzAmI2W@lf)Fa9IjbaCj8o10gmNN=RoX!qL|{C0`p!f@uJ!iO5~-e0>oCF
zm4R@*tr`$`^|1uR%QqoJUz#VQ<h9q|8@UT6QaQR@e|if}o~DBfmc7GqaDnd?6*&6U
zPAbtR4G8huW`<Q{TO}EY&HqsWc*<oekxPFwl=x+r1_VFrs{jS>g*cG=N^dd60}M=D
zKq$2QO$fdEP{vApbxT0#2mO2F__Q#Uf^Z(k>ge<?DiA$8o>6JDoKP@px`a}2{A(FF
z-aK0YqK|_Tkh{dgfcW)#Dv-x2Ye3<H>oS#QWeCUDPvR)~b8AV!vBet|pm59Q8W4^f
z8E~va4++TYSXIZ`RE|<u;der;rT$JRI`pkdt5H2^75d+s8cJT3;|$1ORGSdFvXrAl
zBj4(%(vYEqf9@g!dGENyEPju5Y^MOZ4V$Y#$S_a?^8cE_fV`21#S{le+We#ep%HH?
zKsZUiL}bnTj7pz1G7zj;o=Wc3U#Pe`my@uHb($)sxWS?^<*5|Dx<y6_)xOAqqVc1}
z6vqT#{JU+ro6k~-t^I~$72W<oCA@K82ErjpOmSB9`ITac-xHGWGAfnhG7#?7jROHk
zeKCtKF?p8?6jaUTK)iWC1EM$Ai7Ac=ojyz@wDo5RC3kLZ4n!V=Rp8i1vIK-K){}wg
z)3FSQ-AmO==rl(O-nyazvBw8xps;;y6$m@_DL`KJ2VxdKif2*6G5=W&g=d{tfIyX)
zMCdk>P-yy&p%hd+%mH7Yf2hRIy)A;c!LYrxMi>_)M80n=qXZl8QGr-|Ap=6sTS`E5
zm`Mh5SN_I<@S%VLgc?*4Q`~Z4zvYCYo0Alj$P))4zWq}brJzC?83^6^gdlzZF_h!4
zy+?>uvT7*7zDZ(=TMo2Zt^nb+6F8N(kIFz%#!(f>ZQp<ad0wUgVM8wt#H#CGAbH7+
zX;rZ1Hx;Ysf{RolJO30xT<h3W{S$ZmMnx4Rd}2SNRp;trieq9&^@mfm?XZLrAH9kY
zSg*g?;#FS~8*y;3aHI?b%&Qr|H#Ja!c&Xh~0-G~9N}#Bx0>m1Ppi(gFAwvm1iD^LS
zc@3GcdoGm|Z~F9du8xG&@mZ}EAdux^z)|G~iOSh(R0=+yrl3SdYJ}X-5JEifx{6ic
z#=i{6U1;LKv2S)OK$JaHffKF1kbwgC5rTh2Hwh&?WVe{&NyuIJ2O*UCJx2+;-jjj&
zvY%9-X!$iNu{KK-l)}lC#T2IpuCLO7=$hJ8POLmAqr^X*q5#pH?Gg~*Hbu<hXa4uz
zgyX;I@56Af_av-}Hr7*s69pD3vBcU`iuNuP8*zH<(~B~|Z=9BZqKc2G9F1;LP-3UQ
zP=VtwMLCc^f072o`uwB-C$@BvfxzdNG@vlJosjoz3n72kOpaC5v7J!VxwDG$^WN<;
z5a?Hi0|mbx6I0y7@QO|h$S-N&gv@1B;t9<alxWsXf@@0wM+s~@BLU&b`l?{|B^4!D
z)kw%4a#BVKEbPgE@XF2#koR9{PH4SC1|maC5Tf7cAHDpUE5xSwh5IJNfyn!t7!c}s
zhf3(T(;|ptBA-7ZgvK9JQG(&N1pe+pTJayIDp*BU?W0wJF_ELh)*mB;O%53)G^&yY
zL~2f>k{{|Jp?H4Re<R9W^$J7b748#aADvcE96e{qKxlkGOmPpRAABkSu8gfJkT<)8
z1_VF-g^({*kx~3RvltLsuD>_p`F|@Y;YSZ-Aisu(6He5n;<|sD5IQ)S5WCevN+`a1
zuooxfhga*X{cf=-e(I{7r&aW)s|rfq-32m`d+HZ4#WAsu|DqDO*@mHn-(AXq*p#ap
z5Fep`Lx%H4$S4KZYfC_M(lP~zFFPy)q34eo5R(^5K;Z5(88~|M2M&bZSVcJ5ra43L
zT)Za%d{G()BAuTzAa89O8OR;`r34hb)Qb^z?2`b#db<n+0$c+^!<SQuJWp0o@`I|F
z#jlrZfJ`vA<Upj`Lk0xO{ZB8;{4z>@EKy8xR^if7D&Vu%*MPk1H>rdksS--;a7+er
ztJSAcc>hfm<!E>UmFP=*7)re7`zlcMaY&-_w@m}`Uu{e9-)Jd<xO#ZidI@k1A56&W
z_((zF7uJd?j)_<8A_2jQWi%kL>#Ue!l{>;q2<{)Gp%kvsA7JG1s~jcw<M|A5Nf#s_
zR%aI>o^VA3aXW$NXN26Y^EpcXR{d#-y!a0f#RIRaSn<wnB_OulDyBHSp#K>a$bIp5
zZSzL87)qh<HX(BJUm1m;wou6}S4RYKdia}PspMTxBgCrdr+atYS8+`Kpbi`ehL{9I
zMqQ_p-#n3_g#R8c0fi}rddYc7M#&qV#DVxvBMJHG)npWJ=i({uMl|gn10oCmRDe+P
zTO5d{{G<T|7w!<EvCmbMXrIX(2zGZcApTdO3KYCtix6pOl~Llqjpsn_>*p8{{pyke
z<d1140g>a^RUp=+3n4F~olK?UY%#@MjVNoVgeQzsQKE%r4G0}gW<XxQn}pb*vNB5a
z_v%#g+f3F_;%{vhv-k#2*d-v4I8dV!>rEy1)1`#a-P;83veyXiE2XuB;?wh&RMg9Y
zZ)B8s-KHuKUGCI?$P<l9IB_&b;s4twra0Y~7ACmo%w{OzHTwy%B^3#gmgN+za$ioA
zfN<Z*DiFH;F9-5psVD>Sg_jw?6U`h5eb-O{a^G7kQEA#mBmD9jf!7|$QOw`#KO%!$
znlhB=-o6qL95J3sXhat(g@<OTSmmEvqyhZLIt&P;H=|<xw~`3rk>>SyhXMI#=Se`c
z(Pn}<IDw&riu8AHY(_~1rD%1U1o*4YQ-N6H)eMN1TTck?`=5$(biY>tLQgAmDt-6M
zK&<6x6$so<6;s@w0JkYX#GJr@Ts4scT%D%?p(;s)z?b^R!LziHhL!hdf(+yZOR7Mu
ze-Q_c@3<gl@k1^QX+WUSY6*xgvnxP;qk#;Fk4eyg+~&tQz{8^?AhP%!D)D|^h7!oN
z5<(?QDJan$i^LSyinXarrLb#F2_>*eQ3?5rIS@Phx|qcWPl_-gd|@IXGUuX<lGkUP
z3iuKRD?omm-C~Nff|bWpiLKoup*X+SA7FHbPezG7StzDBIP#vC5c)^sDDm!Zsz8YU
zM>skl!clVD^%7H@6&>Ax0pT)}6(I0sV?uO<{%i!6?Pgd-OaCYX{GAmF5d7djDqdR!
z5yTCKtP=^jn=>RTZ!Ay<FWxrcGLJPB_lCP-in9vV<;#F$?>;KwwI6FJ@!$R=9D81p
zp@iOhNdgMo_XzxSC0Uf>KPeNRKA2jxN;ge3@ho;MAp`vThlJ>?B^pZJ@Ol!E+wryn
z1V5b1fV?XWB_Ob}tOA6*n+bul@5?Ch*YwYBIFd{yFkqyHRWRTaQ(T=_x*!2XyK68i
zmS0pLGGnj|ILdC-%dR>i6h9x`pAy2A7E36B1$Q{$>7n1}@V<dEN@)37T7@Uqlu+`1
zd6|&wY{pOmQ+H@Uq|{UehzE;QAlQE*m0;%;GD>*LPYlS<nN6iYyQZQ<AKqm^JZFXi
z9Q)891EIcE37)zSG?d`}BOHj%S<MJ@rxPLv#_7mkr(qTLrAt8m()m<EiBnXR$d1D@
z;IE<IV0`a>juP)7F`#hV42>}2HV16It1%$Za<~F`js`R!Qm;B8_{JC+CG@|t3cy`s
zIS}Y_LQL^m#O|zBfM|>55)k*4P=P>NJ|SMKlS1X!GKA>dxQybz|0@S_dvwtN-aB2U
zGJdB7`2Xz8f#BPp5(1^>ODM4><7FUsb`hhJypxcBXn}%~+j)lu#0IP<1QRB3lu+VB
z1>je{QvvU14;c`2UC{v3xEo@MM|y1QehF|k(BBf#;EVfQ=?{)ocwJ=~2>vuq0(__4
zgm6)H8O3(wD+MU{@~{R3vUaF|JN*p;uk^Em5_VRR0Dee+;r#jf>t2-g5yL8w=A~o8
zwlP#<uD_|c#`Pcs8?KQPiXVk{RtE<7J126$b)lUCgwE&>$KT--4aGmn&w;qbLWoMe
z34vj^C9J}SOjPnJo{>>PH=7f3SLknAPcs`G6M1Ebf@9*l-lY=zvXhDuZFZ9ZvGuiN
z!doY(L<`atl(2e<1NLjD2=Pvv2n9>aF{~mLE)c@+9#v8DI$o0q*48o*2<p#s@ISK%
z;%RaH(~eg0<pZc3t^J0ERZ-dm2?)H|K_OIIssiCwV+rmxc8N-d4urt0Rvg81#3%#t
z_IFhvcVcUT@4-(RN@&(74tW14kO5w{lLlCpIwc_R!CHmT$)f?G2lZqiHz%wBk;zLL
zQ25iERC0T5CmbDgR>mr_rL+d*HQYiaQ1)jDB{-xb140Kagy7Sj93{WeBr(Mk6*c5a
zKzzYHy`;IQcrRz_tBDPm*oj|iYTcybi;d<eu8F%DaO_%dg7eNyjuObfL&bV!IF(S(
zQVLdizkeyFxWULbRvGZTwTKf6S1Ex1o|{VKu8B(QXk~^~WXR_d;4d9efN-Nf8DQTP
z7E|1^`?b;>h_)(XgqL4@ldPR1p@idA38A;^5+c<;maz&}JEj8BwOt8;XZqJpaOF&f
zRqPc-0z%EcBgFRVpSaksiB!UW9O5`8aN&do@VZwR5E;5mrm`zn1EMqAGa$ZbxeSDE
zey9LGyZ+uV?)q6n33d~s_>uB%Lsf#7NAQ-bBcVj+R#pIh>qkcA$Z8n~Ww})#xbdKv
z;&%LlJ2N0!zN-Q_?|&%~3XW<3?{bU~ZG2cp@vY5a0I%bs5<1a9LWvA&&Vg{kI}GrJ
zmZ*TAl_jM2>BvzcB~=FSN}UucjqMVZNBZj?y?tIr3Dl{t(rRieg-Y}x6?fGdgxIsW
z5?0Qyrf`5;N2x$$cN+z8Ej~#IMN4rM@46?1aPoLs`LhQxT3!58Oz}hoJ6%(#Y~8H_
z;i08uDwVz#v-m|BI8FmXuMbuTmOWyMRqo0~D&RZQRRi4j=Ms)>*-LPw=^t2s%kL!|
z6Th^O1OB@MspOZaNhQ4hB^9gSrnLnB{F}7$tD~s|ntHSZyvpm#h$-$yAg2Nq*Zj8_
zO1QR_1JRmm#4LWK)p{_1-+V;^{Bn0fXk=#vC3>m63WS&Mk%8ROr3t)Ml7tfcEL#O4
z+C(vnpM>H1-EjVKol4wim$8cVf2aVi=M@N1+Y}ka`QHy>7T>XXO$fQU`o|$W@>7Xc
zRe~~5FwRY|{at~hgi9x@Kz#q75|CFzza9VJwK7UDYZM{cO#dnfm=0=KMasOV0>Mq+
zDL`OPbq2(G?BhW2)2|5p!}|=ycO;qMP3fSa1g4!=fx_M!s6-~{Uj?~68ZfN#4>spO
z@b655zaXrl1h#CZ5*mC=M#=r?6%~j!o=YV(-=Lucwrv$tJY(E6n~=Nf6AdLgZ!;C1
z@s*6?>hQY)#NODa0sO_E9S-_pDit<U23!rtav<-<7!?S-Y*K)r`iKL0l^tS=tA}P?
z<A87OYcde3+erfWe{V3r)+~b%Soo8Q;+#E^196xBvnld;rG^q%{u_bUv=iK?FH2Z?
zS6$*j{NpAv;Nc}Sz<Xmi2YibUt3cTH1r@$?orDtU`<n~|-}+ObGFd;^_i!7nV%C9#
z(CNMsjtN9l#1xOzaO1oNL<avy@U7a*P{LE*Rsna~Iu3*mB@m+N`VZFNlPOdJzIQbo
z6TI>%mDq|g3X1pJ=TssULkuN2q7%XMK@$zdS>t^f@N`X30iHfo0vyX85v)%AYcP0v
ziH23Mc}E3^T=z<V^OKJ`m8<6zAh*W@g8N7v6(yYd;+b#pNJfd@TSDN|PI44aMgtAV
z+qOgnj3ri32@L8ip|~Cl<v`&2M^qx`-_}q<o%*Oi{Ni{8@NQkK0TIhMDxvo+3?(@J
z0)cx>gjkKeVk2H1$LEDA;JrPTO8$_R3W|IGBL;-O&EY_vX`%!;_LkECpK&`O(Dc5F
z!l$+tQ{1wz!8wWGc4&b6`dtPDS4%nu)KgHb4f@DH!Kn*kiqj+Q3kkNtw^Wp1W*aK5
z@=*yTo-l<<?x`9CuVJr(Rp85&gkY{&Mu}Y2pC3<VJrTqua_7}!0G~2I1zeN-3J}=w
zHwWB(-=q?Z&!G|wULd&I_mL8cZzm-GD+B(<aRxXxHlh;h5alT50|kWG)Dj|yv%DYZ
z&#L>c?-i8r&hZ=we%(?7_@ylji2Pbl0>V|VD?rTOSOr4!zMvA$AIwqg{qv|;)Xx}7
z7@r1l(ZHzBBotrwW*qQ_E^1WXYs7%iJwFG6@7x!&_yO)(ssQ#K|1p5)l_2E4bWA}B
z2hLK-d;FB6*cZ2ufuiirD&RRfm|(k{CW5$SXMR%|2>lQvcxnxhQ2hI%D&Xs~i2;t#
ze+m#jHkeSD+(tt2m`A8U{=fRSdUQ}@5yT~YeWuEQ|HelIPwC%9D1J6VszjKne+J#=
zQ5s70_#=Y<hn)&asQ)kt;0-ItK;Byi6d-u&U6s&zrv|w5cga9*{w)r82kd4*@JwkH
zu(zJc0dG+g1@OK2zw7zgxe`jKk$!)?Fa8RK;NDbPc}H~8a18h9KREpJo{JzJS9sw#
znO3iVsRBjOM`DUqWW!Je@H;!oKy>^BDz<-SODOyU{X-O5vr$Hg_WeQ3;?vtNPyoKK
zuM8NEJk)^Lls9D{(vCC07I=$_w?Z=&C6IGN0`gCF7PI&UyRYX!ptAlUiVP``sN|$E
zpx}ao5a?2xqwrM~2;OiD1?6b{ieicz^nTTY0ihjZIS}3+5>u>vcm7g<V8tCO;H<q>
z%;K|>`f?zgc1<Ik-p>H*<4$6VWB9Ns5@6_*tO2>w4He+-zZu|ey^sU0FFq&4kA2Hf
zBG0F&fW7dr4ET0Fr{cM_jH9@2Cd)wB6p#SR?ij)GnyjG&jLoTpFTO2;xMP9b#T>9V
zXdnaO90!%c7k>?0Brm356{>iK5L{G;jtNA+Vpwte4Ka(KjrUVzz|!mkD#2}QRFqtY
zUU-+248?2iAOYrcH#oq1Z6bu8{VYQ9J#3LI5r#ym_?GvRQFs-H0vH@wD&XD`mI)`?
zYCtH}r~=Virx*|zq`xyAch6`j;VqxaK=6l_DiGK`PorW=Qh?~>b`s#de?Uxe$MVNp
zsMyy2B%^rR+*N>py9|}=PtP$FbGw#einD?*{)JW7Bd>zu|L8pxh_u;E2>&};LU9g#
z%mDj)(+JVrkc{Fjb&0@-oRUy{8+)jL^FaqO#kGtB_Y(M)MG{Kz{WVkqJIWJ6GgXF_
zD_|r9ww>cBf!xyy;d~V;!4Eo$Aa2<&rxT(zzm-rz)ds16z2iKBe@8cl5{!I9C0uk`
z1aZ2rwEjkl{%PkZy!&XCFd|h<aZF^z1qJYJ>B)g`6{DD973%j?0wSilDq!9JKMurJ
z>MQ>qE~EH1>)#2nmJLJ@rw5!96u@=wLk^hYT~xr`Att6c#@nhIA=kM~MhPtYfCHA*
zAF6<NU>z9<l%AykzS&Eu#J06zC?V$+707$JE|p+Avk2l6&b=!nAW(B0!5{3%P=fI$
zDiD}9Ljl5#^^5X^hzR2J$hfi+z_Yqg@n0@2qeM!qQ-SC|TNJ?ea5*77II5x)oUJJV
zQIm}dzhmJjf#uW06qg8D`!K+@^{fndCa)$0et)8>F{Z+=&My6i4D~R3abOWGEs3
ze;VLvT3rFcDf2iG@bx2jp7zjC{M%wwau?MULEH_0(ODJXSFH?iO{vSNj9MfE?olly
zAouSs3J_X+RZMYuEU5({aDTUo5@=pm1N>9GG7vqKA_1XBH^dZYxhgc%0Q(yqRKQ<z
zw+!&ZQ#8Ocb`Aq_Q<iZ;l>r3Xw(|;#ZS+(^=-hsStMvsLtKirAR|<bJNkwsNOy+>~
z`L7BP>Yl>@-^QU-g4)Xj&*PygR^fTgIS?3HT>|W-o-)9lo+zexIF5HNDuBm3j8Jss
zYo1WN3cNmq0pU`I6u{r-z6|(QzbB?R#(#FB1X#Miqyd5caUHi*j$&DNodLcV|L20=
zdRGK-I<MMa1tMiTb08e~UI84djuAr5?h?E+%W7B!9_gP!K5Cqb;`!n-2MjIFGQi&Z
znhJQIq;bF{l^0W7)Kjkt6?czP8cJx;Yck-u`Ky@X82h>R6~MXlV=Dep?HEdYcDf4i
z_d1Fx4)#5Ip8>9hHx<A%Vu6@q6)dbJ0gl$28Q|P*)PO**-2{KT1VX6%IteSwlJR1S
z)024~1Nekq65yP3QKd4mrVO~_Aq}uSTSBm(?k|Ek-LkYVmEhWy9L2hSl$c`0CuB%~
zJDNcyJZ^@H!ppx*@ch+@R^Dd%*Rt1IljE4&()vf()pWgv;#lEf0B=`aOmPX{`GHiN
z)4Iwi{PV>e@I<StK;D~-0pa$mIA9<15+O9!tDpqlIHdySoh2DypB&K$t2&D*u4Sz|
zTmhU(<r!eV@{O2c<-YAE1cq*rP@J0=6Z`{95~4NrZ&9ZqQye1Ba?ZCg!1rZ<N?_r)
z5{mm)0~xS?@-w4yR25U474qv}_}<qOIf`*eM5fira#U<%`zk1r${SR`oVP~;ENwpG
zfc;@3DuJ388A|xo4Pq8w!d+bfTw^*jAn34(S-hG!Ln3%ua3Ik3lma;ZZlnSJUcbtO
zA757i)A<w);798#KrmBbKxESo8ekf_iUZ;Q)~Z16IJ=nQqW+TYCBQZE3L$#_Z5`j-
z=2+QkwbFo4(<x$#vwWNSC_r#$M-}kDbVSVJ8?;yCglKsQ2#oN`K&)L%0i28L5p21Q
z7>dQN-{;_4pK+AnlN?$F*9T-2`*lSHJkx)s;=kNlLvgMClL6j<{!QW?^*<HG^0j^~
zPx(D`OuU0h!YY!zT1;`DgSYl8fPKL|4)8fE7!asiQvuvh`iNQl07qA0RHkkw@O7gl
z6jypl8Q|Z|;($5%Pl?J!{YOc7{#OjeJ-a{!oKu={z_CuhmcMZ+hT><6nBt<=b^pqM
z^9TK#&3+`RqS$ZWWrPR+QL!~Liy+Rj_o^ZThPRU?z_Q?b4e*Rx!U6Y)E(&1D7%gV;
zSzEFh;Cu16Ci%nqC&jrxQ^hK4;0g`ko{BQys<V%ZH+?ok38ei;;8lmnC|sMP5k9>@
z@D|-wQH<{nmH<m{j|}hyGZ+y3r#}bWy(en`AE`f5e~SVM#nopr!Q1B`L-AZ_paQ{a
z`C=Boa67vxfU#s#D#87KsVGrrmIg%ryUzi)KSc%n<unNhFVmkWcj<!~3UAe#0gl2<
ziE#6(3<Oqxt^)SIo=}N}M=}(fSASfAhFnAOy!ao*<{<~Dcn>g%R>lP~z}q$yv-s8N
zZIl2<{fiv%OdhHMuC<S-xZde0f;gCeGF=AT4Vn`8<Q^QwFO^pS{?8Ikp!iS9WRB7w
zG~XTISbC2xlYn5mIt<`<j&i`(c)JWZ9<P^xoGbYp2;G<?rntUm_dOMGoq0wG4O_@j
zob&W2#<_mFgkoO5fdTf1Unqd{-fjtSwDM|zwSEf)u=W3mVE*3_Lg2@5IaaRupNc82
z?%kEgfVj1q2Jp%si78f&%(^n*|M4{j_|6>@v-qsme{mp9eq9EFwf~iX!0+X#*rr@y
zRH}Zf0HMV(32=0srviyP^^aPlPX~_T%bZRtSEDWxN`9B)Vv37ePj{nYed$$#PcE&m
ze*IDv62A_dm1*T}a-CM*p_ggp`8AagOI^>zPCWDPb)({%bx}s~Zfz$4&Zp}Lp~XvO
zl+gMh1HyZ2DL~GKc~n9zjRe0oQpJjYr@w*$`&J1h<CprJR{z}70P}|W4Dj6kP6nLk
zw`zc^&29#`8%$RK!;$3@;4-i0K=4dIg6qj94aL7tKiHYQMMm*<|3d*h-ak3u-q2qL
z49XS-@V!2oV7Kigc%Pl*SXu4_sJNdMF_eI%CkIS9Unzj=>CYNrL%IYw$IT&xdevhn
z?w<V=z&r3Q6$o~^PjEJ!%29ZeUnL-}SgCmaevJ@5^&_o}(-+DK#V;Ix^+Oe~e|m@z
z+%R83;rYo7a5?%?@jm!lLkY#JQ8BLULdDteCdVqUXrByNhc#9K?~!8?;BPcj0jwMP
ztAyWT1l}nmp?I#0QUG7y1P=JCw37f!?~4RipsI#q-};T1;!!vTzsdmDn;QuJ?%&BM
zjy=sd;pr*~aJ7D<0M;?j7~n{_r2%2h!vS7F|Dc&i4VI~V@~N2OTE2}-38rBa2!R_h
z4J*U3JyZ-#|J6Qg@(zZTeWyVMf?w2-0PBqB8ep>=RRGtZg%S|B_BjL0=bo#8>t+iM
zq_ZDnz}ELZ1AL#J(E!V%T7>ZYt_0_xp9q=NDzJp&JMUe%T?P#A6sUl8oF)OjL8mmp
zK3#wN1ruN9DAosu8DNvIbHKf5A;J4)84<)S8|U3t0pIR&RIDvJ5S-KQC|H^H#yG%-
z<!gX-`8r0W<!4lU<98EmNA*|Gy1KlCV|>S76;oU*(0?-rj9YKYKxn{61lP&tB8X!Q
zC&qEWx8xxeSN*0kifeR|3gk8$r~u9*Mk?;asTxXd|Kkks#G)K<4w@~dxaG);|3sl>
z`YQ^G|6~2mCqH?jq67{dmjFZewK5g?5|zOEiVVf`=yxiPU#Dv*p@(l%;U`~{QQST6
zaDXRQWq_gO4v9*4mj)Q?l_g|NxTK=E-+C$op83fV;8=D@0YdHm#{jqUkOp|yr&6&z
z@kuC_@AFh3+&wM>hHqXMQ#^L-+F4YbhNm1Qc-SJQSlQ0JO~qFETL~o?o2>%;w>(CW
z^ElvW!Knmh50|Kv%jQ7DZIc1hfqyk1(Dg7CpVm+WaV_88s|>L0?I8n>3Dp%KxcyTG
z7)l*f0p~Xj3I0A`GZf4Ay9y9(SV>HAgTb@kQn8=?k)gPs9wS)KrD!P5jgL5Bee0?W
zcuMMb%<JpLP`s}^=0I@fHW{!i9xA4|g#FzH65!5jz<^-m_hi5{WWJc<82`lC62L$H
zK?OXm-k=hmrhfytx>sjd+1d`%0P~X*957rtC}#2V<7}b<+&PxusC`gDan$@s1$^D_
z5u8=hL=b1CzW6gQ-m6(0CHl(m8sIq4hG0s+ub{Z(h6H=0CqpsV-ckX3w?7EEG5!8{
zgZ;%uT*A`m3tIVS&7tD`X9pF_qb6LW58pG*x<ac!=KvLJ?GgmXvwsP`P5M_scAu}+
zgyOSo+m^^c_+Uf=yoNPmij`;mHyYr4<0l4WUoIs9hW#HYfal776)^8_#Q?sktOmFS
z{>TA;6}JpT)^^f>K>fQaU^|(@2~$2Km=<-^P+ZlT>X<x?p*UJRB=8ZPWR!5Fdm0dF
zQA-6J-IgkV@%P3YaBse>0LBISYhl@6lUC6Y_av;GlP6Hg+)_bBaTM;<0AKx@RNVj7
zmQlRc{S5GZ@{tCZTz5HOuUIGn&P{O*i1u2fP-(3{Kk21jl2IJ(&ZvN4Vs8$(-qgQ0
zOkD?SC{9BqF~w_<w@d$uGoBnHquASbAsAc77>cV@cMdp4)*(1YKG#sJAAU<EP}ELE
zao2j40hX%M3ARlA$u!5`6&rC;Z|-gmct%!`0L%6^RP3uq5FAM%mQZ}Kx%MAo3OFz3
zaKMuAwFG32wK5=-@8*D^+H^6+!N%XKP)V&eUqSIS`;JO}vz{u7wd^=C#lfMg7dQ|%
zZqk5Y)@l`SF8W6XtY>CYF)r!FQQU)?5j@jQ5xhTlld$p(JT0cUWlyg%8o({%6~Hle
z0TrvEHI>+s_r*pWY?-^50q#xfG{9QoM=`}p{3$AlN^BbkEN|8jQ>=LN%?z;poIxdE
zt|Xy2PX9%)$v0#a?~LYz4CytF!lmg{Ed3rRD85~1Ibh%LGr`(&oQh(}$zp)_t&3uc
zt2<ZDrxG~WQAG)!|5eQ5JNEK_9B|*cL&cjJ&`>OFFTwEFZw$r1rU4cA!A}VmwXKAe
z`@#S!xm#-y9N~csEAKmZ35NO^1m}mBRIK><7GjFK;dpjj0)j66nfHw<tD^Y6njr(B
zl#vn;SZG%O&p!qkux98_l>I<RLve;?>Z`Yn1pk$u3Rb~KuM;8-qYTA5KVJeo3wA1i
z`=S2h&%5tqhLRRf)c|MTBnb!<jO9RZeg!ebqsTOLpyHTQhoRWVwxwdNGFC+im@E<y
z>StB}+XYD_bp4A1;R-nt;AsAr0$8f*Pq_8sNrvJckgEbYeI7`FrN=Q2SZ^n*gfhPo
zB5Qt;P{JiYAb46GmQjpj;~X%^<s`tdy%xc@bFNHfb4>;~x6YxG(|(7F;@vet1G2t*
zT}<&L7@qf60C&TwR5I3lN5#IRCBw?Mxf!jT-~P=}@()*6fZ)!(GT_}flS=l7uc#>L
zy?)_9@H>|T1S%a-fSi&OIB=wee&@X}dqfcT$34BA3fM~2=K#O*xdvpaGgZKCJt$`J
zosSkMfPcej4miUNB_Pt)paHH~PZ?mIH&_82>*}b0vE*wqV7`1#1A^UpQt`a_hswSs
zn;BMtn(wGUYPnMy;Oz2*ioaGiM+uy-D5kh$md6KFAg5wX0<5oXqvH1!5i;MeCN|>W
z?5j^Sz_;;T8L+=zT>@<HH&g)UrKMCdGExY}^4}5c2?+%A)Y)1>@eP_!-4|29W*N(X
z?6f~P!1oT60rMMU#1zN4{i`_;UE4teOz(73fb=0HRlwr<P6m9tduf39@BC27?cG#D
z36v>K#min5LEN%&g<Au%7vG>_c(9A3xF6{6a7W8L4aK}RM(|J9-|DV(vx=2*XLk-*
zEq%lkx9s%w)d1r^|0sa1;s*@ygg&I=eq|}A@}LaC`r=;z@>d)tVa1nMV1V=TVhylK
zn-su4<5w#7)hjuQrRcE=xCTz5Vw-V-p;+_ADuA)hMHz6~8cBfX{O<~2DzTOUhFed>
z6pzA{Qk?^y#v5h8|6^Y<i(izQEmYuecpsHO=eUMqZLk0K#p`}6qgWzw74U59!+>C8
z{qtr1=}#)&jNTen?yvKx7^PbZie>FO8L)5E@4Rt%Yg+NW6D6$NpS;b1z!<v%cm{q(
z$f>R{6sP)!1cd(Ws!)07b1{owuldgz5WY5<1D2W7B*1&^D?;$ee2!w3{?UNYTn`oJ
z<Uds^yp@>Z1|5a1IB?|a_o(pchg1|#?k5Tm7}Z1uf^Cwh_yWI4D8@?nIS{BZTmj6P
zZZV7R>T7-u*sDBI0LShv3~&!<NZ@x35{f0Yj11T!)x{L2dsaSRfHA+L44D5nMg;;p
zmr${dcp{;gDqJF1Cpj5PVEj-O@XnMuV96@W2-W_e;yHg!MoB6=n-Ex=rJ~pmB`bj2
z^#%tVRbFC%y?Y83%Z+&wio5hm4e;!{MI~kW>jWdyKd`oj$2pGS?;Mr^)4pC*oNvum
zQLKuAV2hm=K|Bdo^C}q#@LdXEY_W|2j<U<CWY-y`qWBW$5RCDw8p@H53c;}OCx&8g
z*+mANo^n*Yqnr{-=HkX2a2#x+0oniS%K+>BOb)o3e@QTAxK$Ky{tyP(Jy$fqQ<x>D
zxbvYGe+blYb~8gUP3o-z{)c}{RGiNhDw|)WVoZ&56jOMi40zb@8sMw^2?y*g5(vJ9
zuc#=d76%#NyKz_nlAZb!ZvI83Vt71IY{VTiFJDO|El>aMwJf_JV`b@Cl~(raEkzIq
z^HMccz^PdY#<I5=ino-31Ljx*F~z~QX1yiAP@_Bp9HYZj%#$(|6khHV26+2alK|7_
z8Wrc3wHk^uZG{Z*t}Pf~{Nu8i;u4m5<2aD{{S*oCzFUflx7Sh?#bW$f0^CK%G%9EH
z8#I^JUqM^f>tZ8rFnH`M8SqR_QvjD~o(ec?&(na=#wY`_%8n!imM;)NobD_oDM04u
z`sdxf;uj4iYt)-E5S=(!1uPp*Q}O-&1Hti)L&M5wV^j>0(GrT|$3s+XrCb`yq4-on
z;G>NyifQ(XyLxbsgkre<wV2|L<$S%50hX&HG{FAaJTb*8<rBLEWDm$t0n7HI9B|Bf
zTLWCv;ws?TI8_0hvr1`zm+LQqf13WSZkjt%U$tl};}}z6TNN-q)!%-un=3d<px;*-
ztyWA^07H|hDqtMHMgi<u7dgSHf0<{FnINM$-~CJlY@gnesQkG>q0*x`6hHQ%lZz#O
z{5||l0c;gk$w2n-JsMzoQdLZGjC1-m1qdd8p#ot`KuobRdnXcn`%5Y)YHEN3mYpvX
ztWz^36h3!{3S_>zniIb0r~&rGY=W!o_Z-D>;I0br=@%Jbt$khs?03J^OWY=cxEr>u
zZ8?y+ey|Kg*M6-5o_>84z)<iv2dt45Vv4hZRd*^tVxOBbU|+F-;Ey(wP;8yeVv2*k
z)$0(fdvYX{q@iatz}NCm23XgPR;j%DvI1m$;i2L!((k<O=?$?FH<;LaEv+&mw`COj
zt=bykJ^79bIOl5$kos-94A@65*8tP=H#i~fo|xhWE$^+=0P8nT6u@4kA{9$sV-<ya
z!xE6C9gu;Hzn5{q{OlkV{@-U*65A|PY314=rnrRp_(l$7T>d};>>DPkfbqx%jY^#!
zGGM8ftP<vSB4l4I!BNbqHL0XdI7={=TEVcgrXLhjT+2RRe`jWnN{~=29aCtP5jSZl
zUc(bA*{|sET>F7xb?7?}2RzA7shF1Ls3`W+;}yU*U4JRF-$)jr_&E&3B*3(yCL!qy
z{p-T^a;k=vzvtH+F#KJM0k-xHWWemnmjHi_0b+_9w9a~+O5n(k3X18U&lzAIIaLBO
z-t%aH<+~xAO51m2z<$)g087`t5|G$`pbU6U*Jpsc@@O%|wZfGKbHLJO0R#AFzbk<K
zfPSrg4<BhL8N(Ae5M9uX0fy?csn}}kZ?@3v(Gpg^dW*#@emHYaF(A}+fdmA{w-ZyW
zLR<8Argf}YL$R<*46s#65>p)G+}ut9?Emc60IO>_74Ozw9L3e9g+|C3Bc?bj!?uJ2
z-g!w95S&zk0k)arRlvDDtO4#mIUKN!A1ML+L>C4)zr3YmbYB_8`cFdzu&0jZfT4ar
zf+u;WisJof8^Lqtv5eyC^9O;y^;khM_dd-5OIHU2Y!>~~6#96xjAA@9R|4$u?hN1y
z<2ve%A_Oz?#YQ}2mvl`Aoc-SAfW5{<3CPJ>rU1TKMKWM)ae_*Us~tyisNaYwZpW5a
zUjwX{s}L*?E{Pyk=HK-1jjYC}8A^71hX&*<I!iD&{96QZu%k{o71Jv{2*zdQY2_(%
zT^15wr`0kfz_?uh$heQ!7oqrW{QOh~?Ek&^mWV|;iswpq6-ZY{D}cR0e=)^b)`e3E
z=8cOalt2T&3Rs%xKhfepIysfRG%>|lzG(v);8@>Q0#axFD+AumUNOZneBwib@uztl
z#d!Kd4an*6l?w2gR~cYiZxK_R<sZFR0W4)Us(^9gFb%N&+>!y7VeiU7&W&CgV9i`d
z#W~<72_@(631W&HJmP;t0qi+*sbqRP60)aVRIxIgtwzOs?Kwlq{NaWKWXLP1q*%|Y
zDCSRE$bjW!HG;kVSca04I7JgE{*yB9m-kRJxhfEBp+&_@@%^{7p$gbe*(AWa%`B!^
zIo|q}0j_7?sz7FRkOZVm$&-O2w-Z&sTWbcv-0XD`#OcNZ?@_S?=Se8Zx2`k5^v6yy
z#WB{-t27`fm`%kwD}#!6!Ur-|Cb<S7<K@a6#e6-DkXfX^^4aHe8CI@GW*NxLc|gVK
zFXAZnf1AjFVMi|oNcr)k1gI(7Re(3Y#{tV3{kMsne}_pZ=H2?cHIpULG3Ie68CLG$
z?Pb8$;s+HlC9P+Gb?yL8=zLBF_I2?|fN|Xj4RGv##sE{dlVXae#c<K3P^o)c1~RvF
zrD9$(k&0`11&)=k(lshs+l&fIVh{Z#u>V`1Ryj-N(aQRJPYuT$oVJaMFSw1N94S#k
z1p?2R23VhNQUGJ&KpC)&*`orMp|do=yL1W%+`nDdfXw=P6~NrTmIT;$u9pErqN)L=
zTcgDkkJKERNieAaj*>HeBbBU??<y#!YWfXk<=voic<w}oReDMN8_2ezJQdH{0~%I_
z7ymgh<KCc*awNKrU|Lj(qZsNP6;s>||K>Rol?iQCAoXhp!O~Q}W9IC+RII(HGaTa_
z+>ry;Umr_^F((utb@XZ#I6QZXnBo#XUrz<FRN5f{8I=uUiq)=^kOJgHOG-fMo)iXH
zl-Fe-*uRwmn2yclfbn)!3CMo4uMAkuUe&1deDQ79i!qeUNKJ`w=rs<c)$gbQfdvIB
zV44)6Vk-ACM=`$mHQBo8nF!*p8u#y&0n^=P8epsVoeCsQ)?Y7!X$MD1UGs&Q#b^C9
zosgd1T0yalYbXQOtCKY#W#&>b#leoxQ#9bn_SGCn>exw4u{u;)|G*kbf2E;h1hy-H
zakoMxwaXzHW&e_E9B`CZ6d?1-0vX^H+jGG9#}Wplw)<WKtp5!aQ(P;1(-;+S_0m6;
z*>8<vD8{*QLgIgG6cnpzGY5R_UebVrId%mwys?)98G}F6fb0i-#T1uNGa`gwkH2J;
zjDcTEfIId+1F~8#6jK~*dAlMN^N@8iO4{DPHNez<iH?Il1?5nuO;mEmb|M`5a*BkN
z$+l7jl3)B=)~3q!MG%)TO}N5=oTK{ZBBO@>aWKp&t6}AOd!Ygt%l=4ZUq+IOl67h@
z!F*mn%NT#1N=jNU1;?b--7EnaGvAbfq_<@bWMwvz0Mm_~46yxnkpsIbKd0hYQjSVu
z<0Lhq_&z81x~~De-eVcazI;{z(!YF@V0wEYNAYcK&VaPp2PGh@az_~mo!LxqtuHTv
zxaIxb8#2K9)eJF3F=muh0mE-;9B?;O6~J}2h>CH<9SJ3?;#xxL_K#E)d$$P;Ftlr|
z0H%q4F~u!sjAJTr_}dc%UkX=HjM)(h$mw>IVEg<7j$*B|Q3Xsz-4wvmzK#r}q*mcT
z+MN|t_U+xKpd@|oqmuowwS?kowU`0M&>1ns{V`13EdlNqe;JT1?Nu4YRI8Z+SYG_y
z(cYE%^J7Tf%CT}d9SlhRBS8gBas~$s?}juW^UJF;kaI4q00u2Crnq`$zXk-u_8uBa
zU~>&Z+8b#ES3S<L%9=As0nEKxP&quuM#Z|N36<1y)5Rg;mQx2_Wq>LC5ta1uhpD7a
zt0;>ken*>HLIcujy~e3X?};f^*2=>Psn2F}l*7aCYJj2f9tlX@t6$6JSudkxEuN<V
zX;UjpK>BN{3hWD4p>lZ26a^*u!?!iSvhE@WY&AZS0OPtuDyF*nTf#YSsf?9<<u+QW
zmG?6gLxaa+iu;_?ww3~z-=0HAO6e?uSh=d6;ecg$bp>#I|A`8?noeVY`^Oy`V4U7o
z0u1lP6(Fl?PZcmb7c;<gE~EgNH_C9}NbU0)l|lM9r1`D73?=oXepYsxSw%6I`-oQV
z7yt0XJou4<Rr>gq5|H-p6&X0tb_@d??<^x^T^=Q)BzF6W1Kux&5>l014JEl=Yl6|a
ziK8SnouvRN`#xmA;oJJBJIkLf(`ve11N=3|s6dw4qX5RX*9h*aZ5fL79Rn59=4Tuw
zy<S~HN}a13O7`u430Y-J$|#1EKM9V~ofVWL!^TQLO2%gz;H)!@idrDcC^@M;6d<E#
zoCBt`Dh$Z{>5-V?RZg+rRRD9_FEzl_RsV^f_D4xZt6ut9N!9AgD2^|i)5>+{w1i@8
zdXodT2d5RlnzUO6OeYtLDXwL>(}jvV`AvpmdhrL|Q&#^*C3)y-g;rzq*Zp9rOcf>N
zxt#-+0mB(!t#^WoZDEXx>yw3qv<2@<3B?yZv_(~b<geaffHiBF3|MMpsetL_K?L)I
z?_??i`Y<5%Tt@}Sdi5n0$X=uWlu^I)YAA*?3lzZk*A))9W;T_9#K<)UWSbN!ho($c
zQ4D9EF+$lCz3fsrm9N?=faUBNF~y_EF<g~_G}jfvp;omS%8_T=G{Blxhf3CtKSU5`
zB_9qdfT7y!5}^L1zu9bMGG&yM4-PAUu~R=4$Sko{rjq(l1CBhNOECTOfr^seRe!Uk
zt?EN7qkAdCD(f}<4UqNgQ4z%5$oT1u0%X^G%n4s7NPv2B1_MlG2Z|{UHWUui2&-!=
zz@eA*N8x>?hK%A!>LR8%_{hbv3Xqoe1C^|CRS1R#*Hx^N_HB{@%b<}Okp7weSvFb+
zQOQ|dC^q5-57+HR#qeiY6(w_TBZbhhmkij4T~>jN9bUb7wlWmUjPe}Fh)*CG2Un6%
z(vqiWKzi-&3SgR=PDuRUBN@e7@ty`)KK@1l+~4{nz!dJ!0BfrvD(Z_r<!bHMU2Mdi
zPx{tFtCY!oWfW`DP!&kMqJN#5YEM&8jGd1Xvi3BWP^>#nXh8ZeBUHe)Orc_!Twg{>
zH3b-uQ#V-wY#m1vO!F@@l+@jSQQ7;yrvzj7jv7|Uy$c!OzVwL<<b3p>0yxgDBDf}2
zl~A0uerJGhZf_14x-}3}JW^Lg)&Sf3V=`dc;by?$rjH5cwa*n4V@4Ad*tfn9!SL`J
zPbj{f%;rU63K)+kbHMiDG8wS+ysiKlf9W5+Lk&A{lp{T_tAK6uKLpbti-KZmp3VXL
z!;dw<+UE-~#U*n7>_af_)n8Lf?23xjp`Y(7K>G5N5|BCWjF`n|6`j_ALu<;&fZ4x_
zU^FgOQIbLfsW@jl7?rD^6H>=oHI$T|`@|GCXd6_H1FmK>8DQ<)M+MCJrzC>61OrU}
zOc7I@mD=K01#pylQvy={Sf>J(Cc`DbP^y)f;$VALcNNI$eoq0CsvIKt+8<&ls=P`9
ztkPXE#aV|+yrlq%M;l2%)~|gyu<P0!1{g1mQvqu?qXc9<+{J*jQ&BO+?Km&p=YZ+_
z9xDDo3o2<X^{37}_M(Ji(!W?kE9=7Agrp5o6|1De84BR`=<mbq>rFHi>!u%QWjv?<
z;7IYzpq0h4T*WcQKxslwb+ZWKju}gpVL*19aWar~{xvbhD(SlwDsW(Jf&w_N-eka$
z*9r*9S~CqLr>>81Xz@H1#o*bm0BWg6VirG{?~K*}>)T5Sw&(9FC`a<LR3JH&OUUXp
zm7&=F+RFjo(lHuv*x#K2hPPL$Kvwf9GGRm>mBiD_B$VuOqs1(Kq(^^N0PBl?SniA0
zqm^U)Qw^)k*gFzn{q`*d$f^Ia1Q;*1=YZqOGAdwAUZ4Rvdsd1mF5xaQkILcWH>ubr
zd_pkXIz+H<FiS$>{(KtO0B6tkR4hGzQBjP$^zS<3$Gt=-e(L%TAsG2BM&-O-oa@@D
zD4Cr`$bjKNe+h85_j4d|!mBD^>)c8PQcU{krZyi*C_BGsrU1$JmS}+K;xPs!dO8u(
z?k316?vaHmkZ3-q06C5%32;?jCj-uE^EJSjTAyI8+>vnPRy(l~k1Ml@nE|}0z62az
z)LaIPH|}vNrm<p*gHz3l23Q;4W<bj1-#L&pVu%7bs!Wjp*Pd}Qkad3-AtS-4pxBzM
zmH_+jZ_9vM?o9=-ocUY@a#&eHTEA`_C1p}q8L;=;t^l^eohoqTl>WM>?d{L0R7#Qo
z%SiqGY`A()1o0>g+gmGudGmA`NZ&I~0|I5{aUf^v5rTdDH4(&Frr!>!K=PVSgv^ls
za5A<P$yg<6sbY$Qf9&g*0K?B6WI#RnKQYD1I&ClmQa?K)1C}lWIgoPh11iQ=zYz>u
zi&U&^hRR|V-wp1R0CUHd3Xr;|x|m|+YpcJbQq4|U9qhAT!^&DqmI3oueg<SLZ6E>W
z)#Jnzrzed#NBDlsM+!>T%?}uGX!bi)(rSK1;4SrI(yz^?V~(`RQxl4BC#TI>F$Ijj
z3}rytH}w?2a^R8#c<$+sB5Pr9j^cQ_p8;lPe-+4SP*np=|7)lK>7Ugnm<IIFP|}XR
z%K+nm%@VNdVJ{9CcIuz*qz6w#D1Lx*G8MpCs6XV)A1hEv|8~5FRnnI)Q#lf?BB2}}
zuwG1Yx_i0)1+r&JH4P>G#_zN;Ef_ANq%IyJ0p>HWQc1tMlVDm_Moqx&oa(Ir26?Ik
z9J+Of0XbiPDg(BiH59-);X4hm<TNE1YwB;d#9xCFR!2VWCZ@QD8IEHdQ2Sk^a-ec2
zDuzLqnMmTFQ2l&O1`MCqr;@aFFh@yt<WkAHl%k@fy?;&uvg!<Bz!B%G3Xm}=N22oa
zGcm<2AFR=ykYn%3QIcN#Q&8*cw>6Z+mIE1J+}V}`_I0y0Aa#>|JGMF@hLV(iSOroS
z^pOGMpaxV@mYyKwtX@E{y7c>ONW4xZ>G@+Nq4<vRnqO;xVP8m0Q4IYi%Yb$1Bn3!X
zeq7ZtkV=Yjj*!;AqlDF-w~R92uzaEbIsFDPAbngK2do>qi&^|UFPf_XyD$Gs#awk3
zVV|!&$Lhd@HZoxS@JkIa1S?B`vFB0_So@!1K<0trR4lOs4JGO4?Nl<qI4+@NE-0fB
z_HJZA+KIXZLm&O!YkRIg(zJ>H(aIR=pwPjNKd596+(yOt{pVDSHvJ=WxXpDbq4)v%
z-RD#wv&PSa!-Mn}#oWC}!pb_gi3XTY<q=GumX%Qq&F|>=BSAqqQ0A@%SikvQ1<db;
zWWrMu2TYap_n0C19~C7jQGezgN#`|`#ADyn%G`LO2*odVPJ4+kS%0%-Y`v$VSg+jD
zfW&DV7+`L9T?4YWyrKXH7jEG|;;f@$iW|(?_L2gmwOP!8tPlTHf%FH7g#FuhX(*Ww
zKURP&O@Ca5gs~!s+c|vUGZ`=q=pX@^8&6S5yzp3r;%n{8VZfgL(=;GCseuCI9AXSe
zOPfmAb8#BMkl<3V%IHu>1v1|W%Yf(In3&?0Gx|=DfFo}Gg)<HgDJc6^ty6)Fv8x!6
z{>CB>*lIn~2>tr0fb+~~4y28D$$+tzS4?pU<Ml2okXn9*1nk@t6jQ8{a_`GPRwPRS
zk`t;BQWolO+HBRSVRgi~msZAmW`=U;=rS2N*nFM><XF!T%+3NC#k^sunBrO~=L#jj
zD3w+K|IB0#q^^m|KziR@gtUpv6_l(xUx_JB-#_?e1{~foQ2~6X{SsjLKTr1^?_>S{
zaoi?*6XFutdne<vWzUR?i^xhD`5GCQz1MN}e1G=-e6G(t=Qze2MF<fYiHocvBSdBV
z-skzc{Qi2~Z;#hCKJ(D`tBfp3@R*Yol)RP3YZw@-GnCM|h8*x0KCA;dN9HO($h=Za
zaf|4M{3#H=`V-;6;fRD1cxh9B$VP_@IKrF76bGAJ#=)Vf!>Ks*dMZ}okue3>U9*D(
z*jqK@K%`A?9WeLnC8oHJJ?j|*(jMK`f#}RtDv)tGOeNf9wuHiL#;eNatW{Cm6KW6w
zKfNY6vRW%xMdp1i0pV2R?ygB4Wt8BG=3<Jw=e+q_n4>8}8H(>4;|I0Vc}_=h-z_c!
zk%#Y7@s~_dP`qV|iYczcx)|?r=vgg>;$A&X2KM>O=|E(7TL$oGcY^z;2`Wmc!yPJi
zJS}m-aMSHNt*RB00b8$B3E0z?b0DO5mH~VAdL8f_ts|y5J+0It6|ju?i%PUlCyo-e
z)}&(Fmm;HB-+sUe-_9WHuGU*c3HK?j0I~3GF~yYw?V~CXX>m{iVmT8z;2D08kn>G@
z6-CSXMFDauj*x)Js3-@tUN=?1yvF$P<%)kMg1AM<Sylm(J<T~_>R*EaxkIYSfM!}v
z2tTX9Q9?Tx=m0zM=E|3LGnDY%^9tatJdy)!_d}J+Ur7Z2iiRSHyE{(pRRGhbejKp0
z@aTZ{I4lF^n{E}b%r#yXPoz0R@yyty0Li=lrvh5i=Ms>2(|9j3`>qo~+{?Gv!~sjg
zs}f*7dxsF*+)@N_jCIUx4rtG+6P(HvSrq(9nZbYQg5rbxu18FP;P>xR@we*8sVx6Z
z0RsIe%Yfrxf(``g{f~<CgEAZ?QqZLWxoH<=An>#y2i#ppiz#jqF5HR%!B_7P9G-L$
z#LD^KWe)g4@d}XZJi`F@g<Ue>Id?+>BG=M&z`44ynE6+pQ<DRZHFgDvo;N<5>|tXW
z#XjO!9SC$ON5yKnM+liNNLYm{8Ly&cY*`(}eB^f-@Q=~N6!)@T8!G{x3l;^?*6fsk
z#F56YJh|;I(<*RlBE!lyI!gw^#~KlWn_Gw=E|@#&PaW{(bWj1$sjgy*RWPZq3}h`G
zOC{28ijI;wV5ABJlZ|hW_vxn!mG3^20JpY61#CM{b0BzrE(6kw8*g{o?%pbj-*i_@
zantC`tqchMHBbUP6CV<sGrv($GH+}bQyiSrx|afYR)!d0UR_%UqIHHTgg-A5T#w&8
z<{`!>I@m2i93rk`XS+C%^4k^(@H`r*0$QCjI^Z5NMgp>IuVo;U^-=-n-}QCCv*0zA
z*z~^{O03J9uVcMmB$Rz|$HWvjHOE_Yz#V8{;B;?}VqR5H1;Q1K3x*PnFY$rN^K`5{
zCEt~S+<Q|Mz+S6?nE5Zux%D~_DNu?KEizF;2~8|ZaIv8ZiZwQk0oreERKVM^IR`BF
zYD<7tK3fMorxXr^h9(otW8D&pb-VFmMrubHO3r_MRKRDM&Vk*@O{iG5RZ&pfWot9Q
znL9!PECU+KfTP;)Vv48ZSdqp6%Xq5{nE#!t0<H)3CBS~88^JlEh>Vi6b^r%b9aRYa
zc_tktT$5ALY6c|~>#`CIh>q11z-yW;rnq~k)|==4#Y=|bYZsA#-~;3P<(WN?R^bk1
zRIGMCJ|m_$-FD?O4%inhQvvJgZ4%)AJ&BN0w5W;_4E;^;o$zv$xNZ?Dv9WI{DAC@v
z7@(b4PB2Z~C81c(brw_H)b&?;1#oZkQ3<RxzIvt(`#Dy@1;-fR`DVWkY#BVzC^0`r
z30xdO2#)V-V8tbZ{n`nFdx!D0a!IH40{M40ZTw9JoYUV^37Z=#K;mb{Pd`ilVIqiw
z{SQ86!0vU^3G8qa872K(IR(fpvX6?T`%?)eAa&IN?}x1jk;%p{W$eQhDptN>nPQ4t
zXp@(7z*6vC8E}l>uK@1w1~J7k-knJV)~Xo6Q~DDHt62OI1_X9&qhh*#N=I=GsYVFi
z{7gm>f6<r%M1EPq0C$neIuLg4Ab2Ofqo8;)yNHRqFX4cxUp)qxZych+mNh117l@az
z+RrUiVt1Q~AWpZby(J)Jlkw5A_ZldpgzC0sfOqK?1+WzPoDdphyv|zfk_;>Rs^Jo#
z4fSxqlXG67a{ZYM<gEEbOmS0dopKVO9qYt^9l!lZCG|_Uj$(aKPyuo-&K6Ug<!M_(
z0&JEBgkX(vGKwqxDg*3)zM`VN`M(@(O^b1?w0_rhz<r{S0_-dsrvkw%r8z+zC;^r}
zjTw+J%J}N}uYF3Z+*0*vm3=5f#W8vP8xeA!b(T@gIsGLda_b+0>3mHQ#Iw+vJtf3%
zZb>VxPemCk@4D7htbZ4wVmfe|ifhC~afmoQ?{aZkCC_W2pzPiJ3kS?ww&}obeK5f?
zcm{#}zFWb{U2C%pIOeR=0spad6)-J4p#XMXRtA!*xTtt99OEc{+XjLuA<9rZ&-#if
z9$2`j@y~XVCH-iXH1!3?>cFT;R6Mt1g!J-93E{KX2=>aa2{wCOy+Hm2ci-<!C40#-
z8O2d;xdK>^#8Ywp-CKnGyKnkM0$d9U$pF7`SO+ZCb`gBLjepQ_e9?nrrOj-w0$FoS
zgp3=dIEugESSpSq3l)@<_Qo$;+JJ9l6#pNE#T0jUZTOu7&ZjpeAh&r@6^I<|r~uw!
zZ$A8Ej*B49Vuw!3fW3O60(h?E>OlOg@e-iDI3NS&v@JT|yn01UaeAmiHx8tG59$Eh
zdRqlT<rj-7j&W>==Rj@|<0)si`9VjquC7C?u(KmW*<IoQ6>ryfWfaFs<7*z=yFdhS
zL37+J2KWv<AmseMLj<vME<2_Ic}vT2AUx-_4#dpgQE|VHQPH}tR<ZIu{X+tFx(X|R
z<E-&wrkkcy(JC9)VV|6#vS-sX83*qx`jB8-VpmbpO1&0SJf56zM-G_yKROUieL&Dg
ze5#_DYkW)azTaIz3Dz4Xra0Zx^G|~H<S>R3{dO(~LR&V7nSTshC;`s<`&A(L+eR_P
z%KvhK1cX1%BDhN$4<onmJ{>FD<tbu{gZaJ*3`i}}o?ywGC8L;J?R3C$IxGRvxyH}T
zP`BX<iZ^+vOt?@|OmV@;j~l3Hxf+$wJNpS1&jT(bez*MF*D{dzZUut(&MySk><YnF
z*CGzdKe&|f4u(^1a1`7Bj89$Q`+uoem7i3stoAG&aDLi=QTcU(O69Np3J`s~mIJQa
zZxKw<Vls+*?K%m_8lR;Are9ZcAUHo;0xYwQS3dQ_+YH4zYmNeFuNQJ4^ld8@2p$<P
zQ;B=T0ZZC72}nKKKnBdWcM$A#rt2u$%p+oo=dP{nqymwzKBwY5c#?|y$u^FaeRUlM
zSUxgd-f+<tv|@L^qm}cbE*HqZpe?(rm;(Onu@aDb>Zl6X_HQJx-xLW&tD`GGU|eGz
zFhze6Q=D$CYP`!ijl2?yMX%3+=wCmmK={~t1u(U3NjOk-poC&Qm?NgRj?4Du1JJ*d
zLaUG4s6a}e!aCsE)QbbwzSCvEzUC|;{=cdUN_b;e256H_GT@!_wV2`-v2MF{!2Hc7
z8Ssqc9N1m0vI2wyZ;2@m_FS690qyl}83<PXi~;8V6_f~z&&WXJo0kfqeMJ@Ubm+^0
zXo{5(e3(ekZrzozVwH1bz~P&v0NVQr9IzA|DQ5o5QlO#&c%M&Ufc@MW9dNHKM8(_p
zf(YW^ytN;2z&_bSaD5ZiQOvDU2-aq+6_hkBTLJ<Ley8Hv+=HP6<E?~fiiKc3ea683
z(Mo~*d)Z7Qs5svIQ$YC6K#rCDv(Hq3jWpg<&)`xrilw=q0il;>9dL}R%K>}NUt)?Y
zn-*vc@HyX7fk53ERAQeER!~A4PEm;~uc#<-b*^&2ox4H;yp{7*K>Mo-0|NJs%K%$;
zNdi277%!ba{TGhnXmy-cIj*l|6nAw-h@IKbQM`Y3CM2DDs-k$Ne<r4Q@ViEQNbvSO
z&QTnH43>fLpeIyvf9Rp3_*NfM0r$Ng9PpG`&VX2XNlbA;=d@G?*tU<9fQ+uc%Yf(R
zT!L2WqzL&h-P_FwyNB$jRX{i1IRCH4=g>0Xv4Vr69n(0F*<+3hXp3%fAabLG0{EiI
z1arY1I*RvjEh@3u#+NZP#zMvR{Sv)E{==}O-W5~8)%Xk{xbQ0-#oqad0z}6&k%8cr
zt|}0I`K<!@nysed8~+}a+}nMq<l67(1@f;PY_dg6fx!K95@5SuLI%9=Odv$BR%R%{
zaZV0ICXZ97-0H#rZNq8_*nPT~3IwWOAo!og6LJPUCYZV!Ul1>UD2^3(&sn&GkUOa!
z9b+G~O~T5fRwaa11UZU(Mp+fGuWG@7^zLmWAanF^F~tQfzZsuQ`?)S0C96Xj8OR>d
zNCK=iZV+-tbmb^Hr`w4sPWNmqD*@sAb2$*{;!**>l`%luI9~$X4X){c`I_-MYmJQ;
zDtk%+6|3x$O~n*fcI-MyCGf`(1;u^#12M%at79SuJT->PfNA6w32^kjqX3~BJ5?Ze
zM4}D^zb#1cl}J)h9IbW}th;uoD5mGeV-D5MVJO<onGz7_bwvgCcxH>4{|Hx4=78lz
zZv_a=NmK!E&C*oTXA~tkA6${K@-?f)0JG;c2OQgsXBv8(Afp7M>r^1O(-#bIPVOrK
z8Lb}?LXnRZ6!(;aDqwoJUkA*UTTls|X+&`BWd!%D`ceTr_&2|i)(SkP66}{naBi8*
zaE$kOJ1XvHpHR`-ZWkNz+|v@aalp2i%Ru&@#>3FcsVYkF_6-H_Rr`hDur4En6FW*+
zY5ZpfB#D1C00LibmH>O%$~qAK;h92sX&mDnvPng84;jP(>$lw`!2I7V74YsIFQ#~?
z?yqm?fNkOo4mj>lQGvu6iDHUl60$4HKs4L<7+Xe8rImIpiQroh;yA|jzj_J~wZ{nB
znjH*f_r3%U<i2<&rns_pag7ROt=pvlwtp=W;LaJ(0qs^j9SD51S_Z6Fo=Jcu-L3;s
z^`@BOg4v&}RS2^2iE^hNQBh)V{U-s@4<;#qeaiO)ZC)P<#s8opA+1q@juO@C5?o2f
zvv5B8iH`A|SSt<@H;u?<4#XDD(*bk6mV`)$y9~t>{z?KI!|&>Vw|;3c#p&+BM`XbA
zNkj*tx2mc@NHboR;Hgz2h=X&pvN&KK`&tFUPwR*&R{pI|6u?vGr3%=e?2rJ<nNMXP
zGOZN{GUi^^0rRZKR6K)TD=2weC#pcmoFBxMJ+V*o)vx=y#nWWK)^xR)qSzb$#ejr=
zPf%fn4^Z*vzNKJg?orQJ4Su1c*!P5~II0hmQP{`nRQ7yhyu6+fB^Xwj4~FT0Z}I^#
z#Z9&Uo^rrmI*|d9PJSwBO&7~3{&!9gB5zezQOq@d<be3o)`ZkH!)266(z`lfGaXj|
zZNvqFcgTMPhwqY%RWR*a6>!K|V&*@dd#@#cRq0Lem%wW$-jv<{<5<~#>cD_djjJ+X
zx|b;dEN+Mj1nv!_5{fnCD0xNZ%0Q^-BpqP6eN-S)^t=KDZdVsmJPhxfe|6^YhsjjJ
z{tp;d?6&bsiT6lV31z=xd{YjX?};EzcmG&~R`!#eki6&`t^A*EQH8|Y{qY(F(7H91
zfSuNLgz(|v48{AAlZv)2hv41bUB)U<?x2|Y5B1P~9ms8dO9EV7_bWhTsqquZv-ydN
zl6B#C8Sr=dkr1eLKm>6ebH)NH+VYVSmHlhQ%zrwg7cn5nXXt?G-(?bDyCGAt*+wWR
zq3mi#sSs3Av@I`Xz<r{vLZ#E23(jcBP_{l;EN1@GnRieItX&3iAlSaJ4(u+`mJt59
z7e{g5Ss|u4D=+qe3OH&HWI(9+4+>zJvr^3bXYue`LVC3_5=#2*Bn5D6tuF(CyUWEC
z2YdQg;DFt^LIwiQmlAv}BnibmBb@=>8}r1>zs0>Q35e|cf)keRmI3?Ey$w8YODM4~
zdJ@8Q{-Tw;vGIb(Zd~U$#`C}WI>6%ks(|bERt`jL%?ZIu#)sC{=!}k)&$e6zv=;3M
z-o%A6if!aO1bgRyIf`@ZQx%9Ef1iqdM;nIXAN@ZWh?Htdu>bvl5ZCLaPODzV4<g^s
z4q91%aS^<Cdomp3Jib{6vS+`RfpD4265y%&3zgleGDmT5{+a>i&Fu)5$todaH{R~l
z<r`HT6Ks<v1Fnk084xgCppso7UO~z67m$F!JDoU?KHVe%9_tbg?5%uX2ePlVprSd9
zm(JegXDYr7i-co*k+)^QTjDB}=$Bnp6lUqC5EkZ838@Ell-x(ZaKN<U0RvJ>R*`}1
zN7Ka2|NJC{8Q}kZ6BSRZpoEfY>!$;j;5-iGsl^q*(Kspto(X#;z}+w;rnsQ<#V-s9
zAG#p{p$fGKY4<)BK^$ZAt)$}adsIa+w~kO@cPCKsK5QXlmAG=InBpu;(M$<&pRC9M
z`_mo@;A~;MUx7m7Xcb=el40d8wwB=UHAzNEZKhGNwJ_dO*VBb6RuNM`0ldqG=zwiq
zj7ri&zX;-Cq}S~t0gm5qP+^t+;3)p@SE+!baRUZu^B0Pl|CCQ(q!QcGMnZArHj;r*
zu}lRB{<WA>xwurO(zyhq(&PwX$No7I%I^MA6|lQ~46rurE(5-Sr>R&Do~4ra(9N;(
zHmjxrq4jGS5U%xD2ckPV6P#^~r<0RgNo>S}*UlFtxSTB{6ibCEGQd)HD1dk4hZ5j&
z&L<oyvyh?W7V0CWxQ?@P8VAfhrxG&K`l%?P)VgAdV}h-BF(6QibHG-<l?t#q3njqw
z=XM=9xHy>sR!1d*y?sjs#Xt6*4tNq~iJAW_%3oCg%jud7aAiy+_zJJ&DE`fV5wZ%l
zRZudH4U~cG^|y5(SdlXzW!@etu7}3E99_PRW92=6O932D7OQ~uWqC34AI49ks6@Y<
zDWkY|&EtT{%v2!lSs@0vUS8J)^8ciae~V8>i!S*`OmU^ag?||k8MQ|Oa(^*ClEI(f
zV<@ghQx(9}<FpQh4lSYLt5}kwSO%U}fzT(7s00o?7eU-&N99%uU_Jd42kdc;b--0N
zMFO(!K2d<w_m+y8{}kWLmVi*@`U+s{-<AXHyUIFX8u_!B;$Y`nS2<zX=L!(5Qcp~=
zGM#a&K-M?46(IfT83}MtF3*9S0&`R#Y-%W`I4e>^mjKHUE(1y0Q-Zn8ClX5N!F~or
zT4`eD-)q2V9nkE?<H@=>Lq*AbI8Xw7H3#c}e{qtS;w+zYrV7|HCh0(Sx7%WhRdDDj
z8OW;fy9A`pex(8~eu%)f7m!i3O%Vym*;1AP(Ha}Y6xZ>N4@!XR7mZ;4&G`I!q&%?^
z$E5z3$bp!0ib~+$cVrano4>9t`iEOb$t>(r0N==Ws6<-rp<*hspAdX8j2FP&M^}=7
zNZWG?z%K0-Q`|IGXN?TFT2<wM?bR|d^B?m&Z&p*cswn;$3pwB#Xnf-X8&=D-D$NM4
zOJAxe-q>0Ra6hQ31A!0!mH~6(DG9J^?HRz9jgWy*nScU#(+=uDu+2X*5Ic{rk@!tk
zf|H~8<i|RYGV^N{h@~2jhwW}Ep;#MUPyo;3F=C46<@?pg0qaXMl~}zo3W}?>TL!H6
zhB1J<yHas)swtspe2fa1M{eSP_xD9QVEwotA*b9!hLSsLl?pgF3>Gv0?tSmeKvw1*
z32;db8IXSU0SDYOpUFUUNxBXs4*x^U{P%0>O9e=K^WR8HSY{lfO<KdS3O+g}0ZB!3
zbillF76(lAXA^c*Ys^rBmeyj5E8ACXmH_MQDLTNOk5mc0jh~2)(c>73x6^cjD|4xe
zl4{w>0q@I_1k?GW5{hrE@rv!gQjUs$Ynp;p%Ko-0VE=v|1Dq)@sf3;}9mV~AYXu0@
z+^qr`)jwrG{BJL1z+7Xh3S`uOECHd1ovAoJaxj$O))E}>)hVU|mNmw=DaCwJM#=bm
zl@7SC6{X_q?xEuSb~wk%k$p`CoZX%(Kpbl=Q|Vtt1p+y4F~zIu=qF1+^uH7;=EWf$
z#onYZ1A=om5dzna5t6R%)v@xIPUC>BP!AQb{_~n(sUBr0{xX-v6!!`S&vL-B&v?q|
zMJChAy8MnptA|!9374<RD4{H$4rrU>#LT}=)_W2V8`7Ku)*X2Sdw*R=@p{k8Kq?=s
zQpvQ7DNeUOF<u|j{=qtmduvn$_K&L}1L<G(W`O1UUMi5#@|l?8blb{O5|G$6r~}C}
zTgZSVR!dBA%$|>oFNp8+ZYqjxizES-$#oULo6t{8ad1k9dJ^FKEKDWi^b8dxoY{~8
z5yuJ%2u&I&rZ~&Kzahb#{)vi`w{e~VgnBHeVrjToMhTq0Bmv<oZB$@?vlJO{#~a@-
zSF_s+O5of|3DBzblmWJ_iV9f$=NXkN<t4ze{I;0l@uWMAPm8PM4H?CKt}4Mf{RAOY
z_I(vAYfc@4c2;F5-bPsxpnX1*U|Bz&;MsLm#md>YwGL<>43~h+oZBi8P>WNs-F8!n
zO=}=F;(>*>K30Go<67x}<-e~PV5@V81F=&-QSsYtBILik3y!NmxP2)d@DJ(90h>C2
zN_Nxj48_&O_^1a{s?o|ZeT#%u<jp^-`hqhQ6m8=wT4}A`<0!FO#wUT*?W>|>I!ZCX
zwK+h=UF$g&rfrh2(r!#4xGyadK|E&Lqd#OIy_@liLgtX4B^2}1OJa&+%%9~?0pDD|
z1Z2)RzyQ}MhXN$7{h3N+dpQ{;Wyfj`*v{P$Q=Fcgc96ihLqZ8o>MR4E4t5SC&)-U5
zKb;mqoMo-}OaUTQOUpn;t3oPZ9sM<x;Hh*S#kF~b3h>6`#1yAH2j<9tKjAS4v@V+!
zAkw~)nBo{)d~ptVA2wA1`-PYSIO9r-DUPv>c5r|d%pnASx~8JUMz>~wzse;Yh|^k0
zz?!`33XuJ3Jz?vYKdLB!zm4y^rG8tE;!;N|K>VITgiQ<Q6HNPuGOR2)U3EYk)k_9!
z>SHS5I%_$KqtPIVaBL{SmwQA-$w@3O1HPqAIFS6+0Uhwh{YS-ItEq&N+kY$rOrL$P
z15O|3fW34hDv=^L3A<htAUMw(ud02?FjgS{r3-xF6f+;czmfs-tXCYcE>Xl3D_>@5
z6-cXcRs!6YH&U_hx<v@zy(MF%o%f0<&T=f6qyly|Nd`7I?8||~f{6^Ux5<_O@t>a7
z0rRvt255g>mViW08yQGyX(o6&JrzORVrPd!958Rs)&a*ezYKW(d_?dqf1sk6PhAvK
zoE2L$QwL0wZpwhWdIJuaBB}z|M;nhhx@@S968EGw!Q1jOLvg?8C;_48#z#GM=pl~c
z8}tbSSgCDdikoIHug(G6-c35-KHh`M9@a)eu~(bRfI#qH1+e)FtAL|YDH#aARg#M7
zuN5MQEBjs~NI+7V)*MJ*yqU_bI$w$)jtMxA%YgIzYX$^uO>|&)^qLH$M=mLVbwW)Q
z@GULNfauQ~B_LSaA`=u72f~*k3J{w7u>@FJA18!wR#8y22{SnmczT`?e!fUTS(jT;
z25b#(D}Z+Mg#=_yt1eS1ZdU;72!#XY<ZlVSuQsSCvC)&n6wlOLrVR%|13ODVxTo<}
zde)tzRiLb*Vij0@MF!Fuz0?6~YET7o+bRq&*LWqSxU&5R3kT9NhsacRO`&3+wvmcs
zOfv~9uU1U~Qn$bPO^&{gO8Afc94pt*R$_`Pv!9GB+rBPA#eaU6idF82bsR`>8$aTl
zH9b0tf7>Jl(8je_0j3(SbNJ0)yXN0{TE;4^lZ^wR;6n*;jXk6SS+ABTgjFBOK%nb9
zDr|k6gu)#EsDN#+hXLlemK+FvwO1ijSVhI$Zk7n*d6}GhBp}3nRD$ap=qLvU{;B}l
zg}P#jgDu~lBKW5c&{4F!7?r(M5;=-_-2@%j_O<cZ2<&>UqQn;Upp~;(stDqOj`0o|
zNEq=pm6QdUI*Mi10R}j`To+Ru>>m?SfbhZ<5|BKo2o>$vS1L;2k9{0)EGj9cILmu=
z5|xxf_Z1XRmFiU7b$u#|YvVWt@RvSnlzB@zN>XHxnEB7W%3?xjXJ3Y5u8}SS-ZK{@
zAjBGSfbG_0Ao-mORLpmVDkxfmb_CzTQ4Ga0z5x|`p*AXtZHi9Cx#BWM+1qxy4p<jI
zA^6_>yB(%V4<xL@^=gVK9%`_sNd|I0`HkR8Fzy~~)rw)2y?D6_xR1=HV!8KBLh(Gc
z6C7`!kWoV8y<&<hYsatXguzY~NII3n0Q-M#3E0u#4k6OBi3sAX^ruH;z}K~j1o(S>
z#DUO?`UFpxbOj|?LLoS(wqq!kDz6B(4NeKg{pL@eZhd4^P&}h+=|JS4UJ78nR8j&`
zhL+-hHf5R)glY_x0aNW>5)hqe{D`xz7{gKQLuwG%)9yNo<<0-P;u(8TLUH^WlYs>N
z8x=?{^%Vm$>*^AqecVO|lE!W(Xw^C?DCRov%7ERsg9F-9<8}5nIwpd6>1;KMDS&%z
zGY<I8nJN%F^qmfP7qlYS&QuaXoMr0z2f?<pfP#{`?<NOKUgKT%KOV%WeCt<$NOlb|
z#p&)}#t<Uxl8O@QSwaG0C4Xc<>e}IAii2JMl#v1LY7L1{^IHYT7@4O7k@Oc55Un<v
z1Fm+(6+-N|1o+1UsRWxxWt3<c<D>5UX|IansXCJbwz7xC6nEElekcREwjja6N{NvF
zENUCChI{=)2_^FOG95^*o23xyETv-aXuKNcqVY0Tp3UdP6xT6b$>cz=*+LcYG#pOw
z9$6@(M5|OGc+1U@P+YQ?%Fe$l>L`xIeF*Hm4@D4H4(AMHfO%yX1=u;@3ke7oG~N{V
z%-bS}gOf+BBY66ZQc>cLwUz*D-Z33;{pw;=ie3{_oaMc{TnB9RIx0XoqW}W}KYt(t
zt~Rf9z&mp)12V4+7Bl}5PJ2Ocv^8ER_T^a}tKjINDqyKNiICK4IzvgT`d9)YH)kq<
zW6&Bg^B<4$Cl&h=<NNL`{Q<+u+{GjViC<k6QyjcASJi>=_bX&Tdu)6FJh{6$inaR%
zf}?s<2_^ZQ@slmRdszh~7~9H#P=lT-;4eE*27>1wQSrXLL_x{P%chbXYbT><D`yab
z2~!l5==W<m;JGk|0f9IFho^V;QyIlq%y?P6C(J5JxYKKbR`!1!CH_zn6@QiL1nUQX
zC|FtM6ytznE7yTMNl^gzu51I&O)^U2TlJ~f<`0xmJdNL`V$Hm#qlmw2l8UWTNfGkj
zi%uVMAn3oW1J38O7!X)=g^IseJrTsgzUM_au-&v*1+;-VVv3bkX*t0icU?y@PgtY^
zfij~Nz;y7846y$Fs03cGS5cgWpG$zZ^D7Qyy!%`SENi+dfO~Rt!Z!az6(!jIkq%hS
z{lfvRkMX?H#@|;_SouO?ibrUbmP^3S;UNY%3w_Q3_P8+va&8#!sdni$7603hB&@<;
zdUU`!<)jK^e3M8B{*Wl6SbCh&0mqDsVv3vE`hCIx*U_pnpe>tD2oCBmp#)ogp#Z+r
zX@rEqWmS~m>5vRW`iy0OR`rz*gjz2mm<EOk{ue$KtK^U8%Ya9zO(m1XM96<UGfr^8
z^<Q&|aH)@&V&&LSK?O`*a;bzawk6np4aiuThm4YdSXxOHupVD41AIb135cIGO-yk?
z)A>KBq`!EgQd!P)z}YH~QyFZ$xu#03C6rLVIbw>_1Me6=PCaD?%P2Oh!~xe&Z6zSP
zj`1_W-SHzTo=YtV*0+9@ag1wPYX#W$_AwQ(rXA7&bH5-19H;)3fxyKtIpEtjO9GNE
zHZ{taY=&YQc0>WJPh14cl(7=Zj;)dkgc{BhQ#^~%+-3yd?lTgKwcK?D@V1%E0N>@$
zWx)M>E|tKMQYwn;{ncWM>(~N?C4kK{evUa^W`$M<-lbJ~-U<~Z?V<6WCV%`l73N#d
zv9g)JVnAl=3o?+pv@5|Lt1F?TH%Vqd+};S4xI@Kt6kkqXF~#GFsYMB)_m&Zy+dk!3
zg<kDY0Q;A#RKQw)9u;l&uR2O_SvfJqby&%P65u~!JTG>336=C}6U0Uw<Low72O>E`
z2<9tgRTNjbnN-4kFENy0ffF*oW-h0qb-b*jXtVoL@jJgDI4pmuSXtiu-H!fw1w{}y
zO-+43tF+`g93}Xy5y4w{g0U()Q^v})vziY064y|%WiKS`kt%VlawJ^^qO66O;ucvy
z59dIn^kzbG<MJ}fy7%@efMdmfI$?qFeRu5JCxSSgFRG#fR{KUB@HM=pP#O1<16d1<
z$Ky=1>L{MbdlHb|sV4{g2Un<oC2~Us?Dir$;BTBxCHPUMf?}$3SpuRnQW)UdlS3uJ
zb624vD>{(1_Xw4&f-_{4!0b6H;9NCa2U4b85>q@KTi0=fjLBOS6qf#s18LGnRCewd
zMI|=kp@fxfVmB3Ft&LBMw|_-OtF);KVD%ag)jRaLgp!wbL`-qhQ1~wy(59YH0ef~U
zg7Z;2L&<t&+}*Qsvk3VQH7!pkB-9}UyX;p`GUI<D?D}q}grZeaIAChtRR^N)caQ-G
zuO+6qV0xL6grK<uLy5!}r(#lyQ?ai;$+2>$WXOPa;sF(Qx43~hnY417Zd0*W-X|5v
zzlHgqwzTrz?MEx~wiYT@u9Ke;vdaFaqeKT^RRHTmj|zmA6qSKUp94CORVj%Bo>RF5
z-;ccs_TlYytlXtaF(7Tx3o7Avyo8K4_f)Jx-QSl0Z_~bFiic|bC7X)p?sOH!J;g^Q
zdr$!?jtWl|tP-;pb3k+TRsn01C1Q%}u+*Cz@D6S$0jBj=6d?S#m6+m~z={%79NRup
zQS3`bbHFrixdbF#|Cvho$yZcdzJCl1G@ga$msPAl{sn_`Yl<n5{l$14aPfEsnBMJ9
z@HCmIqF9zi2wGR;R~cKvG6^fwJcSTk?WSYG>;F-(a_Kdw#4e}|MSC(=0X)CnlL6a<
z(;To=_)j71JthHc+6@k9H$P>7Eh~p$9eh|v2@Ly;16h^d6;nJN%Ljc`z;ChWRJI%w
zQ>?7riz!rY*O35Q>DyEyAC#vOF0x0*%JE0C3gj)cbHGxim<+_l0y<zFa7hBZb8=L`
zw9TUk<o`(-^M-ap;<m1r@f={krRzX&?=u5KzZO9p<A}ac0oGzA2a>*gb8x?6RQ@-r
zi-eVR#S<BbJ=miH?s6+6z%n*b0n(nti7Bp}yY_!FVD@j5fXK;u3ZT_WQ33M<rvz9N
zCMiJV^Pmi*)-qnR;6SGc;)3o6@jBocW4!j3I<r+2)1`N0!1Ccd1&HiwMhF)e&rxj2
z#*Z0yp*cEAo@KAGTHKuA9e9RemDG8x0$AUlA*Q%H%euyZ1FIYAfVtz3Vv3c=W_%=F
z^A<}ek@wzK0Pog~guwW@B8Y>%Rdz7IIrAb1Vg(ONK;YMl3gG|orVcogTXMiwuN4E_
z*PatX1K-n8Y|no+u+Af+WXcyMAp2=k9mv=-l;C~Pouk;i?G-}ogaqWx33I^QeTz=`
z^P~)DpH|_3zjjmyA}m5BSW{9_%=70{F%_T4P<B;+A*OiB|Nk$9JG4BGlDgNf0BM<V
z42X6fr2=f_n@_?Y^(2(U(+)Akb<(r?Q1R#7CIknJpp||8bS@<Rdu!R7RKQ+xnwa@7
zc)h6-V0!bHhy{MzBBOW;ujPQt!wDAa7@bP54h%@zn@2_a;+}-!(+;YD!|YUm-1YAe
zvMVl8Q9^H3C3ufEAY{z=n`31zbW#CQ%B6@Y9-)6wbp`}a&LB8yZc|Xwy7ZHPj9U9S
z5c#}61K6t*DiFIdfe@;lE2E@`2T6cw?qe~<Eh1}1%YeB?1_M}@ae8oPk_h6MV9_K6
zNZb4|2Q05vserX}EeS|`Z@dhI7msCtyIwpM*ZuVpiuSvm5UlLhQPO+tPyuaIL4x&t
z<L9Wgj9bErZF@+FY|170R#sH7a+H6Mu=|4X3r=uvGme$%Qkn{AA4^m$HTzLXKF?&V
z0yFyR0PFF!3b>|wIN)Am(*aNHeKEz0khAHU4A?I3Re&uUo9KYE(n2xCG4@RzWFRtU
z8zFf4JC0&WIH>^c;Vv=7!R80Xw;<Z`nu-#z^d*>H9aK=vV_gzpZPHk#vZ|Aq;`H!H
z<8$cQFjYrMdU1vW{yLcq2%T6;up~a`D2}PuWgyh&0SCM%H;5@N7|Xb;0!gP^=|FnP
z4Gf5#{6qnwU5DsE>eInuinG##E*W42=IKD{W8)pv7MEct*)w-Yz`psT#1v;)H(p?X
zck*ElMBcFyw(qJ=a3?ocvC5FAGr)HLo&vB>-74UD)Qtm~u~iaaYi$x!+`<#wC<CGS
z)f9j`e%ArFHkt#za|;>Z$x4#}Q=F;-wm1JJnn1HMB8UrS6zxx|z3=UiQC!Qv7gMbK
zIo(yjyeNa<8sO5Yd{;{bLZjCaQg5`CP_(bR6IlIk6e^qFl7Y0>TN#k?sd0D9N#oTp
zS2I4?&XYTJ9Ao*S7a@An_!gw@s-a+I%QZd#wm_WTCt=tb^aU2u($fE+WCVLGu
zOJyjoO*sk>|NGYx;MOmRDV}oH^+X-8S;oo0?uqpYi7i{}D6G5jlRsASGZFINluE~C
zz+3%C9Z0GC9Rsq%#uL=;e?cWP&a7Z%?$Ch{X+D9Y#P>G7aZb6VgyPE@tOMFtGdK{q
zyqf`xhg86Fc^C)G#lNA#mei!;+$AYk1)S3ukT>p(PNm>KRN}s@qN2F{2dG%@m*gn6
zg5OGjss7grklNC#0?Ea7D&ef5I*RQ}i2)APcn59QK9EsdE$a}1i=qk&EBYM=_BL)v
z2s|;%D85|_bs*>EHyp4nic<kcrw0<?u4jBQ?K6D}${y<%9Png>3654ahT^?bNCtuf
z-WF55UpwCX>0Q&Wi#dw(Kph2ekG&|SI417qAP(5qSELf~wNOyJy#~uba&Au!SiZO>
z0nUW$3}F3k5Tb>23B~_=ODaCiNpPi>r<H3&L2-z<scmQt8A#6_OvUu8@e6e3q5BN0
zjI(wHa31L+1J>JnIgtFUodjqt=87pUnA2{$0{Cw4p<)izWhjB%?^GaKtfc~It$&b#
zT+iQPiqo}`#&b8tyCjs*<(VAt{k>8Kg0m`1fGwk;nBuHxp|2T`_F{knct6OLfne#2
z1Zz{{o8q0cSj8&*mGQ9&cWEl2c$z<9fIGe+!Nq2A6pMKP!I8T~Me%=jUjeMz3xc!e
z1Q{j%*ZvI9te2@c#uwC4V%4@$Ik@B~6;rbnR5Zs8ULgPJxDNygks}ca#rxGN280f%
zVv1u-vkNPL%iBo??8iM4kb3k74*1f(Q-ENx)+%6`epO6ydged(38u9TBotfQ7{R+_
ztBMkx-;rQhoGL>8>pZ>(mC)Z!IEtgpk5udjuQC*u(whUegI`NPa8emD#p(X)E(Nfa
zZzu!qv=U;9m38_O9mviK6Rfv3sVM9(6P4h;{UV5icQ!jM0l^B1Dv(*eh7Q<+g9yIK
z=?o>f<qpBxC5NLpHd<A{+}HS!_<pD^g1B<z&ICqylA!{&_(6oAU6xR+1y;#GFm#v!
zISYLfVA2nBz#c531OCaKCBT`ItpHg+O&3$#izTIVVCD6m3gF9>bifhaC;_47rwFc-
z+c=7EY!e3f%atU!AMTV;Y{NcM0cTAyaEnG&WFT1i90wvF^^t(U;rFS8w(p=~^E}tF
zvR>Fg2#+hnP@MnlRDrY{<MU%KJV!<evRx98x3Dt<oT)3QIL95M5_|LaY$nY*ui%({
zBLh0XDve=)`Tiy<xvdjL5U-DC7M_OqA79ICR#6;VI!Zvw@D@5?t+Ag2_GT?)Dr*Po
zfMsHy3S_q`F9C^P=~Nu;izq0rgsL2{%|A*dsO%y5?x(6)1)p@00MpvVGLTi|ODYN0
zP9lhhYH8n!kd^MIl`r!ZmE=cbxKO|D@#DuRfaQ`^2CUh8Rlq&c_{M2RLlTO2;CO=L
z%5je3t5re)f{ptdCA}aOEjoot`l-Wuf&6>flRDBW@L3sxcc)dt%KhU21~{(m(SfY9
z=VibZ&{QC*bl`w>Xgw;ST1Ba3J#RwAeWbiNMBK|?!zKaA3kFh&B-Nzi*xQXt=s`bO
zNc?7dO&$Xr-`lA8st=?RDg26K<t{%$2ZC=GRjBkd-X_z?u_EN(w5XK>yQE7hkThii
z75n5(B8X$muly?DO1{ql=j%!YQ|zIR5`4c9L3`TTD6Uk7m9^|X70Bs+hXdZW-6bGp
z;CBR9>s~raY?NIkEYTFeGUTof*b6=&M1Jleq1euU&VlHNJ2GIalSn19sh9}j!Ta*|
zG9ZxpEfqhDQ&ED4T?(M7eF*kbZjNH}PgMyE{$Yf2?R6k+(rgZRPI$!>w{VuesRNb|
z*2_Sut)&FS=4UE^Cmct`_3k%R_V4<eW2M~-$OQib1&B1g#sKr?1RaPv|Dcld;E9A1
ztWr?{T%(6@DxE)(38#Nl0qu(lVv0u?-`98%;@*16Q33@!64=UPGD`61Eh>&~d#Hq>
zO$^k(Oe<EvEfvVWVDin+2$l+?Y2`d>e7n=WP8WxWV{*EU(Sa0poQn3!_z`CbPGnel
zI<?|J`Zu45Db5PiPo<(gf2E+<%g&I2q(k#`D$fThfX!V}0!)V{s(^1zGX+TAeo_ar
zHXR~_GtY4p&&3Z3K7AEK2@baEfMsVf6-fA_5)~W!m80Z+TSo#MO@AV|&h?d1GKyaz
zxHf)4@ci!qA+oB0QXv0%x&OY!fbg9UIADHjx(;YPn<{{*sEYyCNyhif_ML;H99X4D
zfO*mr2H3lHqvE{&E){KKicG5wlWApZeq2EbG;_&-W87K}_!}Rk5|P)4ARgh4j|%I6
z>)Y`vVE5#3AX=a&A-8RSp*T<1(E;s&Ma=vsm@z>DtY6;Zz?uy?5@0iR*MWrj-!dRD
zq5ucNdz}*CtJ{xCN^j#=qCmF<87p&KWeM2)u!fl8Uf!q8R3Mycya>So=V|5irb)D_
zaasm~&9Cb~-Y%v9;i9=5Fg3`N0r!iyb->fq_?~({G2RsCqdyd^(jRgL*la6wf-gug
z_4$*dnAV?FfweE|D!`5gDj{+^hoLw=EUW^iykRn6Ydnns(Sx%kz`f))2OJ-MLh#%-
zULW5x<NM|KeUXe~k~Y5if%ei##mcMFS=L_jbsS@@--iMA8+j_=TJ=f?EN6W(kWp*1
z3Ya%dVSuCSVnXhjk7X3|r;~KRp0Qe@vNVf|?a??YmQ{@yR?&i;2(I2s36{^fR3QHy
zG#9VQ0r#|j7~ojZN(Y=5Ux_J>@ppTu0=_P<3A@@FzlbO0-4`2ijAz`p9I$zgGaxDR
zA2G!$uJdauxv#fUNxpo8O8DwRNhI-~5DH|dKy+X=9k56{6d>4hBqtn**8%S|MIme)
zDFIstWplvRc9H^Q{;!Y@q*WLp1C}J?7S=T@Rg^sz<A;NJ&P@r$Kl!x|_)NX2@coY%
zO4g7nVv2|AZr@e~0t=gPz_a5e!Cl{YHKI?qQ_((HBjFfplioUzl5kc5+$lX}z$4XG
z0N=}PDiD_U%0MWk1_zu4KW9LAU>`BX!_ao$<$$Gcdj$w&yJaAnmPoLyPFGRfEq)>7
zy=|;~PcP6h*(n`$9AlrIEvC5g#&<Sz!2HPy3DCxVOz{3Ni{Kb`LTtprzKb>vI39MD
zfvhD*RKVVLyaG55&!ZArzeNObmi^D097rE;JPXtCe`OSZ)BhB}J>B?PW#0HrMR7K3
zM=&KHlTg^h=Q3caP)!B0K5r!fftts~6!+S7dcF!+s%%#PPqm<!V)g%jUfx^#HI<xg
zi*&5K>9<utoBC7+Y-c~G;-6DoL0Mm;fDD)_9@hbn_BjJWqs%Ih`RN1+2p=j-@VsA4
zLGcuFFu*eZIhFA692sTZ=7xlwwsty7;NEyK#bb6is?LGb{V#ODd2~EM8@EwGS-;J=
zyN7p@QSz2-5L29<vT&>lcsretfZ%6g1+djvAf`Ab{Qheh$g5gO0X)^-(E;1YKUKgb
zA7Oy$+)f$r+^$1~wYaCEgx2@e0mq*msO;)dh>G*!J+Tq@GR@e}fs|g&WFY=nJqfTj
zxz2#3r<GJ7rOI9^!S`OsDE4JnB*0XvBnL9TTf+z!iYkET-0xKKl8ndW8Bj;UD%{a7
z1BaSjWPodb6AqZ`T_GeramgswhE5&uja;b!!I{6SfVpTI1ERL}RI;t_=~QZrV}NOF
zKQYBqwtoA713udyIuPm8NKCP^Bq<D7m++Aecs?&I0gmitVv1uzr@m7G+tn`_kWkdg
z0e0nz3PeUu*9k{C2h4N43<%x5M8!T)r{aCPnAqe$u$8G)(%QX}Q4+s4ej>RS?&By9
z$1RCgJ&KDd&f4Q`A_KN<?@EB{w;?*<c)W~Yo;FlQaj#5L07u?b9nfm`6jNNsQTitp
zF!Q+_$a?d~yLSGtSVwV9PF4Z-<fsHV;{Oy=oSr&)DF>`gt200wJ46BETPJZKlrWM3
zHg}K^*JHg1;&gA9*#u|Tkc?t787u1_cc?^)4P{ta7kr`uTC^OMv?K8x#nsOu1D2Wp
z6H{C{RB5vU#9EFeL?+zPQCwfll>yt0!BnD+_fUz}-NvzU3@pWfU_k@!#W7>`?raq+
z?c13;khRuB$e7(-MR7D3NF}`Tv4UcIFG|H%VyFn>5oT|npa7Xa^=Ck^@Bd`Lvb7io
ztO4W7uCl`w6!+%ub-=Q!GXqji59UCiWkoT?m0g7j$UxkWO{hd4`Z<baUIn8ZIj5i`
zRcxaIfl6O9Alz*-mC&Oc8O3&V8NoN+`1Gcicp^69UhGy5A?bUsf@0fWKn2oGZZX9%
z$v^%o1LmyH6(Help9<Jc7_X|S!XO<bXX!s;inB}!V^zRs>!AadGHb;YEB7$B0z~#7
zR)Ik2Mg;H7X*!DS>S!u~xbYH7deQC-Fr^yL-TPZP6~+GPc`?P6gKdsTfR<N?kluKk
z2x8^DaZU!JA6+42omj?Df`_h3fGa731Ca?kR6?1bWWaUcx&)X<{G$Wx&OC+6!P<;U
zwL}#NIo>hKimf7uy9a)0rUIdMe{n*cQ4B~wXS})YFHh3Sw7P_hm9^$61z-=qHp-Px
z2%fRaRjjmftEl)kFCe&=ekV5Krq+Lq-=U@aU6fWS@4ZDUd+rt;$E1xvs{+<SKXJgm
zuBr@(|FR^NP|s!x%Jx|+7~tv9S^^xeFB8J?mt~Z=zbaE<z47Me|4A9k;j;>sTEAoi
zPj7LQNN|D*B)|Eu5;;u{D#5jT8CK?<uNA=VDxw3v@uLXV!NX(}=jKHWu-i8)RC-mV
zV(m3qE`U!*RWb5mI(33zOKhy81ZP%}fs79GsF)@_<tYBUH^dZYdA~HPKt}o55@1%I
z$$;CvTugC{>-*OnNP0YqO6JSi3W~kRMG4q9!#FrH=R8N*^|}}#{Ah)Yk}$KdnBs!*
zb2D_nf67b6cQitX^%zd@Rtz&C@!Q)9*EtYO+9ziIW6qeR0&G*FMCklC!RociD0$1?
z)q%jSAE;Pv*;JHZi<&x+e)t6g9A_6$aSmBeaC{z<3*fmg?4$ygQIB-MS?C@E++i~Z
zGUnG+fuQRX8OS_!S^+|(YKSQwv*X!A2{4svs{?GoK^cgzlq&)5DIXKEZkCZz*n-JA
z;N5+j;5<A`K`~8!paTA3zYr{Q6o!($d@u(xM-`(InC?|kEE`MfKxXPX1#qRT7Bl}5
zP8qEOna(u?`;ONPC1vJzF~u>_MOS6O9Vo2;2`kMEhz(fD0q23TGT^zsz`*-^L=dO5
zf0G#C2n7@%vbc=|_&SeQf!OlfR9wZ4>x9>LpyKUbUdA!v?;&J>_j<evm@+EqfbG#a
zDvpJ{BozPK{i(1IjbDOHr@v!Z?ff;Y1C|Fa83?>rj*7j^c7_t}a#;eB{>`NltUOw#
zQn)o0`;uuoN;r6t1FXR35)gU%L<QnLHa@+%$y+#z<?|n9Ads9S0YO)92H4vhpRvg9
z-YSZ3{9;Df|DBlP#SGn9r2sh<77%>OCJ80?Kvjapy^o`4&GM*Z-8n$TS@4RC73(>I
zR;J4Hsd%pMSFs95t;Q<r5kv77Zp#6G{n29Pzo!@C6d=v}1Hr#)3Bh49F|6X3kD+2~
z<K!rzGu>q%r^feUitB{WOd&*SH&IYBi$9@ac|K1@u@|+dK)By79SH5}OJ$e&nF!)K
zu?lM#;NG-N25hFWDqvo7h6Akhe>&j2G)D$p#}gR9%GaRcab8wYf_%9Gm@aM*GykE+
zcTfTEiE|8yJbPOQw5hKpAS<#=OmT3q$VeUV*1w?wwst)T8Aoo2AdU&Gj@JRtxu#Uo
zTb^Mk+5g;;0CTA^90>O5OU2jJ$xw369F&1z-<|~4BTR4%9?1&i-@?~q2!YLw=qT3C
z|8ZbX@kAML&-*|D(%aAFfGIeP0nw@xCBV9|t(fA1w(%`gz`U%pQQkQ&p_n?Hp%SRs
zhM~B6b&`S5x&<oWUb~h783T+@X3ShiMcLk|oC4UA-=~tTyd{FTX?TqBv(Epll8$0N
zb%o%+F<e4%CY<DeV@Eq3@U%6G_1sAnB|51W!EtN5f?{2}fdRJbeZ&;^GI#ue1EEPz
zWI*e3g8}Aq`w7;znJS94hVk)pS9wQ9@#bbQAhvusA!opJ1tp_KEg1+|CP{?zoB_@+
zjE{O+g=e&~9gJ77^6y`y0_I~=B_Oc72LoJxuTTKn>c%SI_eLZj@xmWsil<|Bx)i|t
zcpM?jOA(y4&k~~h9CCsDgEw}rEde1d$^h%&%?gnAqw$kJxoHcIlJWRwF~wP?TbT^V
zIvb#3nUY4Z{(V%&Dp0<I4tOpc5mTHME?ZCq>}4VZ+ox?fO85!W0b925vIILG6hWMo
zTJi*~92>S0LVf3}SlLc|DgoN}#&^fkV}J<aEL*ou9LO28MF-6GI*iJdvSNy3vNtEG
zKy>CX0{eG{jN%`(kYIKiR}K}nij6oc`O9<#NO2qYvhTP-#rg9Bu@T3l|5%2~j?X5m
zDAo^mF(CN5uL30OEl9=u{bU^_spD=H2oHQ9X8t4m^|lPy#@yn-t|$LWfVKQ_2J9;~
ziUS+Z4-!+H<=CZ4go$<qNZ#?7;OlsTp?Gb^n_{WEPe*aqc}Va)`;?(X=1ietiO!=E
zZuEhSmHkABR%zj)3W}+!@t*o773L_8=FN4$USzinFjI2|1RKnz;`?`!isH)Ns003X
z9u7D?8x>&Jl!***&rmp^z4>dv&2l!a(hp|qSeg63`BBhotAb+v=%x(VA~u3`z^@F&
zvEp|La9*0L0NS&CVv5($@Aygr?1%2@fT{6TF~!O=bqygV_0&;pAO0o-F6$}@$llmd
zOmT4DnGa>afA$cSwB)fWitV%cGLUh2Dg(kt&vU?i<GKVSo@l`USMJ|piVM0v{!apu
zS2WOpz{^|}2-bazkW#agjH1oXk^q~fD;4iIrBxKu-$yy1J&s9$Ij~p;4t%+bO4j*u
z9L1c`NKA3};5#REAhxI>r?PY@155|~65!b|R0SOKe&c|(^k4<>CjBJ=o;yJu$m%&x
z0krcQCBX8{Gb*8jSsW$X`LPOkZo~+|z&;tpJaQBV0#(O}DITHwi$e;arC#Ac#)sRe
z*oNlFD2~I$bRe=OrU1Fy`g4E{`$z{eFaD_j84r3$fa6o+HPr4jR8c}#eF~xRzXWSJ
zyN=?>ND@;#9*=1(1F~0cQGuPYHB`*Aw{a9_y>3)|50fMm?Obsk@SNT*15Dnk0@?10
zRQwa$8~AUef>kKGj9~7WsiJrrf5-rBluu0Ycp`gdav<+PI~njiIxVJHu~o?&*y(v!
z0<@_|2;p0P9mO@O9|vsqgA(8?K12pmPk$<=xQ=DGmjjN7tOBtkKQci3t%XErP*}|T
zo9bmb;G6T1kPs*&f>>o$x=3KJ`%5TVunPmswGK!?;GywdcK!RI2;wZ$cQpvE-s=c#
z>_iEx^n))su&ZPoK|8WRM+tAK{eP9+2b2}nw#NG=ND?rRb4gYtX9O$|kRT#Kaww3T
zlS-yQ5F|I9^V&J@T|0DuZeWoF36g^(i;N%$ND>6z>Tecj+;hjfZ;a<1j`=(PSyj7s
zYShE^Fo3NntWe2d5@5@_ssLGU{ypB=F<q#zZ=cFoS+fpsAg<0*3CKFoM@(@)$*UI=
zg454RC_&Fo70?d#(E;zyJqqC1Y7tYMWi2+H0j9a@sd&{g1n2Uiw95GD3t1%bk7E;W
zsX(SaOH5I`pUo#YEu%P!<NEhh91DgjD3RC(jNq-M1EC88sKnO2B7(S%mgP}^K>Zgo
zVE%W!4s0$NA-HA^=P2g9HY(t=+Zn(ro#MdOzz&JZrNS~0n*9Y8`|M<bWkOGem2caA
z1#nzjrvj#xb9Ep%VwVCqJJgkdz{!48(y}coinE#V3v_U4X<Atp?Uk|e1lAFP*(Eqi
z=-)O}EVge&5KlQ+MNxpwdNB!bwXC26p$o$azN3<YV!8S^1K77q3C`5-6cm3#RVwz1
z6J(UYD&yPj|8a+olDK^iqgD6^8SqCN$Cwkwa+HK=#$TNH`?u6l()#_*fcUCY#1!{z
zc9v5B_l50Lyq)Gz@n2k|W91z(odf3WgT)kQ`41aEpY0tVQt@`tRII#z|ET~O<$F@m
zX5SD&oRu_vl?<3F|I7)#xe{RYy`=!YN>>RP&yO+`t?fpmL>H1#tS_sIDK4lb7(c%L
z^IN6N{EmW^>v|KFaOgP|@51>KidJGgfd}vEC>hBQRKWF@T}*N1pP~X(951ivD7H~A
z#T2XcmQNV4IdifKIB$O>0Vyrob0D#hO$96qnGB?rIVYw#-QHlI1UOHw=76@XEWtWV
zQ&6l$pXoqmaxn%t3MZ<7>GCWI&~jc=v2E-@$ZTaiFZN+Gj$=&26*0ww+9rIj0+t~)
zbs+Ms@fmZ}$YChnX1i#Wy!Lky#98igR~g`Ia9jfHlg{cu*i}jetkdJE1cHl1$bW)4
zl~ur7`G5krKB+DPIV%c^DUQh)`Ah*a<NHzZ_M0W6gu44U;2aiYfaC605)hoRodcFN
ztHl)8@punWu{U;6aWq_^VCDE}g$@MzbP-b=9LOrB0&(ZsQQ0p4$xzsgQ3~Mc_M8LG
z1$|{8u;P^p1fLl{$8yf@mQd2FBohKh)~ZxuyXpWl{lEd=@SaqX-d;;3SSW*GWxnB3
zf$U0ybilQ6gbZkoc`B8GV>w_x8IpkD#!5P%wfvXhyjxI3@uZ(;Kw|Tr5)d4&iYcCZ
zc>7cZuzbKIAT|82N@c+@8Sr-LD*^6}%@iPgz;zCI=9X0fhcbl_>bzD2ab@jNf(&eV
z9;;GGD$1$UUM~ZlNybM#bjA4feRE=c9jk!(g$nowt45iTDWL>=-&6tnwUPv9!#W(r
z_Ua@-`?9x!V$V*OfRraEWGaulaUk?rs+i(dBWV)|DUQD+6i?%OGT=L8+>a}{IYV)<
z;W`j$oh7C?-L$l{1O(QYbV9#3AFb>NL)kdNL#xcctE(ufk=Go^K0Qze!Xe|B)Yd%|
z6mzp5Wx$?VfB~N1dph8I^LLt&>4!N=sKW~}#r^nxFpf$5=??{kRXwc(sV6JTfYlMD
z0^Zc_9Prz|kpbIjw?f7Gtqvq}<0qT_S4l!~y{ssvc&xNvjML+0C(<hNqnTj4`IzAt
zSK?I#i0@g80l|geset+UPZAL5R9dH!lFR^SVo?R~{aTm<u9>%Fz+U^H0z`UsAvhDt
za1=`oG4gNBUaJuYEZS}z$nu`1qD2>wQDPkf7?6JR7b?LIJ4z_Nc2yYQ$oh_u`S7KT
z;@?_c0laZ<b0FYoM8(uJMny?@>y8X0)>%Xdj=D_9G>y=)(w>fHK*Fp>D&VcNOw9au
zu+R+|aGZUYO4fuu1l#q0RIL0>tyKIcCsT=hF`Z#$Ti2CVJ8n#rQLG<VQ2<BAyHuPB
zbwv=56{>nb0=(N-Q!&3iM@6wt=`R7%I|~!cAMeyrtgTinfVXHn83=XG(t)H8z99sQ
zJ(f|z1!pUOWvTHHTKzs$e1A-3ScUtwrjj|op^BpMwqlAq^j<5c0QQ|7IpC_^NX-0I
z&2=ge=(~{t>|8l27XNgPVqS8Qif!Px1U}%Uj@5?Fe-YgK6<XOR8_#sh)V&Ihac|lz
z0a;aRGr-%xrUPuzJ}SXOHB}Us$@ocY8}y|J;to^yxfEcFeXk58dKz=UzNUwm;+XXD
z<0QbEHH8x{?+{b0?A;5hfcC734kWgHApw#5r8$s%^=&GSHKi1kr0uDMa7GD^5~*8Q
z2R6%1RKPmtsF>m*w(RSz0JeX3Nrb;X6H}}#Q>rU~xA#36@VWLgz~U(<0iIcxsQ8|A
z)=^kR;|F8r<)sozxc@^L&>kEmINKKxK|Dlg<^&0_cbOyu&O1X05$&9gVxD$W1?-0s
z2<DY00&|vDu*xdjm5{gP6CEX|L`4<QcEwRK$1EpUKPw?N;?bQAPI17&TT)4Ux<f@_
zCu&hiU(<k7*}s;Gb8TNL(MSJQu?jXY?k9NXo`T{n^rsF40(T`K@os4u(D+3Kuv9-w
z#dBqpM5WSW9SGKVrT`vojSBdt_K<<NsLKRLtw%aa%DD$(il@Vx{%hcc@y6MT{G(&#
zSz-J_<`1-^;@WO}09>ZqGLCVcZm$4Zqa7Sb{V*&8k&id)fPKY35)fKuB{**0U?{Fx
z#uqI-{u31?WRBxNa%JO7?TUFNLjEgn`GW!41w{fjO)|+q=J@XzVA|YM0UYnuB-o4e
z)lovn&xt9nW7}_hkP@f8qE#^N2ZFuh431;0=Z8vwHtC20IG6S3Kw`zJI$${vQ~>+X
zN*suBd_pDI?0|}rz2pI5(}5}+g#}BCDQ?5^oJoLti(LkMw@Ppzv~IToXj@i`nSXGd
z+7gg{a<2ktdlCpC$M-Ty%ECre0_~hSioH}j6-eqjS_VS*6FJ}<zf=Wmbqi5(Tr_Sz
zt>R*_5f7TYtu`Se)=9@?|GQnmN{h<Yf#Aox7~qs@NPz9>Pb!e`=I_K^(ngMwc4e3h
z<Xn+ez*TZJmGnt}6a1%|GOQBM8NZ0zKD;D?xF7H6L|SDQKcJ&n|2)Y7$DU;bZ~R^z
zB_+M6nBpvJ)Ilol*BKlovST_G+sgAYN`f+2OmVQMvhial+@d0ttkzt|D#bd3inXqn
zqnIAYN`PmvQvm{F)-WKH)=>vC{#hYr{>{&d5G=W66e_-d2(IK_48>FJA2G$j>AoZd
z@U8HxK-$)}9Pk|6r2w{Sj~S5oM>7s+-$_(_<qF9t=GX;<wAsx>5LZqa_Ol9DTGx~S
z)7<A`ik1D~S1RC~Kb8TRWygytR{pciRlsKpQ;CbV$SC%u#^*V5%cW2m*hvCB_d_~h
zsrWqyw#K?;D)K8GNIP^&0n9_55Ij-KbQD{*@#BJ3-^fs6V&)N;cQnDa;=F{F`|K?)
zkpDYn?0p?V;;!?0^dVTgtyfWk)%xgw+xi|A_f+HC?0h{=!HRWXE)m+cBZS-mjuNi$
zuL@W#TV%i#Jf={I++~36>oXGI+HzX~5^7~}z*>BanBvB?c{>Td@#|C+Tlpy*2<>~O
z16p8_LfEs2iu>vW86`7)vjl{G8b!sOds|2G&o)Vbsp(uD$Xe5o5NS43Mq$6-;6P#<
zmzd&yBJ~O?z^X!7I^dP6%7AV0D+ai~n5+Vs58mTI`Y2;%`E{0#;<|D`OmRV1)qXON
zTFgmBYwVX$)<s%yz&(CF6=%eF7VMUBhY9jzT4@tk$T--0q>~D`KAOsb*z3tU;N7!M
z0<4jN97x^pwG5bgFD9_3iy4Zq^EDkX|9(;hyp^hRAh}av1z1}?RtH=wji<u~-sdRh
zA-iam*6l|HC3Wgl32>?%#1zjxcq&l=qIWM<0e7UQm||rg^oRk@Zubc6^K~+cb@3T0
z{%I{$6vyzEVv4g&i~4aOoN7F}x0~^clBr{=jFnbxy$)o$+e?6__jNJF>6X%m39+U^
zGK%xmKnbv{AFl#w^$Tzyp~n&`R{K>E#ObWSb_sA4s;dGCt|1%<6+KGDJpX4ZcGGFG
z5og(MbtY(gQfU<)I9bLjsoG~cu&dDo1xP8<NKA27#;Ft)(5&?p!2Q;{9LT6?e5u)_
zK`M%4VplQ6S*FZPDsc&oI7;Zt8wy~0Wqe4iO<T$+nYOtq!2WC?0j{)<skq}F$S9#r
zcL{mFkI_*!T{|oR@n>ErfaO7wnBvj3XB`a`jUl)?$LUz@xV3--af8065+0mK@VwJp
z!OGM4Z5eRyx+bQ$V02^{!P(ZsQIhNYs{m<t@??PhJDQMQIa)=rojj%kjw4GXAkxVA
zB`8?5E3MXCFh0-0)Fuj!v9pIV;B@v<f!OR5IuLsG4F|%?C4#of_=tLswNbFL?{BFC
z$!~sln6LjC8O1!KgqYRo!uk{JG6M{(RbrV#EOkO0d;<EL??-ZX|{{nn!ciJ1*#
zz_Fq@2W-!pserRWe+HzpX9TO?ETP!PWOBf=zp?@_<!>>?{jiH`2-bOx6_oI&2dS7h
znmLNK(MKxaJ@Y3aP;Y{S5*``k02_Xp0cobX3gC3FkpRExBq689n_nc)8gHee{I4pG
zVIy1UK>YrB4A83gAy{KN=_r;F#vA8e*-=IbT{k|f&L!_salPKBVC7#}l>yGo!eZus
zGEX1ogu-u20K2_Z2FxpBIba&161=tjDvE3NR2_)yJuLxw1-CFDwW~`89N#_Afs`+s
zQgN9-q7r`ZEEUUuCsfS)lc~71drB1Ef=2aKAoIm$f~V>@9mVvl3&H=M@sV&}`G;W@
zT59|X;+2iBl@|JxR@w{W>)?I!-w@hwe9UoHM#)hMU|;@70^;8v!GY)|C1k+1Z4?76
z@nYZ(Pu$=@Xy8|3iWf6=*Z_iQs-#lsr7*yEZIldXYkN|0Y}_rOI1jI*qG_o*iseA2
z4A?pzR{;O3%VLU$2<P=ASiVe@Q8o-eL2!KiQbqBa!eWYpvpY3q!0L*{6~O-Cj}j1j
z>@yXJ>t#IC;J)P|h_gJUhsuEa&F}vXH_T-y)>XQg;+WOZlNb>D-XImoDAQ91(kFEh
zQyk-56D0v_%Ffh*@Z3MC*tYp)lxTg0nBw5AC4N%@Thm%H;NDe}iZAvtLrERpn2I^7
zya?hf_op{FkUitM1cdv{Q2|dlMg|-;bqVkms;C3onjKXDlg)T@<1@`v!g_5PtHcT8
zsJI?}N=U0bU2MdICjWJS0qN0YbRe;KDiv+RCK<)w{5S_P$1hg_Z%0W6RyBG|SpO`Y
zq4>(zQGf*BI2o{Ro=wn(8@C#l{*%~<M~^G|6|Ka-cALtE59%ve{nV}*mE`YwODJ*G
zvQ!|cit&qrXW4m%Vwz@rAh%3Erl2Ie|Bnhd%Nk!sZ_08R#dWbT2h8){SEx+fr~|B^
zNd+uBj9U$S{7ONwzc|YPN9YeiQgSZ|MXNDNO!0Iqzx~1iw&h0|&@4-J!1bWA1Z;kM
zmILu?9;!gv=3NrtnED|HQeJ(f14&9}G4rn+TtaX)9;Tz%=A9)tZQl?QlpBmzN&^*$
zx%IaMI8Kk!0oRSUWWXygl7PUHaU6)8?MZOtv?io<Ha?=8UrbSOj6-`Z0lrm*#LWNv
z-26=e?1#+~;B=KBxc3&3QM|uw7gHQ;ZoEYTazj6=fa6#a1H|7f68xr!gkrB~z`wmM
z6|GDh9TV<dP>#yKU@*&g!PD1R6%?(1Lt3$-?=ckrN23+Mv9LYCH^`=<gbvkbKxBTZ
z0$BgtssoN6o~wZGt0q*~uoQ;EhAmWpjFBBUU|VBmfH|R!40u0is{{UzcQC+Gc(|D2
zo^A6>5X@g(R#9^AKGp&6o8R;uK7U3*$u6)>rqz%UD!>L_m4LL~$El>uD<VSvtI@T)
z0{F}YRlxO2icVONz=6Q=lLW2GV-dtz;ROu|n`34YGNx5hvGTQV$$`!17wLedRxt(e
z<kpga)N17k+r~dtP!i6^bHG&dv<yTyH&Y4z+KHiLroXKMmSX#LAUZaQ1J1Lj7~nNM
z*8yLDRZMZKmOF(B!Fn|m6#t@m3}E&1gE+=p;4;BiwX}|sdhVDE*b1%{QydeTH<4g#
zUtL16pKYxJ!QoCS&Z$KS?yHYER;JD^6d-L&L<Rz<N>K6E`C9~W<t#^g2?%WJ%K-Dp
zKNKK1dY+i#828;G5}^HhL<PdtRSvj6F)kR0yTnkeqvpy$C@WC`g3l&%!27#Pr!wJ7
zDuJdSh#(%qR9w@6<VX<}$o*!DnE8+H9w!6ASq*g{Zdrx``1&@Xvcs`kMG0q)Q~>RI
zYca*?wow&05G>n51x)KtDL`u54LT4?|A+&oP$vn<T3=BI;!G7)AockqF~tP~L-y)`
zb!!m@Y}z|P229t7QnBQIt)p0`uBQ?_K8mAktSIY%_2^9%$nrK~fa%yynaZ?p2*IJb
zge`4ah>dvkq`2NXklf>i40vnh5W)djLh;`+z96>u&MPS1B?D!^_eUlNGFF>az{*Ec
z3G8`H#WSI@j8)F2rV0?;SW`^#=*jhscbUx^!cm-uQe?ndy&40wxcL&`TKkp?cxOe+
zfW6fz2?$tj>Oin$rV5zbtf%6b8$<AYTt~txyrCyyh2=9D#r@|?1@JkSaKeu6gkV%Z
z6~(^zPZ>z7e^vt6pPLxq|L4t*+-lQxl)%|$9B?*xr2y{t6qQP;<uYI`JDiZaVwr-H
z@y}fe(7MDC0<B+gl*Gj=bs#viqXM|fO_6}GbYD#IV!A3D&)qa%r<JFEJ%*Lz1LK>L
z<NaAhG0(a{E6p^E5Wl;*gjI^OrVQ*^6=J~F`c@A3i{vVV#}+ZggJwJ0av-cdBZLw@
zl&Ms`&wyZmT>;$Pmk~VIOX(;nZF-0)uH)#DECJrKJ7gfxX+FXAr<0+$Qe_qJeo%!2
zL9M(3*piHoacKEwDnC}eB{uml%k2>oV5#^H2Yg@MR{;0iW)cuMe@O;H<?ITOw9fd9
zSz|3aN~YRa2E1jn3EK6E5|!|N74Vc@!Kh@FAZV5(86{GB3kSR-IuPvP(nhK9g@To5
zs`0BQ+qH^ToA=Dtu?k5w#S~A+)9!nM>C8U_Yx-LfR$Di(QGkTtH5JI7K3z<4mbq<X
z9k3_N=Y(^EsW>XFkWt)+`w+G?d`hrf-mGI43Rf_SWvz;0jw&FgxUz3SH5~}9kCLf)
zexqVrv7DpWewZ!;;oO#D=D#U*A1Xkq`y~~#md;VED<4UKd&zYMn2s#ugmOnDAQ=6z
z0z}5WR0;2m7gJnWTT_q&u`~MVz~(pqMgYhDrBqTny(>22;EcY7B_Q`;Jr(d-UyCVL
zp)(E+WEA>K2h80LF~He=03ps@m!o9wOHhGeUVjO2t}}jOdu$hIl~CX#1*_1_ClZi)
z^BWxq=5AAfxbr?0aGbnDV6#{0D8VuZWWYSu_&OvxeiA|4YJBzU65uKMOaZh$-|2wk
z`3^C~F|Ji>B*6WmqYear`HcalsarS@{^Y8d;$Vxz_(5%7`VK=ejVnW|NY6VQCF|5g
z9dNg*EdwciHc3Fz2e)(}@N6XqVjFIy;yX7}LJ71#NpMUm$5Dc}jbBd_>vxh-(hIK=
zQ`}GNvGyFup7)CcSiiMUvB%a_P`pjwCs=$@DvJAy!a9&u_!0xOgF98g<9L@~es+<N
zI{QNjE0?c42iBz=k^x&<Zv}|$>!4!k$~j8-#8WDf{vi>>t$LmfGtk5MYXHaF?NqF+
zTQ?H2YIc`V>{Sa=@t*xsM{(VlApv&ZMMh=HSrssqT%Z8%dJAR1TFR~hmaMM`j;Y0D
z6jm>dig$i95yXu}Dpln`%G2^PV0yc)4g?O^CBPcrR{_|OYcimn=t&5bI3S_;j_nas
zT+p35jRA=zyae~1LmXxG;$bqN?J3TH;GYf!$o%PjF~#Yt{_4j8U%~+ya6R~00^+ya
zBV_KWucM@w-N%58*{#GBr)ShHN7#1%a~&lyu`&lzH!V;Bw$b=(*hd?m!(g+$v`Q<v
zO&lVwW4*hWkX>DpP)vz8bRh1J?FtZBrBbooxFmu&%j-JM0cV3l3Xs)xg$j6Ez9ptO
z#*)`f2jcC82(F^JDvGxA6~VUfYZ3A<c)`v9+rOVmfa@3IlVDkJQ$=x~P-qqS*D8WI
zOS`sB2QrIJrD7YUFck9_6BNMqX>~Eh!NFe9R3b%gG8F&YV`U()cC89npBf*?VAN%u
z%3nQc<?mEZK{4Ig!~pArGBRM#sm%cw`%BFH=icuQ16Zdl9Z1|=PfW3jR6VKyzK5eZ
zkht}M46K@7MNDyw)%B9V%4<4`_u@7M(C)le0c*$ER8mJ><|v+xP8|qsDk}rY>J&n-
ze1PD*QCz_)^?>oU@?6=cqL^wA<AA?tmYCu;ylbL$z?HdK0Ww=I=RnGYjs#o%emaW1
z#V0ag`?&`}YxFlmNuTyw2ZHMosaV#$rJz{fn;@omEL&hA1HuK0NC5lgOPO&0EgeYw
zVle}>jJ*WYps|F_-qK1`{&h^{u1kQYSbZv%iJe3ct5Cx{70?RoqhfCG2S+hAOOXNF
z<ZKCWmA^>E)hmNaAY~C1ZCM77%D-UJovk`xDff{IXbpNQ0DJ$w45S^}#{qYPVk(t$
zYh)nU;EE2|$7YKuE*R)CR06bt+c*$jV7wRpJFlrY8Y~tYaj>~=5gqU(SQH@Top)70
z+fswz*}9gam^v2K0cVjUF~#ZL#Zd|n+_zK(JO|5iz)|#f83<0<qXVX`P6^P;Z;}CD
zPO_NdI-bPmRJI<PprhnGU91AkEfGTdj5jX#U9*mrv*Rr>#p&AP5-Q;6<C1_t%S0-H
zwdDz^*O-je#uX_Hu<fhH0ZT#SWyyN^w^0TTP_c3?h*JQo%`5@Vaj^^t{;-|{wru04
zy7g&W5%Ql-k%_eO@U}9FR&Rp}<g{ofrZ~oDPGUet@y05Uys3+rVwGs0LB$e#h)U}B
z;|R|1SBjAMOa0oVsigal$SCa63>~nQilGuLp(!Zdqd(|?tIaY-*!F@-{Jj<&#i=z>
z0qed_I$-ZIj*36^2uI0!$0q^q4pmejv;Q+0u$~VPf_sKCl;FBm1k;cyI*PqnH8I7_
zyJn|Q(H{2XC}z*6Dqzdnp#Z*RW--OV?mwDIK=y~1RUoU!R2i_3Yr+6$ze!YV+l(JX
z*5^NStU^tCtAI5skpYg(M^v-}Z*FYx4zbC<u^l5BV42ce0oX(1W8+KPCZlMZrbxi{
z-_KI94r$3y!rz}%0Bdod3b?nel>vK9OEJYm*zcX;famR1gp`~V8O1iLHv_WzL~|hg
zOMevzbj~GY?U^HjxQ=;jZ3eI!U34HV@rVjI{l?1tYcY=Es<U1OtXAU_u8kSZP}a(0
zRKT}Ok%5IH{~}~9@61u0KR;IiR{M$!1nwSG0sF}-90=cXi79SO8+}#*Og~r|V128d
z3Ybf8lK^LxK{CL;G2WD<$xRemt=uXBTc15)fW3N64%mig>44|sn+!<nYf=F39^(_`
z9A1c{gp0qc1HoX5nBvA#-dZLDOwS_ZtXRUSOk1J=t~reep+ct_iu>4?GO(!uKE3(B
zQzo^`=PFwA<arE8x4o8txWl>%Sj!HgV!t#_M)B>hDyFy*_wrvE5b9{W4Aw_IWE9)+
zGZNs8Y*2ulnLjfibo#giShaW=uu0p*6c^n3u$m6od(4-BAN+0=2;|P<fc1F11Z3>F
zDyBHg`v%&aAu5VzvXcX>(EtgsoSLix?n56bfUWN|F~#Y5uMcv-7vD(&q7%R8fTwI-
z71$D8RtAFGS4x26$)_9$oVm|{c;9<6kQo@P1MaIH2?%H*Dn8SvDvE7P0t1pJcsLNu
zidF&JQZ56jUp^I6+@Z5#MTyY(BMv0?9ZImg`CA`HaJATogDum))B&y8We#{6tdaq?
zrJW8~M<!Eoe4Zwu*m@4r0n4$&RQxw<D<}!ee&v97K`J33rm=)#E#nbWJg9H?J_XQT
zH6{d(?9x%Z*NTfNj<GixBLivkQ#hb~o2LVTs$B`eYl|34(nobAAb$NZF~#ZT!i^aa
zTe&O;to`~*fW2IjnBtf%F|`<A?NeO_ERnqoaKsEFgsXKBK^&ZZ;|nT|^ni+Dn>j=V
z{PV68l9cu;imOhnnBuIAL-#mfnV6&k*01tpAW+?|1J1FN2(DIc3B_jiset*fG6Y}F
zWrh+OJ&XexXIsdCdDb3<aB&zFdlTck6YRS|$11M-GX=0Wswbwnu@vhaD&FozI0|du
zSpuwMb;7!i#&ztEz7!jAmgnViS~(6}R#045{t#2FSnoA5VEe|Y1NM=1B*0hWBthHX
zL<DiL{bD@^nEo=}0#_u1qu6#BC9~#A86{)zLJ6?+*Hs{Ait(wlKkl!ixS|)!fHk3$
z0(cTe609+ER21u;h7zDvJuarWjmWG`DquaMN<i8{HwUzV#i?xRvz?&DHm6l){6_@S
zgZhNfqXBe`X?}tdm49W2DUnv@#5=Syci2rOF=wh8m48gk_9Ik0@7JNSY1AwotI*DX
z0@z2DR)LK6!*xPV5ebO?BU?;yonZYO888i<$$)U3?+CUzt#p*QPmEtkofUjkEQ^a0
z+?A(GQTf+N|6!4s0ui~m41{0Q<3P$F<7Z2-#61zjF@cFs84#TC69+=&MytTqI#VSe
zMZU>_wAgP|z}D}P0^}@T&4AeG-*q7Ptycp4S85QVgEB)|7j;Piw)L99f#^OL#1yxY
zS+Y98R&}k6Vy(4T0aDUWQE?r5NeFk{!?E&wQCH0TXBw<+z`ly1I9<j^-P&)6j^cK<
zkpN5AM^pk!k8_mBX9vU-*RhU1A_0Ll<rKibK0{2g+H8Nx0rTj_28QP7D3;V$3~;;K
z6P&M)D=5yMYgIrS5Gw;a?le(=t&LZRDXtuR)RY5St;;fC>r+JnLM1%}fB*Uliaq5f
zm0-uqDoS|KdNIXy0*$^=0Ont$0-kC!#S|;{>30&4{)6$;Mmye=p~TmyPAhlxWf8=|
z&Ni#4m~v<8D3KvO7~srZO(kpUV=CU#w^Xb=>SF~+96nD>ah>dXy$S9i2N;Uw^EFg5
zYefhtrN36NN_QRA0oyMZ#ms*?OY8)<QjOpUOqQ_n6(7j}+t(k6ng3W9jL&@Pt1Jb@
z-Myg-xWY|!AWKhDs7#z9rZ~&=>mY*r_B@8-_@}i31V>y@fv~-u4p^7`E2cQhf4DOT
z>=hd6fN$3T8A!QVOa<b*tso???X94=g3l!&v%(<`>>N>40kmUJWWe0-nV8}se0jGd
zz_Ds96>Ens1oNro1lQ>v1V>Zj^~tQ(i-{!uVxZz*3J}~KM{plBzSQ29MLAaXCEpW5
zjz<h7&&E|C`*Hye<g5u(Nj$Wlq4=lWpyG*7<|wTFI0pE}<*9&{vWj4>K8m3@zU!p`
zp~5?KV8gUcR5tf&$WdIKE)ddZ-;+?RT9%mkPv`4a9I$UJK*juPj*Q|CJW~P7!qN<g
zdo)K(aaMXiMF+H(a}>fn<5{rJf2CD=QK#65gZ-P{{J5|`p_OSQV^}%6og&007ZX7o
z?CRyC5_+!!l}Mkh9IKSV`&EFgIw1q0mn|j0`a>EaU_HlBlA{*uK+4?O1k3hZ1tl}3
zj0Bh`%+LYPxjSNtM-P=~&w*g=ED3O5eP2wm^8V9U2Ym8+nUJ!Z0j!<zkq9^Sije;-
zDg{-*-Sak;>@S)z6!Wa_Wx$<hRRI6uO)3y-XdL7Fwy2K6t{&xpLpdb_iJwl?fpFtu
z4Dk18zyVXHNd=q(B2?H=vk3Y3Q>cUvWIcO8#kSwWQGD4mb-?6V#Q;ZOhze-$6ejov
zj*w9{l|I6N6yLjqoR)4Gg+(3UfWy;)kT$4`jAF99QUOQjdW7tci%BT%0Z%yK{AN29
z-*;=NBsDo{tU7-q4iV47vN?|d$p?N@0qtdd83^BPFQz!g(>q-U);}pK1Gdk<qLMZ@
zsGuYsbjX0Cw@plOmc9Hj4rCRHk^ozsTtd*+TtP{w+JTDyb7Qsr4^71?aD5&JQWh;#
zfQ(t0GT?ZXDFOaIi*!IM)m8x_-g7dLd8j!D+`}j7fH(P}3V50hkpTAaJ_lUKmx?Lw
z*)2y1nz}?q@jbsU13xNT3C<-0bQDL`)(T+x^F1ngo4@2JX@68u0QVnd#LRy+-u$om
z*p|MgRZen}gcX}NjRP60+KDL+4#fXO#gf&R;JMU-5MA2%6~w*hnI4sYurDq}0xZeN
zgvf{w8A>oAPX}z7f2%;&h7n?lvpk8pGN4)B-0HM=6~#3Egal--=)?h+?H2`b-<>T1
zNqNRc)HU}P6(zMoCapZj)-jau*v5p_vwxRSTnFFhK;Fb146qk|EdkEdunss%7$1rC
z)26E^zBb)B;BUH3B}~7^fXuP6RFV&WPlzsWd>uSTZzwp%yJv|GxF3F`0)YbS6(Hy=
zAp^c|KVv}5;15*5!yQyobF0cI!D%xUAg)7c3GkO*qXU7(l?e&Y0vyG><dO>b>Re@j
zYuqjlnAWrtQ@jQK?ms9%T=g<CU@G#q1b8Oa5K|n(Y96Fw+qhRnG1qQLMf`$DF!$Qa
zv9f;{)B%453l(RtPgN98hbauOwkj*8xU%KV|MHY|M|~Z|95qG)a_o=96vvnjKcM3K
zG*3Y>cX~@qvC5c!R0qN>OA<UCo=GU43M&;rD?UXA@-pjDF;}=Ef;ioy)m8!X&IS^Y
z)nJjBV&yDnJgCc-K&w>!9IaeKb&g{~r~1eMo1Lix&LNW+5Nwq$1KI72FMi@qE`qpn
zY#>Vlk{5i#09Q^M9Y~L=D5f|jsii6dDZ?IdAhqy$F~w?&cAH@NqppmScx|KtSl%ki
zfxxyoGL_jy#S~}RubyN8i<k+PJ4KB$wXlqp`SNrLa4fu`1A%zP0Oy<kUMPCiWf8;$
zO|>UTfc5uj4DfdE!vSsSXo5R$Jwvgq-7cm$EBvXO1Bqj@6u^{Yycap`|J6|(9e$*h
z_tZU!%8YeXe5G%SAg&XsU=-WE*(!>AekTP8+^H$1IL7f$DhDz*7>~|^Ph=F=kTDF<
zvYt`Nsk2!@vHRxfK)mmmO1M{v;CT102;$1lxd8@P8YRns<Nh8FWVd+2fPiLE0Q2ND
z2{8Fz5rT~pb(HME#Z(}8CQQg~f07V*^E#*4s)|FzgQkutO0afqqo6o8U6X*!L5rw(
zKiMmSIM_X~Cqa9^B*B*PGZzw{pXz_>K;o_zDv<T#5(aqI4Wy#wT_HsJpP*H8@-ZHT
zD~}yX#kVR~LJ9nEQwJ<3nh?^r*f@%_tnq0H9Udd2Sj%+fz?N=bDuB29_d4L5Y<yD!
z_gab|ZdFVEKn1dTuF`?%f4|_sj-}P91fSFpK^*LRGK2$3{)Z}HY8-E%&<%!?Y97h~
zQ>FVl;CywSkh<xFjAE}plLNMTABrii>_0bI2QnHQRRLGQRx%K{b68AqjQ#f4RJ5%n
zRTQnlD+VMD->U<TeHS<oD)gla1YB=k4SBzUVjg6CdQ&QdbQEn^yqMx4Ebkmrfb@F5
zF~B>ih7M%6tRe&cDdnj+=4?<<;x?|OVta6hqXa7}Vu}l<&+SZz>tmF(*}E84j-)ZP
z^1KtHqWFvdA_Jk(+r$*7hYQ{0Kx)f|DiCaFWq|8|@uSJsy`qAW;5sX2{tI<<y$Zx_
zFGVGgah;)Tt5=c`U-cD7v6*_wz?R%^RUr6w8Hw<49S7`1C(3}wTSy1APJgO^sdhm^
zFmEzLiEeaE0sK2Ba=<jVjs%$Nmz9CsM#jqAeKVDm<{^fawe~)OdG1FNmF-11U|I2-
z3b;D<mH_*-?F{h6E*Dcg_v|s%Br2C@F<{g7Z7SgXytNLP^Ng>R_LGyLY<zo$nBw#-
zdm<tILT!S*O>-G5%eV^?;Cl0WRXlIMMI~;@4=PsLaN{pdY{9k)N@%h~2;W^Tqa?2!
zO<>iViXa~JKmTv!?r{@zDw{TlDOR5Gixt4689!h>E!<QbV}6sc3T3U(fwZ~ZIN&hv
zAS4Z}Eu*-rol$_?3RiR>t84`c@K0~V0JdPInE4O7qcS1<RYi{CPEOVVE!0j-ag0BC
zGy}}tPsqTQ7r#<rQ))>lEa{m7c=Gm$Db7l2-9-W{*=7~++D9=U_07NGBBkZ~B8Y?C
zzsy$w+mA<ez_+)U3?zIQRDmtGH|c<>gz<D7i}Dnd@DnBjmJ%;HkX7TB0tB|(8IYrG
z)d835HxBqRFH1n^L;)G_&hKTE=}wN~jB7&3zLuh(<SdG!Vr#NRMR7N5qX4c4#;;bc
zscWd%7q|$4rzs4_1m+q)ZQQR<=_uAnEe@nUmt-KO@sBFtpIDBHX+g4rqW#fO0$c@i
zIgt9itPTVVMF=VTyKt09-be}1PVP_v*0H4wZ2az`4y1k4QUx3Z-WF557vXy67~og$
z5i&BCi6B;%ZaXEw{m)wr@K-p>0bfF*3OFitkOBMs91b}D_)AQ2y3;j427(P@sMy<Q
z6FisZs#v+&3?ewzE#oNGGrMHKdo@P^0^@(9lI6L`P`ow05@0EISWNL)ky3xlKw$7@
z9dI_SM#a(QYZb-u%{2vxOG=Od^MfP<BUkDu)_PT_xOTNAc-tG_I9s>U;t=szY_;(=
z`Bv;#QT*mA9Ed5jN(S8h`lx{U-3MakzZ#dzDuCtnJ_#^AnnloR6;e^0tCFbL>m4E_
zkJELmQX009f%q1K#1t2_M@`{C;6_meuzuW#;Q4+jL(!fZ#oyzsiV`}tPX_|6kH`QU
zR8j%7jr(-K`*tZAFr6#Kfykq`3C^r~B8W#fS9qoawmJJ0z-d(owqKv<D4q{uWx#j3
zfeOTS*{uMM-G^kr9ot0(LXAFCfQ%W>Rlr{M3<EaTjptPUk#!(A;7bJvOg4U81on-Q
zP`tnWqyq_qD{;VG<~uRPJ^P2bbbv)|l>p1P#ye<g(wJ7}DK>_c>&Za{FeMe1fxKz!
zbim!FA;EHRtAY|(=^)tLyE%%-H&O>o>02Zq;ci12@K=~Drnnzh)4fzePdBM3fej-V
zkY%kQ0Wtbr4kVTtr2v}uk(lE2%-^0ez}j?x4%kK-?`iz_I~>J6qM-_y{8boW9XUb<
zGEDD@DXz0TDNX?#Tjx?Sr!ACGB4Ibdy!U4bCA->vDwbED5VXQW2)>kI1aE=&^{D(S
zJHk5|V7Zu0Nc&t;Q2ge>1mEG+5{jk6hg1@`bx={XDxYz{)Aoc6WG<Vf1A&6K3872F
zB@}bgEd_8a7|a3Roy8Ime1D7rWH#3sV78PIQ{17y&ovcD`CvT<%uD`~fZ&xK3`nTC
zO9yNdxC(?UO~n+an}4)X$#~q6qp;5xQ*mG1NeEUfz_2nkmI?M7#=B$TLuITi>HRpM
zjTxW-p@AkH@UGaZ0trLEBE$?!CzxMtR<L3#VkN*|@HZWBzx9{{t`<)uDz;V(NGx?;
z1;RDY5kgbG5J5a0o3w!t>X6D%oEN52@!cGuqnO5=lmVY<5h2vHhJ+ICbx;RFU(^;;
zT+rltPDQ(ajEbjo4Wk5m%OZ(?P<E?316(`P#LPdd#TprKw>zN#_7@ez6szFYS32NK
z@4|rSlJlsrH@^#*4~nJYtiOTaNm#_9@=s5DoGPY(eSD%01k>h8g!TgoDW0k_3jISh
zaj@q|En0=P?crFZx4SE*Sh=Dv%0Tkf+Y;bCl0!v{>BLa97gtmu_Um>EkoNio16<{e
zZ;IAxXZ|Yxcgmy{DWYPTI^!V&Y(d6>tce?B!1{MIm6&lEB8VG_wY8;UzPwCAF-Lxs
zfh~gvbHLr@iVXOx_g4V>xl5d|th<=vI+le;7?81D(gA0}Dh1G1PL>Gu-+cKFZIV%3
zzgoo<r)Piru>?3Czf=L2{Yx>$D*32~z&cgaQB0u>3D5%N7_g;;Urcdu{P;c+kk>Gs
zifL>NA#S7b+G|s*sW?Wv(U=4F(<2#>arC^H;yRgI@902mxn>gKc#w*1=Rqp2BS*wW
z92^)hg^F{x@d--Lvv90}i}r~rj$sck>VWxlBP!PIuNjKzc7hJ%#&`(6CtXAkXN4na
z9PoWUPX#vZD@w5aev+fG+@}hVRQ$G>;w=BF3k1_q<25wTs=%>wq$Lqd_y1rh&VBh)
zoE2k>$bflvMS`_yaS_BS?OZgKv{SDbN?xlmR8};J;wTwCObQU()?Nl;iZ14Wsc(V;
zIDT_6!uH29;67)(yb(!NP&`j(62jwJ6Y|QB<h0TYF(8=SlaP1lf<(n>yz<_%eL0G@
z&G-?Ud+C&d5}G_p1`;Yf(gEAS22=ul1~Zh54Mhm{ZQBj7FBGio))8WgS0j|!m=RWf
zM8#YwQAaVIo63One?KL76OYR%xew2(fJc5uOmQ9ObteN7UO6SeVO>D*zrR^V(GK-j
z0Q2C*RD6Y|5G?2FNmylM3}is!iM<3<pWQl2?BHvJQ1{*<hzGUKEhPb2Bg*PP!b6z@
z>CJr%$Z5NdN@$q)&-rJ$P16ZUrE?iddI{rOY5#pRt?a|9=~xA2T>?C>D=L7~`ho-7
z<4p{(ZrdRNo_f~^Dci~sQVad7V&&-cqYlJPz9Xi%4d0*d%Rqdgqa3hRtfd03(W@lD
zn&?%4q^7rID&67;_LqYgN=Pvt-PE`gM`89D9k5?IK=9gChT=;xK7Lu1Zqh1La~#LY
z^Wpm{5KeC*1O9o72&SS#b(GLp<K+#Qm#Qe134<lTRA@dGbDeAv@}J<diwY38-1r#B
zHeO9DZP#iEtH7879I&r6J|qd}+A1ia)Wc$m>jdWPPyy@W-7;XyD9V7y`X~k1bbX2r
zu<W^1wsvu=D7F;^C4iMGrvuiDqf{Vy@>B_MpE@s7d2~euT)W#)VMWF=6s^)m9Y|ZF
z%Ru_IR-BOdQU@INe<Z-UFiHhdyH1v=>^{c<N7p$F2+TBYBdolmpd@_g7E?TBPf6p4
zp6~4;GK%|bXIeSkcc^&YKcZk|op*r&?wu(zkg-bFfpE`83Xrq4v<z4}R3TW4ni-1i
zGvk%_pSdA|cyveYl``Pkc1r?S$-82TmHBB+8Ss=UCIM-Uvk2}h?Kz5X>OL{W!I1(j
z34z|#6%=;lHV1+S%Zr)+ObgZ5fe3%b0ms%~RKQ%%`0;M860f7!++7vGda@@6Ld%zn
zDXtS<RaXTvllBu5fB%)>YBO5FN?TNn5V>!bP|U4+6Jkr5brjFhVH|LzuBDP*dOCp}
zZ>(Y!{-&vz;z8HPER+CtCW}g9=lxVND~ENg{O_)&V(+Lkl+aI39q^_sqLS750Y}Nc
zX;T45bP)x}nD!pQ(PNvA;;gzs1{{SxPyl;oA%eHkMTTNJkthLiyUHklTk9mIxM#L|
zx(aNV`cMJ_<Bk!m<x&W~lg}7dk(Vw7@J#P31J>I;2%BcyWhf!rPdX4=y^sR<Djye9
zJeHc$n~GyuTN%ZwC2+v9s3ReAWt)VORQE%It64)8#XjB1fRNIb1Ll~zRD7c!t0;jZ
zi#g!<F+l=cCC-W|9$gz#P60AnT%!_@i>N5}A)j)<nLAtpY*!*;inGN3urmi7^QKa<
zJ-I2PWG~*w08h)0sH|u?g^IsfIk6F^o76!P5Q?&>fTMQ<ol4dSF~u>=UsnPASNF?+
zvzx>K$CD@th}Z_mfaSmg2H3v)R|4XOoS_o!n?vyKydp>CU)fu#wMzJDB?pqT`w%=g
zUlDf9uOVUOR=<*gq$(G6AWL2*0ij;T_g$;fkE7TwI%s9?YrOKIr70>_>343>DkIU&
zQM5(<Wx%#>uL6X#Ix!%!sF@CA|GHUBanIWl?x=v)`)?h{JeMbC{tMN4rwsV2HDW+O
zz9ptu<<z|>1F=myP>GD0#ZZ!yA4x#y$-83aKSZ5IDqvrJS^@$eO(vx8YOkU=UldXR
z_dD|$U_J7^PWX5j!Tj(85yS-(P8)A`($fAMCF}PeXqDN_c%~Vvr^{HycAZMds5^s>
z+46l3$I3P-gOEO|k%ST*nxz7kT}({zSjq2IQ~}TW(lU_Rs~o4Y+NS~*dqW-YT{d2^
zK+Xk@qSd~q0L(g22DTiwNr3;Oiz*PDaYqI$W9KM<w#N9rYvqmap!f779jn08$zqBd
z3k6jP@T?f60RCF*#1t!*+<^nh%e)d`UYRQcmd)RAAo}!b2H3tRApwCs2gMYpo8vMW
z5FBqj3rC++D!k7x3RW3ADiX4P7)0<D$W^h5JrJV<ENYJgIJ-?HSUQ&ADE^dO23Qjg
zNI*`Fb5z)qtul(EflMV_WEmmuiATrE{&1!OxO2~|fa{;{B)}WfoKfjnLkEHbuW%qz
zVXFc-I=$9`!1dE&il?0Rc|i$Cz1xt$1{{)6tW_^6fTi(l9Z2jpl;AmWl#uqP#<9{K
zy-UUQ4x<uw-Bz&jJS$E}9cj@~ygLubfTvP36)-=E;ecIjq5zt;s17(ig=D~6>^KL)
zr&}?=vGrdaa9NF?W3HAfRTTGwxg1~@2kU_4M@dZal!FB;sDNojdmS*3{F;i>Q&mAR
zwd&3Re<|a!JcYcpGQZSytOEWPRI-=W6G2?rHr9CQ0^fh8qa^LUqyldL#~kpyC{88$
zUE^E8?oXjrp!#35$~^xy75DAGl&JhG`%4X_64O0Q@I`_uR>}7}D}d$8NmLx0>Wd)G
ziu7Ht12K&?g6CmjjuO${VE~){QcQ7hTm$1aZ1e8vD3)DiRUos<4GFMr-l+iQneQ^d
z8JfreRxewk^6hgm#RW~Dhh!jWPE#ESzB1myl!om&iZ$Fy1#C)01{?(#AwYkZG_KKC
zy^0n~srI_m_z4B0^8fs*SE0gDQN@ah!~fSGrK74xO`S1kPUA82=Z>2(d+fN`-8xr}
z8v1|!F?G`1xl_mGUw>tXtCfm~tL6Xse_y|3luXzEZrtoSlV(is*17%v$JPJOhyBl@
zd5QeTK9)Y=>w@BO^Z)$6AG=CaOM2`^V@J;&-FVKp*^@?3owQ)|TyY<xCXMaZx#|D^
zTWQj&d8-yJTeoiBx^0VAEt)iK(z0puCitUi(>5)|AI(~{YSOYvoBX?+G4I18`5&wN
zKfmf#Apd6n^B(-qcc)bT;=lgDJCc9Xjb=`n`oB;8pD%pr|2i#8yw(5xX8!MU|K|%=
z^1se4KhZeXc%%OJng8?2RQaznH@`Xa|NUkDs#lRB9~&oBFVyfqr-~0G{&+L-6%eJ(
kB)m}K6)RFS|A1zb=|2|`{;Jno`5(8a!s7q(kEdh*AMK>OI{*Lx
literal 0
HcmV?d00001
diff --git a/5_dpnn/results/serial/results_serial.pdf b/5_dpnn/results/serial/results_serial.pdf
new file mode 100644
index 0000000000000000000000000000000000000000..64b7e951ed51e2c4e32acdc7bc45f2f3835c03a1
GIT binary patch
literal 146143
zcmZU)bx>SEx9*)F!5u<y_u%d!xVr^+AKWdtLvV-S?l8E+;5xXw5AOQqJ?DIN&#hZ~
zS9Nvor`P&D>#x;awW*ZEBp6wlI1s5S*GMbc5LroCNbQZS5d{QDnbkZU%t)C<ja-au
z?JY=|m5eORoJraK2~<dhgb>XDrvDN0{I3lX_5c@B_Wu$vYbhFAo0+(fa{u>K+`~mu
z)y2rgjFjWQ7nF=#T+EyRq@4emA~LI58kt%FEJ(Tj=c|*wiK>|ksm{N&ViNyQ%sgC3
znWgRiA^iNGQS^UCX;Quag~R&)8u$<Qf9Rde{?Yw6KC_CMv%Ra6$v;2;;s2Lk-ptg>
z=%>BMzl<#Z9#&Qk4ki{JQVzEN8ub2^;h(pE0h~#B|C>+*0I>fzWBrf%f7jzbpa1^?
z%9#NyTr5dh|0k)0mF+*vq|6ew|LP!SW@2w@_8)R*7bi0#J4BDH@y?NIJ87pJ-fl(5
z&~)hR+b+CFYj}9hG8HIJbPif2Z1KQ+e5j&D<`+M99g7QRQDZTYAb9dL?)P?JE3?7*
z!NO|p`@X?PxBrW#|C|5YzJbqcl>f`FqQS?V=Eq}!-)r{U3pdIB?#0>W$Ca}1`|SH$
zCduIYsqjasYxI=U*_8jst0%|X)aEM16v+oTj^yKNDEGbf?UDqt-2HlGazDjuefnln
zIkn=Ry*ND4BsBKTZWH(MVslGF2oBo-p0ZHneO2;ZzJ2$cnXjtmhSlaJt0=V&RDM%%
zakVVTUvM_RI^_Rx8u#(COCt1sJC!++$o)Z^3*LCz^qrexhX_v&bvGY>z`cOmiZ;~m
z@qok^48X<1y1ErF?<QQ1!<XI7_eWvj_e_*UU+wpUT^lJK%0K=YOK-o5N&K&FXziyK
zh@d}0I@dlPuW<d}uS)#&gil^O3L3ehM;{+={9o%<4{4p(Rz@2+%X5L37L}?)@Bfw`
zufnc};)Fle<NWU|Cf@yrtUqcOKR()5TvMlJLADqC$2jkdr*90yoi&eB(4E3>42vDr
z&sHBf%aD%+=I>rFdR1OJZw>PcE35RKf4j{lKBSE=7~aAEc-}|v0Ez0Y&MEU_?;&@@
z-4%C0!K>wqPS@_L8HQIKjaSV--Uc_ha9;k;n$=z;FV6Stpp6NskB*a5_okD64WpdS
z6AztY&Bx}-=Ni(7t`lLj+xOcV-8<INlBy@7-Ns-3M)Ok_*7Jv!x!zL%Ro_Gu{jJUV
zlkUK^m+}vR=@)Rf)4WCdy)Ngu${};L7vogD4M-TdHsV%rvnGex#lLzn^YD~8C+ZNX
z{P5=ByUqt;XXQn(!IpiJ!TSN(TVUrbS7bZH^MW6O`$zyu?e?_nHmqs1|2Uf{9Cc^9
zL-a3ryXHoN;GTuJRh)vWb#}!sQYVpK6<+S_b8HS_y^Gc_4y*3pHba1#|7-tLJ+ueJ
zcf+c?XVbF#bwcz7);n)~>`lNd`R<+3VB3AA#{Xswk$H&$vLS3^_0Tn@UUSL4!t8%D
z?boYsbz$>rntk;jp?%{&%DjJ+(?E(ht6j+d*?oD2dzS2`lMC@PILEyu*stmKfcF=v
zagqNmY%%^GbBXjCb3L{xP-sxAxw1%$kEC<4d4=;X@v?7Eyu!SM1ksK$IR{U@c>0HI
zT6Moq^q2h;>R%6*=`;m3Db|f$a*YJOI0!57j{-jnO<L#_vnJa(@%TI&ye)oQYknNN
zyNU>vU;I>UKq~7xoiw<t=^+20f$^8{UG}%4%(>|13Gn}kzG>UgbZ}<lC&v`Xdz#iH
zr^zMkZ_qc=pf(S<t(S{?G;i_f^*=rFr+7lTe7O6#k9r?Vd}K8E3w&CJJnCXik3}l7
z_?P5v>8~>F1K%K6KHJ?-SRIM4u#Y8*m+$|jme1WnrI=mevZ^3V#qqBx>CxIR(aT;}
zI9;XrPi&e;iEG@gi`=bM=4WQboo1gvb3)tHdeCwM2=_6Y7z*n_IBMi;Nw<Fh?rXDO
zfIX%E=l6HmuDAS;Yf8T{+#^tLTOH~7&PndyH+{5TTCC-|O8bXPq<0<RUhahMPyQb;
zn|?F?o|xVHSjb>%*>$(X&0pR=flt*9UE_~FOqIxdL12lG(07i_#luAwwoSv2UQz#>
zfRc&xn0MUg?Dy3hgWK+ptvr7@dsYariTl~$$o<)V%75nLJ*XzLXo=*a>Y)SW66(XH
zq2q1GCOp^gnp0vuw==cj<>i~frJ~8dRkP>|;U|2P=qX=_7ygBbPp@#y)qPYdYuCq~
zO|++wh4b-;O)!lAKTEercMn6|mYK%jgPa#%iq0P48UL5TG5AgYYfeyIVhFtY=FnrH
zKmINRm)XSUV`Pz2`{gjrF8KlF<sX53*FS>j{}6<-23yl#`E52Z|6c^y$VBd47bUol
zC9E5MS8Sk%T%Q=$p+kJ{rU#-ss1FX%DziT5@jbDIRpF(wY3=5H!eA>G<nfQ-+kX*c
z=K{lRx-QoLcL2TjeZNUTFaJ%=f4NdWR++m%kH<6yCXMgQWAtX-aaXYaJFfZD+UdQ$
z?jG#BRdlmKNA%U(Kk20stKVyv`~RIwD3^Ng?7{9X<ej7M^oAMQ=cO0-O-J|w|E=Sq
z>>mKee;f?_b1?Gdo7u|7+naZHkFYeHeYXb&TG!j2zu6)kwb5IHbzt&a@x0mWmOwG0
z8W^!!`crdL)Y45Z&Y0osMEA>Mf9{7R_vTxN`wFa<{Nu-CzyHTBT#q(&IM7e{{qo}#
zBK)6sf<*W|UB3T4f3E4F3bKp#G35U=<o{YA9qa$j{Xrl;7aP7J^d3_<h+n_cPWrsF
zbI8C9=3PXb`e;hrzO3b5u;PAP$&$qpup4Qt3*fDIzZC-Vac;GnW8wkf2`#+uj4yD+
z<bLQgY<GfO9yhq|9hQJ@%P-LEHiBc)1S{u28%YYFzV~z~bvF<~vCyP}_V{hpI)A1l
zqD*8Zr2#hH%s_O7__Ec6Paw*7eBYrtk+895;^XZ1MaPv(`aN0Ta1zO;(JCrcc%{?%
zt*XaYosn(9j55S8tDtXgvM?B6f4AenxT+#@<Hv@Gh`^H0A_&Msc2&7V#PrJX><u~x
zY_)0*yf{@{t+c&?`QW*)^f#jR{)HE4M^{s=7i2>O4E?g2u_PcPpm(w4wW47sO&Fw;
zrrXm%ExG*bBK~q|t#&<_L=|f?9rZ!Qx5o^6C3EmM?X+<oS7(ELG<OAK66DFM&*`)s
zCdR4&-O(6rwQTu`$!2wk)g#Sq0W!5}1fH_PPIKnwwWqXpc4D&M`7yau2_?(F4%{Q=
z-q>3bnsXR*)1adWZ9jr5eoa$ql=$AAm*v0N1^gu&T$3H3qA;Jjzw8W6?ACQQkjAD|
z9pP~0!r;b%tnb!6wGQZmUs0E}sIPTQ{C$WMAAzVR@#h@)RY2}F5FwOedR0z-0dv`%
zh>CU<IQzT77vH-!oA6Q#AAOW7$|t>UmHMmkk|ry}7zX0Vtl1L&t>GJMr)zP}ImCH2
zay@Mm81tR{zHy>$@wa@5fJC95Ba|=;mx3VN3YR^9N^{Nc3k-;PU(Fjutg{>*XY~Y)
z{V+~(>&EXmJF{^t*v!MO&zdINKy+WZmktSmE%OrD%i44~la|k$6yy@#6FP1J>vruo
zF3pTqANC-y@4{p(TMEFUVlr)OE3!^>I;EQD>T6I51TE86QYahbsL|7`n@?&xY}{_7
z5N1cgnR5>OY*XipS3A*P2puZ$ZMeO^m;o0hqPaENV~HnO;_S?czx;};#3ja(1g>`A
z^(-C?A>9&PUg9V0N@fp0u$Tbu-+XDh@q$Nc(0(Ap!?|5S$oxgKo8xhWTn*afr<6;t
zbYnc4<(g~BAvR~XRZWrXaU;2@&3<IJb>_3q6i=+6)d50+oVXW2{p($88+=8fD~?Uw
z({Qzd94^EGkiWT9Mvk?_Pq%u+=mD0dn?OE+7|}|4@->i3)Ne=}dfC~ZRt5CB?oD)~
zD-9z4@U+k^_VipgeH(>UKhL}v>5}C;VIp8$Q98$XtT&Byv=j#yLG>+GZJnj%(y!I^
zb$dC}RWpB|w=>+Sgr)N}>g-{6Yvh{2SGmV#<_cJW`Fp?dsmWivR>7o+4Lf7HFA`B;
zewNAG;)up6<FkuL?QXZ8!5;y3yHTOOcQ*Mm@Jk>J3>}%w#4}|}zU#j>c^P)61dR-#
zrPM@y0PXg4LJ;9Fco3le;v7MSQTw&zfwV03t7wLjna*mCKBxdtbJZB3?Is*K%o0FG
zZ7eClVXNIrb!}(MuPA6+VYet1RRU2(i?L+;xsQta4ew;w%j%^@Hk##|<R?W<ogaxh
z5(2)qA&clHwG5k`oIT)4j_1U;8PtVnGCNVdoO5bP)EBhrckOZ_eSnr~utpWlwzAlY
zgaA;_$DRP0pf%qJ@dKaHkbge8B*reI!+H03RJwC=O3ow632Gd4`^1Qa{l?=Od)VV-
zQ{iiy%Jaa!biPnw=P9cY^aB&i%dQ}8+?2z37AS4x2y{FUDgin4Gqr8wSLj&?4zD0$
z=dOqop&{y4WCZEi(fJamk8OFM;kP3xj2oxd1|8Yin37n&0M8xxV}`;=e@!U2M<IDN
zuyU>nP=QuU)y51*`q>ZMQuLsU<J08quNRg-8|dH!Y5%3ST2gKYEzZ8AbFlhEoH=E{
zjSARWk6C^u$ZWBBp<@$Lckfv*d^AV0CQn_5^CQEvy-;cHI%PEi+sk3cjNq)vgS0h$
zMvZnGp~<1^%w7(GnwFxX_!wA-a`p%w8YrIWe@BE6R%56CNHy_uF^@t>fq~5&#xzoy
zN?Ta1@bzQMWaOuQ5{*b@AaF6)5`^(3oahk)dPP~Bm>qBQKA4i^2NSIK?Fs?v3tOF>
z8}@S#ExQr?fElmt7#Sw+yvc-C@mM_xBv@VEKHO>v#E)KAWe8VF3E}6yF=SZEa~-3r
zEpq|#<Cr>FUv*m0J1`1~<d4X>r`TVfjEhl|CSGS#e6!X4M80@_<EKl>QU`LeYrlFr
zDkdIkS4;!6ijsh0LDwj@nPWV=^+eHU=25$=_AAO~S|1SzY{qrs)F1<JL6b^BDkiwW
zn?bS>g>bQh5ClC7T%kVaLO6e&9>9&~N->J&BIW`v=NVbY)m8jqI7`vhEVL8hYG*$%
zlC51%fX#`IYL=5Eu0<QrTm5rIUo5l1sNfr?7s^VUKYtenS#8#(S}m63cwCPrGX4%S
z@8FE;vK<3udI*FQ>jPC7EthD8(2{;7vT#`JCR1Az#H_!4gZvo6i9uOpnJMJejWVUg
z>YO>X7P1?&<gSY?J?1Kg;kSL=1@J_zTQ&?{g{3<89?e%s%kbxNB`<_uo=n7M*l&e=
z6&4$ywwz5iP=TkEWN-C3{Mx+#bCIe(17X1ixg2gSFJJXHg2}+?5@)@q7Y^!!mdNVu
z(j-{L_iQ6JneBK=xCX^aXDvO%P7LjUtINdOGW5^Ld(!h~J5_m$SuR(#M&&a3NPbp(
zENe0EF&yfunQDUF9co;^B|WpT^+@}M7-L%oquSVAr}yAj;}dgS`(E5LZ`T6og6hl;
zYn$brOuUYE%l^+N-~B{aCF;eoFshqoUi<BXIZ!;|@`_mFl)pmbZ*nPY%0^o}XMlEv
z>g>3n2X7r3ana9iP-qzJ;H;MSqQNP_+c2Vvs|U_SQ(Nj7Znv)T4lD1l)8@GJ*>+Ns
zuPX?vzBJ?=m8VUA-9ln2l7S4A<`z5Y#l0Dsw4Q;1$A{v<xIQjZY1vhtB$HXd^@OKL
zx3YZ>YduDdNukJYqj{#Rx=FVv4dot_H4_0aakHJnDr0WKC-I%7SQbb|zm)05Eh$P+
zMMMP<V(VTIc+g<^R~30I_Ro4ZYBO6ln|E_76tYkBTkrKRJ%IJ6#{MCP;wb!wv`Oxm
zuu;Fh)V}XpihtpcdGOC<2pNV~U@K-)<ely`OaN}NF+->0=P|^GwZ~sUaxRIhZsD@u
zGUMvX$_Kf#DfdWMSEwDY8oviMzu+c)rCzz;a2ra|rC8ll@OI8)Jd|@|Hi9knyG$*s
zx{AEt5bBMp#C8%W=Du(2q~fA$e>0Y9Z<1G-kxXLLRLrzxu~S__$feByMtS#?M##VP
znAT8uZz;1hgMo?wfr-?Uz78>+jH*s=bXFrXTT4ysrj`$;UqzM;;8VJ5NrGsEw36*o
zx4727(Y*DRQvAT9UVh8Wjm=*IzAaNk#KrN+)l-JZ4mBQQ0mZ1IJ+T4FOWRfnLv9PE
zE7v0>@`MJ1ydh;s6<_RLpyBeXVo<8Gw6NNr#R=3WU=nn(b_r0@N|d3oxI{$7&ZwLE
zeS^XwX8{+biSwEYl}V7cU$1{_1b?d4UM$LI5p?ahU*o*Lhi#liez^enDCa}fS`DK2
zezh1a_DJt_Low_y^g}9`M_zg?3jd3K?;R9gcSA*PYo!xT;Ltx;)6iw^NCfl9F=6Tu
z*%md9%ere#-9s&qnik=&zS=X4Wdj@2WxETDrd4&=2+`_a=J(EN-%xI+|HVTTvwq_C
zNe={FycK%KM*`GA!6ZN^IrB(6?cm{V=)54ux%{yfJaAU>CB&-oYhRR7&xYi<U2=I8
zs;q)@0ek6NbXa?~YvQZ<w3f5=C_LPi70S9_2&&1aQvHbPa%uE?a|;d)_<}~OV-#={
zYal^|?L(aOH;Mk>3;wZ~b)&4f)f}SG$;9@4K<NZnRu^#eH*-zd_Ol@n8oX6WVZ?8p
zuP5v{`4UwCw5vO)vR`|}!nZxaQ&fD6kmt++h@xu6{$8(IKeS8DBcf|>_+_tK%BDg3
zX?Y$_X1EtU=rT<Hr0&hcsgcpTo6|6`ToL@7ZCt<W;z4fDC45*68aw`M8;wIwUAt5$
z8BqdJ#81d{426suf=%oI83y9%Q3(0za$Rji{@~}qz2D1Ok2hw@-}o&_`0ZPg+pE8T
z6QFa5#ytAz$1(;_3Bk~@{AV%3bC@?E%eA+GrF3`UC_0t<5jg}la*0)d5}pGxFiJtO
zAPfDI05sPIR%7#(*N6ExD1sHbHCw<>D(DTTStr!!qSK#TbJfDjXtv~9mG;p~+qHg^
zFJu`YqeO49Hbrd(v&;@;@a^ru132r#Alem_4KV_<#U#4HL999PuydNmuE!9u)eTR^
z(jpfnJGS>#k(xg2SP@hK@dX1KXA2Ki(<5{tmp#?in`bcA{JCN6Rq^X!exvrR3aO=%
zX2!G9s;<xmGyqb$%_kQ?4BQk0@W!pjZLT~QB+uk9Z8<-==%f|j_xZsWy@#cL6^CLw
zaF3vHt!O<KTPO&ajX48=WtWTNZk#IC9N032GVh~i2S{Z30B1~VJ_+d+E|(mpJ?_@7
zQ!3!2p;qVqwxfGlc&%+25Lh^&j&DUaLiQRQTiW^C!nq%+_W=A%uz7GCQR3ZNF>bhP
zR?!w|T4}ftT=E_5a-~lfdLPq%g+p_j9yOCgH?M9@5NPsruA33wS%8q_42Qw*m@(X^
zXVq^h{^cZVr0SmI;KcDQs<$`gz+$rprh&d6wH6^^Jsl6T!(RIQz5)SN??^tOvDcMU
z<5Xtl{+z~Wb-dz*_ZM)96>Ifxaj<L(;OxNzUlmi#LGw(<^pSn3Itv=3O3k(D>rzs|
zE^b+jo(?+e=egi%w0b{WP{g-ct#Hd#CRT(BUdu|cBsda#&;wQ0-v~Biss3}q%eMDe
zsqNw)|GE#)Th2h=j6PiG46%^uk+E+LFM)>xH$A*&5;PhF?OH?`5oOh)`Oo_~4ec+_
zF)GoTgB7_#zEaV66UnHC;6*Wl<h=a`u9%bj1z*$hyjxb<)d^(>yg_hrpXI{68OwEL
z8eFTQQqIl?F*{0h@;EGep<N#J-pYYf)A(Zu(nhF8^s#Dm@51h|2W)iVe(t{5UXg7!
zD7P4xs0U396TY8EPZ)?6MDShOT`M0DV@)GywN!$RlBj4axslCcE2av(+0=byo15gw
zTgTZ{LIarhvFqfW@Z_gGiSr>?eOnBHf7#<#e$iO6uAmllAD>jq_0Cp~g|y+7mZ^+A
z?KQHiLxJuCM_QH&9XkeY&&5pdLbwzPgX++wxy)5xbp<;rznIK)SMMzClcb*ZeNu*;
zIf&X2_ZwA~1u8%2)Cy(kfO<0Y56It|e?!1kRaTiw(+Yhz)>7?}HX`5A9=BZ38wO`M
zpg^=?F>@NODch`gJ;QqUE}THK{#y^C)w#9Aou#2pw7*Myv1#d@Ki00lugr)01o1et
zg<_}$m6~NNX<b!HM}4D%{Ps5v5v(gOk&}lO2ko7GMRIcJ_3$vMBij4P;ug7kVZ7>9
zm-N}75Q~+DpXr_c6`eU#NUJ%HW)+6zYr;0+k_L#15u305u11!?dB~k3rU5w~khUh5
zZuwc!Efl~au+g=S;DU#bdBD4u`gM(;udAlXii0BbEX^#-ZMjSv{X>NltJ;cS$el~`
zep)Xi*SVD$X3hi6Z7RQ~oFGZwb~iyJZH;?4W2ehgB}a61v`*$eqRn24|6^X52XHW9
z7_u(=t*wCi-Kj`D^JlgroGms(;J{s#jAQ1vTIs*?SPh-q0rH{5+D7SOWkb*eKUk?8
z(&S^=i)W3mn5!%3x1BPN)OY0_Xbn?)(PEwy8~E2sYAH6n@Y<b>g4XR?oMH4wYwXTX
zLV(upA*_|B9XW<X?PE58k7=)H@N#VEm#**aTxNIOY#+#$YJ`Z?oiJ7}=o#>vrb~F`
zp9{K~Q%7qX2EY<O8zFXbrUp-3z>?`sUSP0F%QI<St#KS@%Dg;`f{V4JlomU{*~Bl)
z&~mHgXKV~kdWDg)%{7wxg9V=L7%90JeVU<4AQ#eiBnMIIHNShb=Y#{zcr<Sk5rUjv
zuw@OkEHG_^tYug<KMFywOEE6{FyhCMEULhKzWj7-mBeBKHkUMt$H+NDd{0;Sg<&Eh
zWPA{u{q6GHVRwa}BpDaT-^*Mcbtfr?NS<MJDN=|<k5OhkkUsz}Ft%}XT+<qiD4CfK
z&2=`p^&b2KhABHeTvIrAsdk;YAy6#%x<t1rGSF3Sxhi=YY!5q{+V;$|2Y3Ndf=Zks
zNVjaU7@?_tilkoktZ>*@=FaG8(Gv8+7EOAjt0EL*r8ZKi8LQVoP7Fy0KEaxeR|p93
zm=u7JR%~`jY+xjp&ao~EX9KdmNzfhFcfm02y^AR(8z<MLhQ`81xNbq4$)%dIzwz)p
z+-GYUDQ2@DBYw%ES{Wm3^W)rLWayyyEa}S3-!;fN@uMje5{K3I+SZ6+Y&T%d);McW
zupq77ZX;x~nHh9PWa=_Jql_K95NYtVjm*9hB=`!&*#PV_X^K}>e#LtK`Q4l{dmq+r
z6l_niDpY=UUoXJxXP2?WiR1=hLZw^M_{s)+!j(+`vr;*UN?42BpjrMTJCH+(8U|B%
zDe#244zEf#N}o^$Y61rqKJ!qA4PF!YMWoMNH3#Ww^Z?ab_O#48@Sa@ych{T9So?E`
zJTC5YBz=;#qSH;!+U0X-z7GI>zQ=A;3q<*-k8k4#WWsOesn#*3W5>J50p?sB+9-?;
zb(ZnKEkrcNoy}ij6Td{6!~-vekk?v9mAyiD`d3B{m*bb%kt9uW;w!8gdi9m-GP4eh
ziu)ou+Wef8!K0!>v0<PPUaSj($Es2LKl{b>Omq%9D?156b_s27Pob3KEfc#!3%-|(
z!2V}Np+V%eV2l`>ECiYj!3dH^4Dac#CmssJ?>U1F#t7Om`(!O89O>Wp9bzVaI4C|0
z?TZjh?LO|t0y7{pHPQ<FS}b2s1K?<3g%J2dKXdr-OZ$?ct6gF2>YVQFB7LAn_YM=J
z3QH5`dOz@I{q4+THj_;A)=zEGgJU>ewctr3`d9mZM;{sw0(NUML~<yBkc&YdLn+q?
zPd>lQkr?&BnzHSN7wM#K&E1sht<pmReHS4MLx~M}ZLbZ~kRwAf+otFNyBkYgLeApa
z+#X0?EJh+#U4_HizeDEHuV&r_IX2pDFP~ah30V)5n|tj^<wNAGG#G7eq`WEy1r9WM
z64$y~Km#tIy|d>F?w3Qan=0Ek`1(axi&Q&@PvheAB@{uAJiOHtpZjra!PxPbH1VIv
zcTfgDpC9dO@0+nr3&?Do1Q+c4oIUSqk^f0M=dqq;6{+}H^0)Y`&UyZF$o<f;dGmUF
z(}ihf$Z@s^ia*v}#ie|0o@#SUO^Cd{_vHlQX@I3P9pR#xz`cKimnF=0s0Fuc1Hgjg
z_H$$@x^}%tf8wcI*$=hU6o{03+@1w1SN21{|CXl(O@v0PvC7>JD6@v5wh@&i1d($^
zB9gU?;e2x756?i(NU@r`qXs;L4BrTf+I|IS=IbM7eh^N5N7&}=4`Er=;~e5F>n{m1
zK1!iCT(t(HeciFo=Rr-`z>cz17!)GjP~$nM<0s?cWc}SXfrIYgrJLI?_J_gWVG+0b
zp_jo5FBu(WO&YBu#XOJ~^yrLg-KGg)`Jq;_?yWFW+!VY^np<(er;jCo*ThBFL$$bn
zj`Z~JP9c0Mz`x;&uGI{feXJ$XPji4OsJJ#baR{mY<$G+yJYw^i+2p(IspYK#3t?a?
zzL7Q|EZ9w_JAj)UG5d30mra7b=~fHTGG#15v}1mWrIzk==jI<&oaNaRb|$sI$+r+{
z3!(4zM9r;OBpxZc*W_X>8Zm!Msb%>}7K2-_1r!~D)Y~n0(=#CiSVFbrghOs*hd#)I
z8>pvj?{&Ek<b=Nu9V1iQM8nhOGRdJ7PcEBmICE)^i|w;9_nzMcx+@NX;w>r)CATy=
zof?G-9xT@HD`j2e62$E%9@bwB1@_q!@~55NsiAFzK>4-}>lT7>2)+5D;YNsmAJwFC
z8jZF|Ph+!UAw(VNXY}Bai+iJF)GtAXiK`E-j`d1BG2oE1kPF6KZddW21IZ{nX4p`l
zt>f464mat@DvUC@V%=`FzZ8cTnz*Ne!zfa}mim&Gjv1r#LIo}kk7V+&nlNMenITUs
zcgi<LQ?jQOu>Yt|M-RKI%lEfoGo=Wwt(7|dF>y%1rMoFTRCrPt#D^mZkWTLH(^;0F
zc^aPe988ftSgxWY#Y8eTo<M<6U7G~2QZM>MIyvM#8DK?a30xi8P<MzIp->6%V3?)-
z=!gCgI8v{*g3&U8<NeBckq|Qjy|UJ$#R}*%bqFt6kBDP8#KQBK37qoGiFe>mpI{o4
zAw8`_Near~v8<0=YK-Y3IpI?2z5hj5BHm%SgB4edio*>WcJGsENQo+hDID){Y|%`G
zGD7TKLTN7e-oPqjvt-_*ncaeU>KLZreL3SJG{D`+?zID|NUccR6>`Vq%dCoHmp1_z
zjG18O9Eftsd8T0l4%?jHgDLh~hTMKF#briV9eQrK4Lh`khc6{xn4K-8D9be{SltGA
z`xK1p?pJ$a(v+^IhszJeC?_L9Um3ngu%G0ilBBcShlf#@!(i<ND)*3SI`@j|L33~a
zesZvm7yJp>cA3MUGKemf9vV<4NkyjsPv#z8?X=IowH@I;mQ}_T+P%<I9B#h!#6b~7
zC->euhA-2d15oJCXW+Qv6NAyUgTm((uHCNX=y07IlLfi2q<7L7iVJ<R6CaoG@54Ql
z!(ydz8`>}szb-j@E)-%%0lTy7U*w!IgpM?v07IE$4Py&Fro>E~9zt|qod_Jy<S-y;
zwruyVos%35HH(Az99dQIomU|>gKrgJSqEE%!#_t`w&ml_HVTIT6UoIIC9KRqGs9y}
z-lLP#jN#0DnB2yf7MQ_UFiBQ82d=Di3kniOF6!=pq&m5Ufss$|=d~#0W9a#JA3^_=
z1JOPbx3sS<^17!=`)sRM+VFHX-#ASt=|+-kSgAr{^CmKdFszyDOzcAT*On&236j*h
zSAC+erZX=%h8xmX5KDgw9>yk)4mGM*ai`j14slUcF7NBwOqyZta5fEOUJg=<N~ci^
zu|-2Fvyh0R8kM}W47O@5aqAt%gd5f#{xlWo_>C4_v^bVI>q$`N?>yCqS@xXxBwiNi
z5;z(E6>Y1tOUb^e)4Y9ByMD?+b8_rr-Xu*D7;bLkInp~Gj5<DxuXAA09FcI?%&0w9
zW9#sSA5iGndDr>%Qs1N$u=@kht5+OW7RcS9qNAI;l#@A?`E|C2|H95@a8*NXou1D@
zn`v(VqMz9pv1`*#oo$stbu}&iaAP}l^9_&*7&I6Dv#b25uELmqX7&Wl<tyHQPA0NT
z){P+%8%jlPcu;PK>)Udy44*r6g-E;}aStnaCc|@3UR^b^dhfHLWsB?Cm<@k|9otzF
z^-y2Q#2c8IhUVL+GxnH4jk-wG#$q0X$Adt%8w^z&dqIRnaYducJ(HO>o24_{xR}>(
zvpjTHsJDGTL_jX>Zqsat6+Uzq=W5~pR+Mn#5g$Qc*98KDg506Lar*v5czX<5)j;B-
z{P8?eb##%+Tuz}Wr|9pni@ta|NZZMMXu=OZG^sbN0?e>@M6cdl$;WE%T;6Y6PnWyy
zxtz_KF42=9144Ts(`t*HF^s;Qkm@Y9$wP>dYDEj|B&y9zk9)>I+<3~se#VgjmbjB^
z)b0H>vD<+V1Mm4H#uq?$=A#NRzQr_`ymMxtuIrGPKW<YD(wWZ&PuseTm;WjNlDC<e
z6hsZiJvNUpX>uBgzpYt^;_iHmL>(HwHxLqrr`20AyKfs3w{y^$%bJS#Q%>)hRsPjW
zuU&baM%_%Gxw$2zBVi_WiuvW*sJ2<#Zz8cM5S4e^Xamm9I~-B|i+TDKSk-8~a^<SW
zI8QQEl9$XYl0XYCnufDmPVNoSBB=6<j<b<{T5YG(`66}?BuA%UBj@m91s{hq?6EcZ
z5hBEd`s@@ayxq=Tw2snWZ!y;sf9K}#m2JX&OX5>^`(BK<WMvT)?TTA_2c2x4IQE@L
z^&ja0nHf4>Tbe)M7>`lsi9>lc`V57D4?Wxjx;&1Q#32fU=W!$DS@n?!?&w<S?ZR3#
z3nhrhG*JtylL>oWQEe5qU#2j#q^AL|B0shG6hn@4mCr^OY^Vk>@R&8{k%7=6986M7
zA~0s+Q>36-Ol#&qyr#f^8<9yAFU#c9FQCoE@53D{^e(^r=S*clitwxbWkDS6vh$ub
zd3yG-?s~Dh@r;zS0EKnacb)8nxz!%=U8Y$4FP!OrYz-7UPtPfq2|t_eMj|IbP%*I9
z=K9aMfybEahf5>-?pWG1=1y=5+uthJ3snv6hL@2svr?RuQLT4ieiC~7(Iq(x+x&LG
zaj~u0%o5DiHl^iYKN)Y*2K^uttwPj&Xyu8z{hX>;{ML{tps}<gr&K+q3;X@liJp`|
zO)(F*I^muvyz$g1c?-17NrGJGJ8`I!t?wk1WOmG3E$u;dwB*va6M5KM%{6|JRk_~p
z$dK92*jVBE;wUf@tzFA5lY6{P`3cC-&fS~bvhkHixfJXAHQ`(T08N*eor8%fMfJ}#
z?J_UoWeG>WfLx(I*&oHKLK`+Swb<Nv@t?_kT;o&m>m??ZO&%C&21?s<tei~5ubLG|
z%~ARHWG<b>wMJ3bf=lxc9l1BrLwlQvf~#fcQAH+>(SI|Oz^FVQF=)mRPVbD8A@EC^
zxzGP}w9q4BA4yXpKap%-VXyCv)4QktjpPR!ZO#v!x}AniLZInpvHv5WC*8Jc`bq<s
zm3DLl-OI<dn@j60+(;IY?o_E2#M8R2>^P{|&(=|>d!-F$4E=>_=F>?i1FKGjBqFoB
zpu-^)OB1tyIL`kGr2p67l*G5CMc&>R!W!}mLB<;uleA2Xcfv7D=QzQq2Jt#}pG&$P
z$u!7Ga>uF^)6yJUsO0#b<|O`Z`HzikO(C)A7S`o8O4qRCVo_hq=B><~e6f{bOeRb^
zE5<ti)DCfmRVWjpy<b-L7V<mj0jMM3-wA~DKw#!h2=DI&6JjF^nDi!rU3N6t-uKft
zgbKUL@2{XaRa5Sv3nD+oE#;Qh@@rgq41Z!8JF5`b(h%_y83@>AYb)LmO>T^ym!MX=
zX*ZNho#Z%!>hlWq-}Z>#Q5i0zhlM~LO<FX9{y$Srbg*<9VekTawOJw?1qUtJ*l?P6
z;Y{hgZweyxxK^e3E{Ui}lQ;C%W&|{;MwnmGLm5>|C3mQ+ROP?ujSqAloG+Yo&DB0@
zrkPIW4nXJ7r1@Z!VYjTO2alNiD2yFy0E8tN0ZM;nGf8irOTZMB(~G;EQWCAafA0>v
z!W(QOIup`4aKU4g8Ri)1q{A;{vJEwZv!NM4@oxseN7@Xq(xg>IbghRM3m=S-XH0w{
z*io?%GqP|fwEMnI?IJfgLapGCm+A8OQ!tmiSi_7gRk|EBAt{wfR9c}Y>)Md8yTaMJ
z08gXAKjp0~f%+^<9NKm5S@?Q>b8M2-B7%m4%<&=4XB?`54$V^9b~?$<e~OlE682eK
zX4%P3XqWggfOshV{rgAuBMj&FIwBg9ZqQ&^cw3S8m!MWectXtBFRO1bw)_#53XDRS
z07b%P;^-1|U-m$fbw!*^Z6l7pU_1{R?vir!icURW2U{D5@|d6(FXIe{=*!T=nQ%0Q
zt=8m*l1qXt5^e|8I7etFaqOS$k~>NHj*bDZLp(Z5-yjr#hgqeF-QK#Zuq0!d3HfJ`
zcGS_97=sm2--5C2*ieX}|E069ZnJ<84}fVR(dBwYgK2e;d)-AMCKopqDSWgkxs!1q
z{uF1u2Xedbjq^Al{o1I!9-?mQE49{g?%HXt{O}hx_Aj4q7iqr5L5mx@w1~zHUz*Ae
zklMMWHX3edAHxBYES4+N@ttLkgANCYaKq%KWKW=MuJ&h)H<VjXTwlm?HmJNqfuzMQ
zJ;YO;cbq*Ct0$&m69Y7;?36)0zQt5d{{t6cwlCfzEOc<_G!&Qc8fa2p!HIL~Hp{Co
zf#hQ>Fm`PCsr!7GoMytF)-$^qq+Z9O%Ibv2r0pr%z8qU^p@!J3BVAKO9!loLfgK-{
z|F<HLus@zBqRy`v!SQ11MzR{|TJ+bQuBIU&k$X9pBMAcl(c#2ao(Lr48ci8zW*9sg
zG{S4h?|&vTz)T-8|2rnxiUF;UPtkT+=37<xpBJm0Id;5pq(xo?k)pfaR7B;c_-_!(
zY|cLjjrU?93xo>?O?}n;g|C!=e2Z^x`vz0XCOv<Azqxr?jBuKya^V`EO?%+`-EC!1
z1=*`&a9aF{F&&gO65&b49~d8K_O)uRw@u*n+M@1K2EfiBi#E%0Z&7#XJtU^n^O_K`
z1QNg@$(vQyCB}{vhB(p*i|al}ueE%hQY|vBu64o_l8Ytb-w7{ZX>mwIaR2r90iTzI
zoTRocp+fPv2mev+EJN*^t{&5CK;|VS!I6q<M9O3RYF=y6zSC$zY*hrf<o3%eodZ<g
zuhNVCftXKz5z~uJ{AZcxKJN(#+>QQnm^qHctdy%<4qt0a-Rd#9XHY8+`mU5afg~P9
zU;o-raZ|j7B%VNW{wl187J{e6U(YZt@L%MM4rpg#<H&xL9Qru8b1%w-uMkN4D}{_T
zebb3Tzro&D9E8uN*h+CZ@?u|q{MFO|O?^B*)yR5$)3MNZV%8qevVq4Xx5eoE6@=NU
z4%OsM8@_~#(ACp*j$e;)jPs1o61!D~B}2O~j<e)MiVB4jg&X-vupAvGDqru1zaf1X
zO#FK1d7guh<98;6-we6bN^HiZawjBpPg2WTqT7mT&TWTvzcp9fHCTxNi773|8eaP8
z3pXi+;Dep(9esKj=8b_XBj%a9U3G=o<B+8BKQoffaD+VR{nb~cjoPypGP=M)-cc#X
zW&EshgSg<<QYlV{_EaerQ=-zTg~Z9)D(v1_9`9Y~gbk0nxZW$)r{t_D+eg2qXBYse
zgnSSqMp^^!XGxA{h5)0*M#gzP!Z(@`HyatMUTaFBWYYbIC88;|N($%fNi~{RUWeOK
z7iXY;uGtc$XKNbg>4c!-uf3~d!v%pN>ezL^cI%){r-Q?M-5wWpC^)AHPkoD|=|e>8
zR%C3S!Aq$k+&RJA4@xwXNQGUb)>34-a*3AQSuX^Evf2G@m4>ElwP*rsHOwnHd>1f2
z560aPw|Y#auLFMq`4p#Z7Ws%vV3yv|59}K0Vk}&d5*XWhD?G~fn$n*sw>9r9CuL_Q
z5w`=Q7kJ)wqc%DcE$`Mab1E4LErS(;3TDwQKj^frbWAMS3GtGl4F*hFG(opzrY@>(
z>_^|J0=&Ek9)T~xhpp(2mV!KFGE6Yb%ltHp+lE(|A6#ANAVfnbo8t3C*-xelhbcvo
z8>NFLA*v|4Wg^#Q%ORJQeN-%cO=VryGR0!85Bh6*gCt&k&6;TSMG9);zRDkqW8qi>
zYR(C?kM6DQJpdrai(h<*gPs?){$EN*AP1KQW#zKqPzgQVp$Z4?Okee_!-h&k<LKJp
zY?_bzPX<z*n*{jJ?q!zvyx;FK#o#3HTG8Z}^GiISq99K~PJE9)irMAUshzC8E(}vi
z_ogX`iE&>uPSH)yt;BMSmzP(3;p`?uh?qK7h4pO1<baIoKef4QmZ)}OCQpTxvXbNK
zN;V?Ve$5^K6g>@?<`$>m5;8JTzvV^^&mmCF(z(U0qZE`|C?rxzm`^T{MP88ZASst<
z088D;Ur(6QEbB*<rnm3!63q|V)faOF7oibGe2?v}l7i<8o*8b*zX=sfpF(5v^WNJ`
zGi<}$Tpb!DZ@Z4&t(B=u=<$6KLPK>H_WOOAvENcGYN;DZo01O2K+QkTr}{Swm(I<{
z{)rlcBWCkF;UY?z!?^A+);IvDEyc1A`dx)LW}QdjG?Rkqa;Dz5E{(}eRzhqDh_=s#
zOYIu7Q7-DK>J{5i5%A#i7P*$~z}?I;<z3exUJ@(Wy$+WqVr&`9OTd#fE~gC8g#VON
zu0#QQ`mgbo_UZtm<Q=u<m0-Bci)9fR+gzd_0*}aIvcV)iba{y}wppI%lnwAyPB>;3
zXn5ON?d3dv$KS-EH}##gX8WuTdn2$0cSdl9U$g^Fkg-JfuV2A80<khJjbB=r0z3cE
z)49r46Ez<Cykov!fDRU_7XvEfmz(h2ycNu)`5wxp*<J%#N~gV~s|BWfYW1TB9_D-o
z47^%eN|EFZHN?IW%NSBB4BZzW+7%5=Jbm-mL$shidp-#HIpR9r*;R0$PzLR~Gq>v#
zV0nT5bJ-1%>?%m*<fcpQ32TF{n=q9BviAuLAT9Kx_T7bb_N3Nb7QBBLM4-EdKOESh
z!o#N(I|mO$=F~*hz_im*1=H=NxSZ*gAySBkDJ>3gkI?lpk47KegaUQ4giy2F4}(-S
z%fgcVk`5cv3N<`)%4zN9<8NB$QW3o`uSZkV(+BnTQrVh2KgV6Fw~c2z$e3KPU&=iF
z$!i$Y+dT_$VQF89<JRB<%D>j1$++WW_v-lup1P0LKH!@hj5a(o!Er-W%?-%!RfH+6
zhIwq#-0A8#oZr*fI27l9qVdCO-`g0iq_@F|y1+=7p}J37zj*3d9|tO$hV9c$Ztd0Z
zBy!TtVz~q;+EeW4!?Ta2vUz&6PD*9}wm469z~&Yo$1&!qjo{{%!{>}H&`ar|AYh=P
z^b*qxFyP!Yj7-i1P<BOROiMK2gcHg@&EOkX>GYQlq<ie5d%s4b%sFdiMEo%K5+6$5
zlgGF`-fF7C_#qW0{9SWWzZQY_0`_7kY_Kew*!WT>yDQ`M{LoX?#TN(rRo^M4e=Z1P
zcKCa|6+8elYbQ7&$MP`!U;t)OpeC}g)yJ>SJNq;(?KQ++SABKjW3i$d5x9!TzC1AO
z-FZv&m{;lH5n<k<LWd3b4r!yVhbMPDNz%N4@u2@~l_cemU#m$ouHKLr`kOsU;i*E!
z9!<{=7OiA$PM=P_Sys(ztGMie<*~Y*qC9U*wVX%_3J2qc-|>aoRs?w=PNKM;lej#D
z)1;KeVC(+POnOjhkvvBu1niL7`VGK6bkju*!<>y}EY$bD>F8u^ju{0}!}hxr!8n44
z65Q7&E*kaHI;TXv>K&SQ`Eqt6;qlqESs@cJ*52UtfU;GA%|U1FHZx$1X~p4*)dfN5
zPKu<W)sS;SMbL}Tj3`B{E$weNgjz+;mLd~kJ`2{9XeMw|<70{*XpfwXh?y*%qVq5s
zaSSL{G9$loeyn0L`Z<8b9Z+s~AQoLOsOHS)!rL3oN2MQv&|cxCk(!>e;>!soQFhdq
ze7wt)$Y3)WCet~sU06SFaUXSgQPj*><&lfDT@u-08s3_FY(M4OKOG?JOhdF2^NR(E
z{?Fd6os~F=tk|jZ-Ywepz9TYS#no1hOb?r(WyVvTT%eeL0+TQW0?bm&r5iOHyy2bt
zH@0gI3q+VRNbr*bf1v46aAHe*5(GGMfJ+dWHCnZ-lwE(;;r+$vcgCBNWQJ5&A&k6g
zQ7gWnY+x&`JVJksEFWV&Y)=UtMQrADVo3~P_6B>-LfWwO^Bz_d>gR+03TZ@|Z-3jR
zeme>swkq=nMbVlIT-Ax?0(6@8%%r@?LuS_tMO@6$*3>>CAzw2)OOU?^sS8EcILUs|
zo-W8^nAeTuY^PuR%-YdvE7ElIw}wTrL{Yv1=Um{DQ`}&pqZs!L1JHNn0ETxL8AgOh
zy=Rorb$n@If{GOt#LSjL<&8+JcWgq%5obSM4V~kwq)+ve)goIDwqG$$fm?T=8{{N?
zYhT(Zo2{-S$X0p@XNbU3qcLF~RjT4XR_a_TH~}419r7>RMoj-v`=K0@)jo1iTM*JP
z&vXx$f+Zzl%bz7U$MLr!EHi<BJ5QQWgJm5>?%Isdf(=D_=IYYb2anf1-^G)d=6-4S
zNy$yVfST99K>z$#OI`7fn<06;R~N&Y&@p;a(v-o8RU0xZo#O5-f-=5L1CezK_9#uu
zx_ySQMDeVSki9*_^EYz&RVVe&x`J!ET1(t297oMY{V9Lm#*pV>JHTS(S?X#5(mu*v
zdIYT*#J<;0u{Bi@I;!?2RKC}^Q9~6?*YNXE;{~mWelFGM8?sS{BWyL&{=_)_>KbHj
zJ-T7)j>@4{79mEs9W_L5XV;W<0lT4hw#AMGrc69}8EHT25MsI3_u*q5*K+MG!Ws&(
zih*Kqy9P1m(#hqHVgNV2>bd3^omjBfoxlugaQxm)A^pG&G?u22qHhL~!e91Iv+zsT
zV@qnslHlboEPTrLnD4!yV3KRG48WrJFdk}Iw5}MP*iOTXSGtiyA~O7JVD~OeSp+z=
z+Vtz}2E}nfi@O1=3};U85FR_U8O0ZT55?uMNJ%@&=ojm@-Qz8Uu~hQ&yO!UqO0-bL
z>Bd(AqW~Wo&9hSL@7#DHY6)DN@FkK-I8NctL{ss{5uiz)Ud%2V8~PnEjgRft9EYO;
zF6WdleIXR)u?C$Ymp99L!B4WQ`_$1zMhEpXeXlGL$sLEWT@7U?{h!#>)$sfQf^qiG
zz{9H`eC7q$vs;jIF?v$u4<6ZX@wdnT8iOx%LZJv2oQ@~QZOTJkrzl;rF(cJhOgMYb
z`HJoWM`)Y{Rus9p`s?u&j{^u5Qp~()848Pc*gQaKJd^gm7}+jJo?%U{R(be_uKWoS
zg~fdZOjla*8{FX-Z-|cz_f;(2ulDGKPDe2{*7@!{T=d8qSLK%|A@ci!`9M*<V}(Hq
zX}O6Joj5*o)!T>-f%GU;g?fA29>E9EZs>L!%lzj10SmB?nyOQYr6X?%zI(n_4hw#e
zbBu?qB~y0d#TZG6ByjrGkcx=EqpC#{hLvj}{q$(Mej@pBA!n%Oq+U~fW!zBg#AGJn
z;Dmi;nr@k{X?aA`jC#FLl+IRgO1w7TP=V%$6`!w%>0CUnpBjb+as$M{#jXAz_DJ<4
z1AEx#9e?7y*KSuJ%DG16V(!vXG8}Xn{RiGya5Nb;>u12wuEH;u1&m4=KN%mu3?=*$
z&!$JUg7C!OmX_kL<9k5b-=Bp>?!h4?MwrEadtDCaBQT@4lAih4DaYAJu8Fu^%3bDl
z0)MLtas~>@M@8G(TxuU6gxy}Jcn3C*iVoj%X?U8qo-GWj@r8dF+z9Y=8VDW*cOFtp
zT^8`cJo{K$@jLA3KS!z(o1f%itH$TaLif@>X(aJf100(OXd}&L0{(OorKtS;>7>Qi
zVuZ0@;>P~8O*lYBmnOpn^bM2@m<mCFewR-NXI|#nbZnlxoJn;!xvUulOB+#nu1<RS
zl|xTkN)rz%hOUZB4WFYa9I|Vk$c6tBJjPo2jzsgFeWeX?=wneltV^wq83+GIC`L!!
zj94M8(3Ps&x-s_5Nd$A|Hg6p=*();VZ%j|9DXZ9aOoO43O}RLyEK1z?Jm@6d-T>8z
z=>o3vmOOqmaRFn1abi;TPy84qY0~p*ia6v~^{QXVL#b|oNGk@-7@_QO{Lv5NR$Q3J
z7{4Jq?_x=DHLyB^E(a-4LVyIT7Mbj5X#{?asu9r|Bi=89n(3ct>J+ozwGEC4VQsJH
zXFi)?C8RWS<`!js|94LzAEtMH#7RDAChdSFvo6?Gn(lx}&A2>Ijts?3LLz$?4jLs6
zL8Z7A@hweMSEP5rsbQM_llq~wOn6Qg5jy_*zGZ=HkSQ{jrYpOEmPwV$YbDTnKi5mZ
zxx><y+JSpc3}_+pUb)NtcT6qE8Y{JP(!P}QXid=AW6iG}mi94-X{QZkdF8^>vkPHK
zF6TUouElb|^vd#G{-_Sd5j#@puCk^?K%JYQW^@4D<*oD5zyF!(b5<(2g{G=0b2`{2
zY4Nw8)ihs04V(G*dzG8|ukDkn;#qj;V&oJZSz1lsoS{^PBFvn8)jWsDSMEdWeX$rJ
z>j2H<--}J&dsFUNluxdGs5zVuUqqyiZmXk}uZw8_P>^e9K;|I7>Tj@YQJZD33?DR?
zHq(gGD7OcHR$7-yb<GsP3~kUiO8nQO^CiSDGK3FD@a<30YSfnn_|CKOnscfIqA}5_
zFX~4sTnQc7$SC~#ao-WZ_{XueBpNkVNT}jZqFRkuPS`xx5|%QcN&8u{Y=&`^hT&0s
z9^x;)E9;7HfyW(NUtroX$)O^?kYUSZ_hPd-XvK)iq@7xd@j2U_xy_W0!U}l${cC+P
zlGl&d;C~aS+7Z;eq#Tp&&|?S&d1K%WC)T#yc(%7on4YfTZEFjV<fT6n#^mnp?v?7f
z57=5i+z!Nu+iH@*vquuKwMPF!LNUuCiu%jv)c$m(E4EGPdzYXE&}c+5Jh&}0q<iL3
z{Z%DxG3<LBOZH+lV#b1CapcJqP8|PfD-n??UzFHQ_L*A9N~qD4berv%S!V)Ljzk8-
zd1#e&i#!DpUCWiu?HSSH{+>p0?A+F@qbW^6*eZleQS5)`Y%ONoGmYd~MV!oyjto`8
zwSD=CkzQaxyo}s{f4mfwM1k;`F#RDo-A&wu{E?XZr(wiuZbTx|k}fXUem?b^NZIIa
z-@$kX@1?r9|2Ai*rwE0S{fuN<?#!%MWWvM1K7B(_*Un3VEbvZ$;QKPyF`7bOS+3sB
zP3~b#RO;zLL}+9S=4QDGMeK}@=Z4Lch`iqjw%F^z4KSSF;rVe;NV{HH(ii$p|LP(z
zL5YJzD&esXlSJpHm#(%UCv98(NvEQMQ40nA;x5T7HYv_032NG?>FmTK>ye&Gqhj(%
ze(t=Qi^A<Ll+}4FEv26Hi5>docRT&asDSfDmUIULXXvDJlFU-Xtn3I8pL^`IVV$~%
zEA8^Ql8pIQLm4&+E9G40xCpDu$Xuz?)_u>tax<LWbUl5<MfI7|pg{a#g1wAM_`x|L
z{gq%)#&7@Tut+7nPRiKLme#u9(u%~>v2h^Do4D}3&$3PA%=pn;biCT!>D(xIC}%G#
zu?1lgqowVaITtb;Cjp+>b@Nu5+~fZTK0v|0)+;+BHt>}@>~O4`?4*P((aAG$yDA{B
z9Wf#y1pJ<{JtT~58MkB7rG(l+9;8pJ+?(_Vj^?eMs_H*EeL@}x{TyUS)bY%RW@r)O
zX=@hNhM}s{1mVhlG%@f5zU+)esAzC+qJ3&>kt+kkTY+zqAiFjz%0tMzs=XoFTt17e
z;lw>aS)csO<i;4qMpfJ=xe3wxNchE&;EFa`B~Bk1hVBU;p)ur*eYKBxUI+J4bGYjI
zj`&Vv>0dZyctl72QbQUehp^BS&O^2{+8o?ezwMR}vRa<!2`l0Gp(cD$WmZi=mxfi3
z-pp%2Qrs#9gB({|@@|E7dyu;_r=m`tq>t5lC?R3U9oGddcLUrRcF94hWy;=p8He#)
zY?rEvXZk6z9T#Ec>SHtME2c;#IufnGiC>wKq)+j$lXy{MYAzD2Aing7kKO>43!6Br
zo6nQ>Jzcz~P<YGd?XnjgYAM<Nbgc!P78%VM+9W_Ga8#SQ10UX=MS5IOfDF+GlPf*n
zYtlNPg@2H$oPjUu<wi~nRm5iE3)89lz}ePGhcdUKgFjIPp#wKvEFw1T<9jbP@m0!6
zd+5>XFY87T&*EF3{6XjU=uU%^Ai1vb8CT-u7IgOtlRwlFzoV8&McpOrA5$d<`t*ox
znx4&sDc0%=yYZ+<fahI|T};3CX=r?NI-MG=$GOPv-bBQOGpWmwQgU*ZqtbEMBLqe?
zAs0?FYZv?}o9JQ-_+qhiE!&xlO2Oci;v3pkbCF&ZxLr~j7E{9N5j+9Dl|3tG-Nl9x
zIAz|M4`xLBVaw+ijl(>^lPa<^S8W#=B&BrUC@=_2ttoeUkhVIW$IsHR3gj-i37H!W
z7F|K9dq1mxJr-F4$$Xj4y^2ai9w{|Sn@mD~Rr6IM{t6Ev@V=rh^U(+|DMn?t?~&kn
zM#LTh_vV9AfrE=I@TPWXlF;so<St;+3Z^rFv7_ud_)_s*B|;GEJrlo8babUf*GLzy
zRl|ZEf`T-1*iHbqM}$%h)g?6bzN&OZb7Am~=WPe=*{ZhGSZRx4U-WvG`$<{J^WeDu
z!-qm^(j1qf0~&X$TwKk0F3g<u!<C-2z@bjm(=tMJ&$#2j$%J%;j9TR*=eemDwm-aR
z<#=Gab#_m9Y$lA1By&klbLRw!q%H+gy?~^zh_}EZFcZJ!o*fplkZaNE2mVvvf%EiW
zF9g+Z&1T8nvEw`w!b7cX*UVR@?txR+?zoqin6r^MDG%M_iKt4Rbm$y({COHa<m`Jd
zOJ+-gAJc**Q!Dj?J<y0!>q%x%i`s<|(no4e$FzDMmWwCs5R<~?eVix`vQAiIrkhTy
z2W_;Q_6Bfn<@Blc$Tz`rhe<l@7WQpdzOCKPH|%~yjQ5;A@r(-0Y*GOH62%xoT%xn7
zaY~*c63eK$=a(Wi9P+iuE$E@Okm;9nF0J}JXoKd`aXH~|zfxML0-j1Ta91~NeO$V7
z&cJ(BXohUYz3vAd5-L35uAVP^sr1@XAVoOdflbSh6|yPyBNmy0<I=3A>Li|(R7hul
zI#qPA2jwIuUPvwdq*?k+c&hZShadGBY=KKzVJ=f9?x=)-<c$-|h4w*Zo1~7La=DAM
zoM>1r9B+|3Kj_r9lRWtdD*{4C<jP2F8zLhCk*MFm$)6n&SANqTt(;|rK*?}ox7B*^
zQ+w0^Fzy*%3#3|{93Df4OF!0<=$)$9Vs*}4L7ylcQcb;dY1>Jkvg4#`dIZeAVeLTx
z#17HiNK5QQj@v0R=QV>cpvQ7zCoWD$-EO0j|6~K8)VuGAaNrsEG<87c%<-8qKvRvI
zHA`IPCiIo0LY~la*#*yWODgsstNJm-*}ii&7%(E$f6&7GBz~}4N;V>E!5zAjsx+wm
zA3hX`zKKsRP|1{pB+s0WZA>4RZCA?7=WkaddEKStyXSjd<A~S2B8~Vkr$c?U)n$hy
z0gE{VPW2vethc>H)xoJgV&jguvf6&0r*sJhN6lwPaYW1jwJBg*o!aV*JA#D3CORZK
ztqkG8vqYw&8#Ff#=AQVNqOU}OELtr!RYCnzndDIzRMCQJ;UjJBX8C{ctrG6ilcPho
z$4nB1;d;A?#x)=Jw(fFRQ!yt8-5xnFox)Cv8!o`R-+MzotDe-idhIHbg4(aSWTzet
z>Jo57-AomM!hx>%Ta!_CF2PkHKPAi3Z_MY2Lm5Yv3tl_3+~ldOBg~w%d2(|d_&jp-
z(iz#N8`moNT6ibK?C4V2l1_EAkAJkxrvS&<SS6DYTDFotnWPb0I*<@o;M8{t-!N^|
zPLA7l_4%l)!c3T&D=q2(^|teEiB29~KO4ny<Sc2!u!CDrt~y7hzP`fu5cx`d$S>*}
zdzMJ!Uag5=rKH4-)}FW8vxxpuu{^hBPIcS95zt!knkRYNqD?kYF47Uc81T8MRH^9@
z!Z;E?>Wd>A56m%N_b!wXj~Ah`y1y>5H%Pvk4m!SD3&0m5sNc>#eK!a=#IhWBoY5jF
z>5xIx+<GxvbU=02pn7AdlE{kgXwpDv_tD*!Irjq(inRFUU)v*qs*SUzPyEDO!;Kw0
zPg~!}{vTOaQblQC3-0$XD@>?tR6EVrYn3%D03&gNphpBIjxfPbRRrqWivta8V?FVv
zh6R#oPs&DYuqpD>TQ|GKSP@35e4s)i<S^v)5AA~87i$;O34H;x<ZK>tUdhEw@~#*u
zn|eL3WSTC$fmBKi&#vQy`OxE9OVbHN-ic3-hoC}9a7Egq*VR0(V#2pp;N%~>dxlVt
zhc<TZ?N%2WKR>W!B3%+A!bdO%DQQ@~(Cld_*!1A-nYkDg-INJ_ho%DKwh8W>3LMmZ
zGo9<YB}}qNtjd!p2)?PL=jR*o`~*(6MRS$TB-V2u+%|1kP^sP7a=9ka_a5RwW?{+c
z<B@3*4|Zo{pMekZ-u!-VdCth*QtG6_9L|WE!V(izJVBuxm{@Z<*l}i5Y3b=B!LCv8
z)c6-^+4}^XXs_)WTHC2*6X;Xa{f*a}D{S3MHMbG1b(M|SOsjaw?d|6Hot7JEx$7ZM
zdyW}6wNj$X8(pK^IF8ubRa!$zoM{U#taH|DQ)SWV1;+gxgR``rY<tmob3PQ8lJS%F
z=Ige6-qeV`J&Db>%Y5cJH%%4XsDMt9bQa;RC*p=O#S+}E)aLAL_7a_NNEPdZSbTU;
zdyh~>a9hISHI%FJssblmRmuD$5hDX6PP^?JF7p+ee98OYrYl9Ja##QCOEyd&7xt={
z*1%1HEUvaJo4@m?xG-f!EoiN|4)}rU3L|hWS=A7HNjIE3`%F1Z3YTk=6qNHLx>Vn6
zM{a&6cq(yjPETr@REo{XAlUGDS7kl#B`SByQdMb1vhN*Lpvi%|_=qLX&@X%?`O`)z
zekPC+Z!W?m#<}WchZ^Asr<6|-9e49aD-7_yMQ19ZC^!%JEuOcF>a-*LA5G2<^YLi&
z&25?VEBQ3;Z`}kG<@3$%gX3q5m|XV*V!fjQ267q!?|qgBr)|2I!%&OS6F8-}iic;J
zfL9)zY@{%+lDbn9M*u&WZqoBmP&j&F+td<1p{|(kOIzKO#Bja4$^qVP%Y^unX&>O;
zVNd~UWWfbYmB=7-T7BLMQ}Xm1o?u4jNL~~|vl;4mv;6K*)m&A~l0>sc6hHz;l=CZ|
zGL<c7gNLrfk;dHY;}SUEl0LP4V~ZnmPW2R%pRwKj;AFPAee#mKp17&N>M5oM-)p&T
z&ZCQJckiZ2<MYAo_EC#7h2@tpl7rj9)VQ=6E^(KGQ)b)ErFNpDwSbPo*fCE#tx;r!
zTw{}}=Zz$z9Ys(uK$h)vVD>wDnJon#?YPf&#wGI~mCO%KN6y3~e+W{W9#=)x{Rbbq
zwy^5AEa)v^aXT4Rg^9RLMU%=8$>L|B7En77#|0Pi6@zUKl7(kvrmmSwFysf8d6F_$
zJ1RkdtMk0(c)TrL;5Osx2XyTg77Fm&)esD6h|Y@nys;@Y{_2Y030E7q9r`fSp4Q4U
z%}zjG(#e7z;5Liuno*HH9yp$*7#Ck}$@!CJzam_;M(Y{ivotKRMY~4pJ+kJj%2X-}
zoP?JgmNu1H=`)-_d<LQecR1SW{lck+;I#|a%v=!gi+JLxn&jl>BG|pHc3rAD)WYZr
zt_x3By~!z2*%?oiy;JvMK{UpGb^@DvxkQ3xqvhh;M$mkQENx+Lgr(Dtd`EfL7GYbm
zN~lG++Z%US?7`Kv+ik|JMW-z$LG6tuy{>x-Rp9ilcE2nAnBDL~up=d$+}C^d2-`W&
zQ&Pa&itSPJBNFt}YS=}w4aXiQ8BEE^I-eQHEJxH?5fzH5V&<^qe{}ab>P}Abok-a(
zJ@eG;Fsmf{Dmt40trFTWxY~HScj#m{P;NZ>Y`WxOS4T4aNgH-G99v!rGRdnBQ&j(H
zy*|@iM*+7Up1X5b(HCRPIzr!D5&eL&Xd9!t<-!hAge)&&b-V~_Rj5Lo6fax6+!R4&
z%&*6%Y`u~schpP*R+nkN5s$jxI9@4Z=uu~Mnmi9Ky+ZMA<Yl#lX(r<u$svn2$s6-z
zl2~K#ismFK4D2gOEO*GDD2h8LL)q|<9SIMTeH@o~sz#u*Xo?C}IZ<O$VfMX;(G?{z
zr#RW6ZY;fczVh)&0HlYD*uUlG95?<KPv)mBNG<?jFEzYl<YOiIOO;};{55T3wIok#
z>EY%BdNRi6#=zTV-4S0hBcOeGh27y+i{%~Y0;epkvnT_jyD9RG3K@v$P>mrtJ{eXR
zLo`p!wa@c3H?KQvQoDF0Iw?!5m8eQd6^N2_=#a6zKLiKnCLOvP$U0eN8i>=mNq1J;
zEEUx$QB(t*GK7(bPhX>auSx85=r5+x#pREgeyVMj43f1y>q9N*;o}+}32Rc+hvqYt
zc@oY{a7v&Z2@z@>o6pno_j)Goq&4UaoD}mBh4qLf$aOfX9p#~VJFDlYeA-q?2`)#P
zs6Ef$p?+FQN$DxNZ`B!%1bQm?*L<%_l_C$TCTM9s&+nmrTB`Z9BcY#<2Pbo%G3rI!
z#P}c7^H9vUXZ(=TW{1XIYGfP23ppR7R4?Yi1VYxqy|y(h7x8J~(cpQziAQwcvE`3t
zIBpPejbz<OJH$Oe`2|0tm>8Z}j;3r!#Y3Lxi3m)!1GfRUkp;KFk9wUT*o_jl6wCp{
z(YwI|&vRUSgyBka$B1U4TuCZRW=6kH=85uIvMPA$DA)6e$T*G}6lX-XWZ}i-0Spmg
zOOLXZ<c%Hf2+^Q|T*6p0w>wuTs=V@D$*;U_)~IxeO_(s0s%wB~&z<jj$qaM{cE39^
z-*LHB1?|aIMGu~yIB9X-RfB@ev2Q3LyRxfTdwTXzNxMp7mOY)Y+8TIM0*~9Cu?398
z_Z6M33wDg8-^n@g#9G9gme11y`?`5`@3S79+UwU{!`rTimg7><Xt;UR#Hzf_iB75~
z%(vq_swUw{ReiaL@W~7q;6^yIhi;q}6?>&ddiXREy!5Ni=eGjlbx)FLTJw48{e{)_
zw8wmLm2|Y-x(e5Pmidn1QB<m)ma^1!wbOmW<uO)+bBVh%sb^DYVAJ}q=af!DG(>)T
zwC8po6FF`!ENuDudjz`LX^YzBD$ZO>1lSIu)obDr>?LpW3Tu!o-9yt+T66l6xWc-s
z>3L2=`gFie>l_imf?X)**$6mtzbg`)?@sl=86~X1rocs_8~DW$=v-~?NdKUo<189i
zHEC}IhGPLQ-RWbf_<G%y&SW@1d8jP;Z>0s1Zk>Z}E=Z4q!8rE3tU|RtG`87Fk6dIr
zE@T;jtiq>#VQgeZ;2hQ26Q|?Z6G|V>WU@N(IQX!n@d{NdofRIY-J^3eFh<9z4`|{|
zI7Pr^d{W5Ni89#HGpM;2oR2*NR=kQ#;N(e-K%1&Nh2voM=V{{0OBiEjT#($l?W*~f
zV(~{C4~{V@oJ-UO=z&vqTncO<>^JH>f!kT1ZeyZ#ZOb>M)57!$?UX5Vwu?%I|2z^N
zPLDk5U6qkiT81(1lH*e0Dk5N!g-SkuxoJIV1;xp=H0t@n(Y%$Own<BTDo##fe~rxR
z-7-p65@d@g2H>TTV+r(lu6y7eci~Z;2BJn!!nT4t&4fMbV1RpOvV>3j3eM$F-swK#
ziQQy`@5GHcsQuO<P3c)ocwWiL+bt1&{kE)GYNth8tKB1z9sF#y!BXgAkG_7J@FTSn
zl1Znfs{ni{BU+uO$Ye|0j(UyfX(TXnTq40_z4fl7&->n@V=C5W8@c4Tcgm3{n_Hie
zVFXUQy-0mKHJY{(06^0<Hw|%rxM30esC=xdA`wlTxQQ^)wQTMZ?YDrt+K=cco9G#w
z0`rlIA8fULu3zw2HInvGEMnCNSJKBFDw-cE0ZyW}ikDF6WCqQI?U+nI`gAruNgau&
zxM;;~@K7gyHmXjOvN|68%!rUC9`EF<&Z|r;_nnU@rc2|R-PEUw8;tfsIV^6;6JhjO
zvrr}6=O`YyMrudNm$L1x_El=*v%_6N;ZZiAn~dopkSk1#Z}CtU<dmE{$#gR+E1G&W
z^Kp|TMl|K(%a@#uJ)%+3I|=1`p1)N!FjZTg){nvo6*y`;g@(CDG}-~bN}+PA=?<?X
zG4CcllpXslqT*F{uzGVWYPH_{vIHyD65+vAi#|!7G>zi<f3~+*Y;U5HMvd@Vd?+V}
zCFCG->ghq|!jFPvhj+D7=Yo53+;vF5)_q62y^(bAknnt0?bt}}0KdiQH+8}4IUj0=
z6&D|<;ntJRo$|x$+P~QeXUjaL22+GO+!4YaI6f)Q*|*u`@~gCoI!s~lB7-1txJw^l
zKD-f@O#7InJ)liR$a#*7F*5>nDM%FOmWs}8hXr`DJ0F~Y6?bV~a))Q&gmy(e%|kSN
zz@|_SkHOrSZfe5oiV9jZnYa14$*s3WK*5+RAIC-0J+=%p;WOw#4szUP9=YS2IH1`g
zRFve3T+6$vnlxdzDA6cQQtQdqP?W$7iH;g(p8TWQIw$lBLWN8*>|Ru`BAl+o&sMV}
z{@7Ydl0@%R#jbFZ?h#f02ux{P&60+yiKJ{nZAaz1Dj1UHi#eZdJEXf^fhyc5aTK}k
zv0PZGVk;28y=%l;VwssaE@25Lim}uTy`k{%RyR29UIcl;WK1pZxx~ehX^^TyRcbs=
z*^@Xygwk~NR#Mjaj>IPLRZ5RaDtDytwpE59jEA$9$l2K5cXT;(zsagekGh)tQX`w_
zUI$LidcP)zNk;DM#ZBK<c4DHj(#=urBE4ZFw1tSG4B-2!QxftnMeuW6ZqMV_e8C@(
zt1zcj8@*LM^Xqwf2)i}F?#qy136pDT+%(-_yz*riZ7;&8D92sEc4n(LmKZhDe^rS`
zMT*^8)53+iw`#%rVcz7zJ&*epkJY)QW1FYxzU0@W=iNJeskb|2r$Q19$wJkdyqnXM
zA1fpvbuSPZxu&y9x;o!^vJCYq>93nGJz3p$fM0qg8Sd%oewc;GQv<hkSDPogOOl&5
z0=Iiiedj3Yqh@-6kABqVNt*Qv1yI_TI-sZ|(;BI>P;Pmv{gc8>c6d<1F%7JUN*bdl
zokgOwUx@UpTKT#rj(IL#6NgRmn~L*Zw2-o5&uVVc-o0vvX<;=k8Y;Pr8tt4mB@mJh
zZp53F%rl4yyl=I0;v;VPf&*WpWZNmMOSwFBc)6Z#X0-AGB~GV%`zqD_3t@!JtnJe&
ziX{ydjio9Y_NtU_ij7=4+q+1i#Vgj^amrLy4>>z$L>dq{{5kx~Yek7kwRc<U!gLi3
z5G{7V@q&8crTtV6=Tq=9Ua^?^R!IiwbGU61>uX5Ql=|yx{v@FZ(^ld{1-mtX^wW~B
zWwMcmb<cfN@l)lu>&>OqsyZ6FDhhn5%szFy+9M^Cnb*9j)s@_PavAK3_4!D(UMJ^6
z9nYv`%558)4U3^Fy}0N8Mi3_Wy51E^DK(MMq8^D;U7-|J*HSdaDc-y*)RqtG<mWxq
zU-g<myQt2K<F_>LQpI}S6l;jnnQD%u%E~)<31Lj-Xk5gUc|!3A?^EN7eXQuCs0OJ`
z-KcP!ldaGbC$z;232d5>&cJQaRy2UxD|if*C{1!79_qd#!0sq>AqoZ3RNI9SqT7yk
zxB>!aRERi=--L#=Sk#83J;?LcIUIxo=1ee--5E;ti5gi)9d3t*#iFh5d?AvRhkJ!i
zt>IKqi)EFWK)4xMC;8ytu4<ztb7W)^0+W90LOXtOS0=&tP{S!o0__>m7KY=B-cjf1
z3DO0(j}h^v9JZ4<?ExbTpDrrTN?yXvJjswwPcC7RN>8_C@BZ*~0avF9Wsu94-dYsr
zQ9rz`k_j@x@38-ZucH1`bblCUyCqY(A!|pV=-nkEEy0x%9=r9<cr9Xnis-92!-=3A
z@Oy9bc9%pjj-&&Ji2ho{akdAaU45%CL5|Ex1h?5-Z(~6ENYD2&A*@7_Kb720u7h0=
z-dBos7-15XBGyD_N`;G{9Gak4*<Zz&u*l>(9E#|A@{VemI(;fMPtidFrIgAJ+|&}!
zH(AQ4T@Ow(ottjzsC+@aL)$%JBP1)%nJhT{Q$p;(ivB!r|FPYMKm|~@hhcD`*@)XB
zXy`GLj!o`H2<hmnJoBe>#PEt>_|<FDF#)%RZBn1>Bu;>t&tH|ySKx}X^7BEd$3$Jp
z_u8b<OXH}<wOgqvZyBGEIHRFnQJ2uGtO^-r@53J-H4=|{-l#j3GRj9Nfh6YQ--?2z
zL<H2tDH`M6*<QlMgMpC>EFMGFQSW-<CLlxyDJqY+NI;}|8znkU*|8hJp<JNd9wt60
z7LasWFWwQVsavUA+Q6w*;|`8mZ-@%u#yX3~*U&35r`hFAbyE*jN%C=0{#*DrY`(#}
zDCgYg6Q)z|^IGFBZ#bnS+pv%++o-F<lj0$>R+Vpwr388os^UN#r-5|wu+yPp@rm2s
zYi^z#6kEvoFJej~L(g>Qo6AYa_1>8SluVGMva5M=EmDpy=^s+wTjUs&y>mXM`g^A+
z8duKH(^Ofm`TI8YD?PdVa7tq}#^KRQf@$JAy0WO@2vELz;*=<IeaW=32pqy%Y4YSg
z5$t~>?Te~vqlpg5W;^|l;CHIwRK%i<e4ctQvbAAU$)2{f;hs}Hm@elxQt^+HJw_PL
zDVQHuFiqa;j22Qj`CQ=B)m!yE_2+9OvjW@N#N)`|`Lc(^;3$!W_K`RiwP=wvn_`H-
zRX}@rf)VM+E;SzcG`*7@K~g7gTgr=XHF8o%g~)b3#KjCfj@`tr_~1_z$tfCDNtO>^
z#Gs`D_;_iQgAi~pTYa2VpnioL1H8*$PA+oK<~_g^8FGK+D%qZ1<(oS_l4!sNZ;dsW
z<X)>alcF)*N(g$D5g~%LM~fSJ#<q1LK($<=Iuk#oWRIH}Mez?QT=zzK0cw*yIJUtO
zG1%TsBk<Zac{&D>ua--BXaeskWrGODTrx@kMnffBt@-#RuGF<3;%T2*l4#gl<mb%1
zfvKCuJ?&bKG1yp3t~q5IE270DBUc&~0KFKNhSTg-S2Dr70xz7^mB`xxpI7t2(uV<H
z2i_>1?RCHGGKu~xhf-xz4U&|!6~0hlqIx`J`%t#{d7RJ~zkMoE?1>Y5JfrNF#Ky#D
zRNcj(cakSilF*CVazS!==kqF|6Xp7zKG)#$thgcxdS3bb;uQG(X4^hD75uGwzoxD_
z;4|L~6GrrdvNw;u8{l~4Ty3$l8PUX;cpC)(BrxrCEd%ePc1=p+%*cK!8{wyLV+@t`
zVtI(uRPPxb=bb$gz`aqRB2j7v8LKzknAs)@rLF}|!NaIzYKmEk&l2)*q$0+m)Wv5c
zLa6$D2$!5uA{-pe+lwabmemib{DcPY_iDSjIaSr>hS?HVmRlm|2aJ<pdNy*;cD6@N
z;Ew04VQ+{Ff;vm?K2VdkXi<Mir&XN~mXDV3IhWxL)f7L}qocRF+wUBcxSyON*-%xw
zj}eYWIcf&9V@Rw8Gifhsws;01^43V=OS+mzHrWd8q;8_Jn!8OElL&P{w%XC%f}%P!
z`TiW4<pC8YZbZc!aIz1b4w8OP@}}9S>;loIYuwx&a%4pA-wum5@aQ$_88kv~AABih
zN%kV+RdKP=Q@(39Qr2p87X?S2F*Z%xFiu6{Bq8KmweTAPQ}ik7q9Eoc`8t<R2a2nn
z#2Xo^?kH)*3nZOBA_<fuz9ZTMe0k#KRyFAerUbuJEuXIH$<*h0%4_w|eYk;7;^X?b
zq#xc0xy2So6<3c!K`eXbKPf)abatkirX@b-jQmhC?Sr+DhS``WTTRtAci^i}pAR9C
zD>`(M4{4J=E}jtV7dpiuykGesV%p_CbST0S24j2guw;NwG7==!wHkS*z^AH{l~=vk
zAq%VdnaIO=?cIW-5d=icx|RpaR7{iPG#U<hV&_al&mlNy<aFrDGaejs*Y*35BAX9x
z^7!(SLwx9y^P%S#7GP;>HOh+7!@Jb7JRF3o-ASfn3iTd0t<-4Z?k&I7X-YkPX2dof
zGR}4myWG7HIK}lm@!*~rWh20;#OvN=BBb1xafm${n$A!Lf5d*$Wz*)iTmw37J+oh4
zUXoV1c6vitP)d7QZqC7n+#bBh2zHf5_bk2fbhW)&s>dE63DOZ(JG>W_;gZmA^uj84
zq(-ZBI*o`RiAVV6<6~*w+go<}nBZ5TdeS+G_v?WZ2P>m9T{Y}YbJ%8j+=|RSnuG4d
z&#KXYs`JkYoOHV3E@zW>i*G;h+tjcMPb!rb5+C)dVU^NT39v|fH;t=#j*-(G#PPcO
zj<}VZ!_IxxP>K%{DG)RnD`e`p*}&xf`QTFlh>|vDiL4aj8EUK9z5QgfI`A%2S}S<<
z+j{7mFtDO%!43}$=$PtawH4`lXD;Ot`}9yG5~Gp|e4<7$;zr~aPFT?NCRyIh+&39`
zuNw7w>b5HbQn`=XQmLrmj%r<4rtIC9SA#e97p7wMD?Ms^rEK#NZGO~=Tit7P{j3s3
zDsW={dN|m7edg0>S`vk%s2N=ozgum%h~c&~wR7B#$26+J_a0^Kz$tPSjc=%7o#UE@
zq4{hN1uaS6w2Bd7yEM<MIUmY&gsHoSCrT}?_ln?OY6|q64RvMfi5h1sJinJBjU$_O
z+1($VJi+NG;{Gzc2Cj9|h1Im2G%TCM@yv1=N4Qaj<YZ?-@VmV`7p{vFH4!dJa7sLt
z^R&(IOMuU!=3*FJdsYLw#6tBGUA}B<o}15`Tsxw+Y=6{X@VQF_5f9s&_n~w8iavR!
z{(S91B!}t@b5Kk==SgFP5q?QJ$vn@mu6a_rhk!qS9&d#77M0^vIDNoFoylchHpUiw
zgc4KA+Ea%CN}vZ^)u+Zyjwk8|0K-+iTqLdCU2O*6E1KJm+oo#+R~<Qzd#E^8HHkNG
zk&D-znos#Oev2D6dXf&3I2|BH<bZp$a^Q4+mC%e&!DVb}TMxyLEVXHhd-ig)n|g>5
zPqTVt)PaxNqi_RRiw_y{`A4Lrk+hO@uuftDhu2c-EEkd)XOk})n&iuFlZB(eM3<*9
z@vcoFrH`tX4{pNh@cK1&k3}gE+Gtm`@R7!78qvv5LR|0l=SM_wfO5Gaugee}p>^PQ
zdIN*%>GwHHB2TUt$FnOk3LJua1-j_=oYxrDV2<}3Q*ZmI-V1Kq#hz~!M|e~kpz|~J
zm!mb$En$bIHM4Cf+1Clb%RC`)lD47>&PpQzxZT-I69d5A(eoC!v=JIUeZ<W;JSvN>
z9$!;s*h-v+&fTRNipxtf*5`RT;CcuHc~&D|i$VYwiyY}*=k|P_)@7pO`rkAEiE`?Z
z7Bo?qNoStDqUi1<&Lv<10>V8&Jxmwz(KYGV9L2lo7~c}#!^wLjtIL3we!tyci~3yI
zQURRyb#7iK@+*!rAzPiS6Z?XnYQ}sMr((bJwAI5(Z7^-DqOQfAx5dRrrsMJsWva;+
zBGx5*x~LP`RXkf%9RSDn<<-8)uZT^Bz|y}uqumyp869uUC_F(ve?D$kxTd25Q#%S+
zf*++KT=h}wC1*K2C|!xU`DLj*zSM{ePZ6PbImhX}BM@-imkQ?0PloVuYc9)@I`t71
zt!6&Gy(l<-hLs(suZpOt8U@Del^wY+nTaZTthkX@B3=dD-o1nouw}hx@Y{NnfG~C*
z-bmir=CIw<Cn@B?>Fi(}p`o7LsK|aPFFw8dStG$U)G%<m1w@pE`bY(C6ZbA>WHO)W
zw#|zYR8R9*N0VQ3T*E1M9(A%unqaP^Xi5m#oJd-j@m_d^tGat_m5*pMpPx9?oF!0l
z?SZ@8T9RL>u9w4-LN}uhKgkIv@Kq~K#Z22E+Wp>KhD9>ycJ;~r)N&aJL6|-)m1NuV
zyxrGKzc%$Ri)(|(<V}buwpp7`ipZ13<Z1?{_{=+=<;R*BH6yn1zmm&oSDTuI-K|D^
zE$Pq?Zn}XjD@xy`b}^Xfaq!p@2X!LZWCB%=K)=y>5}c0pp&R2SNc{g}>ddk%$8i|C
zl~{siiTmGJ@`9vaesrHImP3~PWQhX^c--4=xPYjS>>39N#T|%jnjL%H*H@G7%L^Hm
z=fQsIvyL?8uJoWmgWaS-rp(dEdM);?o&mD3XtK2$2epe~NfL0A_`@i)Pic@*5Lo5G
z(w=C#t+JRcT``JMClt{j_!^l!#ZKD96zlqPoTwjR_oJ{y*Xe_EyN36eQbKGh9ikFh
zc0c2@5Yor4ZJN}dN&Jgii7ytb>se_nwgCSXvbBPM8{(44C;%a^VQ4nway~wf%3i%R
z_kLg2e9gXBY4i;wO5u~7#_scNk$U5&O26XL3$BFlmMi$u+S>?WwIM{stgHJFy@6;F
zX3wvLhLWu()h5}JQN+Z|r~IJ6Vk7gW*gdG-;ry&m)6#;Xl{DkHDlq0bd&S_N2xxk9
zZ(To+oKuyGZ49gVKmpD5T3VxiDKaMgCf1oce{@#5T^QW{7jZ;Sa(nCpVe$wPOd`l>
zr?nFIN@=^KF~I)1<^w!$&6I%T_@i~OYqLXI@z!&}!_k6G(BrCXeZo%r-O%;ERa7uQ
z$K{pJOCD&w?A$UqFX_7T-AHZW`+IMfUaXKxq_jHgxd4%;!;Wl#)%7H8^1i<A6Upj!
zs)s4m=G%Xb!;b;*P|P_*06+V?A;<=(^;I^5Jm|E%@8LY|WnXP)mBcr-`Dz@1!|i3=
z?bf~qG;m5tO;kn6PlO!gouQW|#`7e9b32e}F4yr=%<QmX%^A9D<b_XSo$ZAT){NIr
zX>6takZg|_ZPU|I<HLHU>IPk9bAOKu(aMZIb-+u#9@kG(37Xx=f8-elbWdF5L5Dck
zsQW%}?cu+{$9Zf12FqH~j3dxq_XXL#W7}+G6FxgsDJ$L2c5-h3z-l3|hALNIKf9T9
z&5rc;@->(U@n*!5q?B`YLaMH=OF@2Ds;tlp+4fHM4TQVKXWvh<W_TE2D{7cI#f%4G
zgu6tu?h1~51^}b~Z>L1^A|^F4m#2B{6mToSJB+vMmH_F9u3p+VfcfL!5AA>am$V;p
z_z4}|h2t>w=dQGqdCFi<O=vB1^}Ziv4?lKDXR>1yJYm@(1pdYuwap@bu!7-PeE{Ru
z__l?`A`Ut1?mVn&GDLypWQ{-`VE1<}@iGf#8vTN@!%r30BbBN+O$qnF-zriZz?ahw
z&b$Xw!RLQZ-PaFL#Lm#E-_d##cEll+d#8x)0bI>zPoeMhJ);C^Wk(}s216Yjub&<E
zotMmY*&(ZwZ7CHhFUXIz?|G3p?z4~?_Z#QZC=qejl6#l-2irE@8VcM!A`#iqrU`%N
zcyO*IE4%-PGv%ky&vEN|$Vg`}@{zl_bv@F`Cva2Goj7@+>?$6K66NU<ix^*Ka;gDp
zJm=k2X5X58bpJi@?&DO%-jPBLFnDJ}`RpjDS#*v;EM4RHw3!W0OWo=0vJbVvDY~^8
zid(Xiafoeaf>WDmM<?V(m6Wq})b$hPp1_(uXh!S4XnxHY|AAbpx8Px;^vhl}#kQIc
z&@Gb-jq0G)@cW|c9!?)<(VKP%=T>0Ez$ewq^c{kBPtHVt!2P|Q>~A3pO#65}6o8kd
zP#D$CQ*%U^j8WS9wew)NbtPVQqka6Fq+LN<RBde#p~*gk$ErNLJ1y^~ohYK8oMkqj
z%08FW&cG?Z6+Nu7zsvaUltx?mOglX@j5C%v>F-WE`rE2Qz&kMGh?AWpPY=#JkdwXa
zaQ~n9qWxrTH4c8WQuD@9vqy~s<|8>WnlKh=M*&5J0xDuJU?kYD`>LmU%%i(YUz2Yi
zg5u)VHXAh^ce*5BCN8EGim$RyX$S;yES&kWL#jR`kH&`!3HvSkM0Pz;=#;5O**Q1E
ziH6M6Y8+s1Cvh04gsrl#qx%wJUKA$FPN<g@7O<+<IDWWW%C8MbF?QLZ_nL_W8fVn?
z6z&sd4c~XSw6Dp2qg9)XoU)$1Qyy)vWCO##0sC4o&+HZjZ`mOaOL>zw>upls9dTFS
zIlEhp<9nY}+^8M;Nxx6p?-t>V)Z?M#?PZ6AY{>&y?d*EkcY<7&&KC<!x#U_@h6Gk(
zzu@~#OesVm&|x!qWd>-wqfeGKRFm|bdPOg?0bP%|fR>YNmzb7Yq*-L%V%CVHk?38*
z?kh~ABZ_0eBq4L;Ef*=w`SQG7pa?XIk~qHK)#$#FhPW7xodwz+I|PSfdeFRfz5)o_
zm&DFMLROQa>{l8S>f!K({;cfu=cmXTy{PrDzBPr|ndpL6t9e48u%crPF|W1LzT329
zG4>}3o+sJa&7&{f-^af78rWEq#TYKmQRAF5<GwU9b$exp&E^JFV$w2O<6QrhM~cv)
zfL!m#_fm<bO;MtR&>wa<xAZIGIz$ho%oLh~m1|Qcs^QV*VR$vZJ-amwuGq~dS{
z**A()f#?|DagNVrn$8LB-FwpByNDCD?tKq`uk1_mP-=ZebE%iz_q`Mn?0nZXKH5U$
zX=f80?tJ!pq{%jgQz>@=+@N%9Vz*o8*7y^$zF-gZsadP**W51|G8&|(pZ%aw5rjP_
z`ygbeIXSL}P4S!0CJXP^k3i`3&%RXn2yRFVo;}&mlSbSC!m=!P-;HFv<~ixJwAzov
zEh}xhNiweYUE{;=xuP=aXH9;)8?=n1-6koTQC|hnMA6v^qA_|1nP5joOV_hw>^^In
zG&8PXbLa4ApB?4CrP}<3Wwu@qdsd?5J%+nh_w_{^_Z1t~@3D+WjFwxJTJlT>n+P#<
zFlw|Ltn;!jB^;9~Hn5Mf!`c$q3Nq7W_APhaV*Z-#t+J0N#k$F{xyGvD;RVm>1HaEz
z4({WIf*f2T>+a94Ed%Mj*XPfAx)kL26i>7E+20itqn5&Kb<;^fW^3tyJjhQD77$X~
z`U(B@eXnMLXx@4zSOULiTBR$8TtfcfG<qv6oobDr`S}BcxN%SUDEHM|;1|6?^Cq#&
znv1_{>A^{)x}R@&G7vj<FI0s_O~!YZo~#|}v8!<|RF+5V{r<bYRj&s58?K*Jv)^d-
z1KLm*Vju7Y&Kd{q`V)MVB8!;m9Dy1z>Ea<#cr$AvJ6v{&1Efq**Y`dp6vt=2+D*5B
zsN5t(Hv}U6U5`?-t|$C|Cr=^JibJ$ZQWfmw<FZF^2<SZnwwz`68IK&tY=xcxcJz?B
zzEjb%ieCUGj=HY|xxVYSs6JW9ZSDTP>qBmf`70As&$au|d?t*Zp#UuN+3~ofU-8@0
zewHxSArE)QEIUXTQt?ym)cw|Dp%%vVlZc`1Q9swc?@31wj_+$VOJM7_Skaz+!&58K
zFaX1Bf5S5f!;rg!?Ea-4<%gTpusH6s?08~Rs|{As+|{UdKizkz43!?`(;IBjA9;|I
z>t#1J!n0SCW6zgKK66Rxu0u}wX1Xv>dHfce<0k2M((dP|x_SqB?NsQN-Dfjm^YXu`
z4d4lt#5nwjWF+doFwQZ~#EJLgca;6CQ2dkfmgV9Ut*q|bkGg7VI_@5Yu-{XOX4P%|
zp){R-<RzSX4SOuzcF9@Vt)}D00w-1%``xI~(WBk533jTL^^&Pi^6DWB*sawZ;Sp%b
zx#n+R*Smp`f?>K2n)^Kyz=SydeV3Y}%rBH+dBZ2=FJ*=scZZYZ&RwrZMfW0jr+lrx
zSvRFQlTp|-Q-)6Fs7RxX{VI_Hx>1sGqPnC%VyE3-E?mm24&sp9zv0q90vT%ye52Pm
z?l1}2i)JM|T8-mxJ;vvgoxl6K7bfvRZOJfYmtAE1@eLYvyR<V3^BF{z-O@t=t~ewz
z=&O~|PHw%l`NFLm@6ulX?z{Gh`8-KLz0dx{7s#sgo$lSbFL4ei6FfmbBQjn!HF{$o
z^pD(%GbMI?XHrJa-gDJQ_8Ke<1rLvySB-xW;*N2iiAFT-C}KwgBZ@kw#v!@SJcqbi
zE4^S+5aYgsZI1$J+Knc}XoaG(GHRai2<pDnnp*#!%aG@N+VcIkiT3Eonmvgg4$;m;
z7J(${Qd<|V-`2>-#UAn#A#;YDOxgkJkcb}*G_}R@>7qW+@bpcN5!HFwx0p|Qr@Gss
zLWh0QMfLRB$XHp8<6)uX`$S&SgTX*{Xfu}f-T?#1OFKLGPUd_`e54sTMtD-dyUKhd
zhZip)7_-8T4&e-c)!mEC;b2FMD(>2Y$O0=wBQ3ZZpiDoXeJR<U{LdLaci2{hI`0wz
z3CyL;SM1N^v5p1ojKo`~cV-7pmmZc^_9;!g0VT+*n?hVBuISMv=#WnYEvQ*~dz3a~
zQ^yXUTA*6}*x%tvbwK_mDZZ|F@7Cy4sCqhHzR<&=aXS>edL|a41%apRv?m>XZL>mr
zQs8ix=B6`J+%;3<8omSQIU!RaJNdGsR|c>zvdQeh;^)(Jpv-^uw2<)5*epm-&9X!5
z($kNkUu>q`cnKrkjxU6~FSrVQk!<&Go#?5M9m3L~&BZsh9v^uJP1mIWIC)b&Ad#|@
zY0S-~pk&wwMd$RpUJD{~+38@5aCrJDk?U0ehnSD56zKqU0C-#T(FeThxhTnZiR%rd
zWX`B*PUwd?BSBD}+iRLUBFX$YM4mFrY--fin(_~z*IjY&$hEW7XYa4BhdP#F2FdW2
z!El#agKj(E5?iYt_P7_wBqzYymN7N<D~KV(xc@c3>p?N}h*jBYznA9TiIf2eE~-UY
zb{asJrU^{r*7fLpbA89<86hAiJ1jpd-LZat*WV_#Xx<v^o;ALiRN(j$jT&V<?$mzx
zBFOa9Ka=D=@=<yHxD~8or_<bI?<9-rdfyu~{#3i+vbx@vtj6!sxEZ-+0=1hG!jaRy
zqj^_$6T4)W3`5XiVTL^e9DG-jFF@L|hk85C>bwD*!@Pz_3-qPKX<oD30X)-&)uV+;
zm(wVp!6)=At1f3>*m{}$*y%N!(-eTO9SlWfox)GW;oO1&Q+673x0V80#xu=SDoEMm
zw2}^}$rwF`5vFyby|^BY^GCOhJZB`4M%n>oYaD(Y2@!;SfPBAyA&I+!i~#!_>ZI>Y
zeeTQZdUqKm0XU?x6IEK-fyc+$4+?0dS=!-wIb=^(cJSX<XQv+=Z+-UDr5H%m%dc?D
z4q`c@F7VA|-++N*_wrrAc1;JeDveLRdF@^E*?>{5-tq=j`qEBYKjU=CU7qj02lQhm
zlP(=^X-CI|l=>h<N+-Ef64EQ=?cdkxUApfi!9b!A_&jS6112-IpnQv3xwYfWDFW+F
zn2MVZyA>#|U_Vo)V3+-<#fR|+kfc}kcO~36Ni$av*wuk6;XeH8z>cQfo!z9UMu#!K
zHk`Smb`-wO*7r9b4+kYANTRu=h&J|(#zTZTNBIA;`v%Q@yRapD)qEhDm+;_Ku65P;
zh!B(DW|O^b^t%3#%ikK`Kf~x$#7=9%h>RKyDD`@vnXSu->1#T#+lhpL(YRf2A4pil
z@#ADN9P3qIS(k88kd2Pcej>$&4nxvA`4v(X8xs1Hy`@Ku9k;Sal7q5FUC&Fx-qo)#
z@`-@34N2>Dlh>NQ3(nRPJb@uc3mhf(iBDRN*J<g~2!EveveT`TsekNvf#EX2teU$8
zeUx3ogS@46-uGMp0vM;MUt#YW-yPnVr%+G#*7I(T=aBrOm|!&y%J7-P1~c%@A8t5u
z=T!CK<p1m@?i$877_fHQVcxp2y|^D*=g%RyRJDhE+&lKW{ys?wcInrZ^Cd@Gnt`Mw
z-p<kmc0QG%&;P;F_g(LAWyIP!^JRb^NdV19e2jB^@86Cxv$4Gd=fBLja6`9!9Zf&k
z$(LWf9@gC=-Bb3pB7BxsknPWY>GaYu52Mx3ZdX!*r2E#s`I_k%28$5Whf2Mt-4~vu
z1iRA2BMMO1XH#l{#LsL2s*!!)m=BesSIYqe-PC+|4RiJ~6D4bS?ejam5%PNTWmbS(
zeD04(6Wo{%UL=Qd+5IhKremP_@Llgq_(l?|!-P=x-FU>Y?`$+0>b^dvqzwmN?M&zb
zo0BhiPHy&JsL9LDTPC;7e&k!<uK7UWt?T*KlM@3wQc4!%QwN;&`e{NMk*z`+C&MEs
z!jA6SC9Uwj_XeoY!n=bKxZb`kN%=B7hw1$;=h;oG!B@NG_2A^hW6WbwZO>H#$<dJ>
zAe+W`FI^A5gprAC$h5y<P$Jea%9@l>Aqrsrdrl2JAgGKHz}Dnpi>u4+h`3?rwX{xn
z<#uEb#>~WOM`B>4nXaqcrjZO7QVr|@l=&`)6g~6Wvq`@y?Nj|&cnWwh_q+bEhT<)r
zRh|fVb0Gs*d_G?5b^TdgGJ`DhE?hv`De3npQ=#lz^8~TLe7c$#Ppj)MqMtZqyx!~z
z8Ty5L+;>X5_M1PvwmYI>z{#WL^9bvXjt&rRuf|~kjW~kNUiSScHkafOsmZ)q;jO7g
zM;;tk9xazub~0>3d%~?2VA+rK>XaXMm%zU3eWxK0u1cbv-17M%m-G-PM-yLU+R>4|
zvGtm0T+>d6-N^E!@5$ddi#~>#<<hyn`_eZ}=9!A>T)`%Voh?e4ZAcHt_@3)XdMWq7
z)PIqET<I;cJ@2zeJujL%H$1)`g3fup{}vj*lZ`d2ap-o`IP~nVG3t7cpQMx6)F^y%
z`Gdlnnh$%a2XM3Q`+S3Lk>H2j@_R1#=JkH$=<nQ%49t_{5S44&uSNl~ebXgttz1fE
z31(MD8ifDN>XQ;+XJ!H(*2r)Xhgc6sswARPJ?eV@7Nwpj?Ba~3AK9nHICAZRn@V=~
zPRNjJ{1#Y5*-^lX`zC2g*n!C22;tCq4z~>ZrEN{Cr~7=`cfE-nwuv_Kwz}Lm9_cD}
z6;v`3#J&}_Qw!@e-7m}+_*7Iy+)rvXOS&_smN?bJD1Km`L+XrdC1`2X%kj@rlOzhi
zBN#WZA9+4nX$l<6><f8k1@R>jm+ZkvtHRmKQzKT3COHUocET}frt*hk7;1q2R!`ma
z(jp&0@;LdO4^W6*v>?PKAd+hs@!j^A%{;&1jSixtm%u;eo2Oy<N#c=|hJO_Ki-Qvs
zg$nyAsoVG2Arn4Yu)er6U!{~^?|FSci7LW;gS`Z={m#T_u_v}ACC335%ncl`?6a%4
z6krBC5jD*T*~UmuIpuq>`z9r&Nbs5@U27bOd}joRcO}>Wv$dNk??talO>?hRw{1u+
z)Kk%qJ1ez4-#2}@ulLQY*9a9;>>#!?H4pd6_m-g`n~^1hm|@g>@PKw^)lL=8<OnoL
zMRuVR?l&KVs$Y0;G|=@L-(wj`MFP8_I8)DMYl_7#(hHXNg^g*(_>jb8cZ=+lqMl9(
z1T6A92XQ#1fEgIp+IK(}in>i9B2<Sf?SoAHjQ30j=x3j&?z;e}^s=K-va~uI3g&e^
z3Qn^!M=2<Z9fDx;-W1&~+i%&2>H47pb@XVC_`W4MDNy3cDl^Fm*rY60EtDN~U;3nH
z<djhEh>{g{6!VteGaAmP-Y08p#vVYu-|G4&W^5j7Ay`{x(2()1uwGuhecJ+@L^=|}
z;<7(6>j4fzdmt_cpC}JT;rUAx1!jM|8Ji|Jf&8MU1R2T03%onh?Uir5q(!N6^L{x=
z`e~o?`2-+T$*k;rmPhXrPMTeIxUE%10Cv~ww*rQ^R9%p#zFxj#N{2VbQPMGZBZqdy
zB@IbFGx&#R+1;Cb;X(TPKi@dWaP{&aX9k<t>6nWY+Q%lZPR5nh4%zQF4q{#}JevnY
zTa5!pT#p==g0h+?+}vD0)Y2OQC9=~489{DRa)zBQ5A00}v5Ar$l6R6UB;iFLp`eD{
zUyR1-0`R`v+LmgE%5BhR8tqh={gsGSl7hiI2?@aNTj~tQsoSoDf^?ushOudVjfmZ`
zBYcwQ7OXvG_fcAptpr7Y{@ML#%X1E`^-+%n7r8S1v=_A6G8UvFnkQ$Ee?hGtm+x5f
zS)%aSWv6=m!X>%n(?2^crBTc0N4FZE81%)b4k_8H=SAylMt1FXbuxt=aa>(5@W;&0
z1^nOEI9>gmHM&2#R3wK`Y4^LUT{XPLH<|((gXF5mG$$OVrG;1a+2gL_Ols+qpvWN6
zNgVT=kV(#--X$X0#l@LFAu<hu>~psIhZSiG%!k&rzvzpt7fdtMlRhKKk(BS0c||Eg
zf?Owe5n6Cx&&q{=MXSbkpI4HQ{CaC83~qq>Q{jT@^f5bi&rQcx&)Vv1R>Qz~)niXE
zZaUL`I-T`#{P$ana)Zm3rUMy`Rxo~$6NuEuw@f1P)%6dR_ejJcv@pu<-fh?T962dy
z4J_Wh5vewk6NW@-ryZ^k?1J4|(8*&TkZ_DYx_vm^Y8Ys7Rs>)4;-wda*n>&@k^Ti*
z@2%`p$vxF?`m<5v_`;L8q+0n-BHpQ3`k|>bxkS~<tv;&Su5eTCmny)H7Rrmf8a4Wv
zb_xVNf)Pq4dAt#83^890)mXFmth9%)L|x4!Yqq+te^L^KR3_zQ#sHYaREIu4S4Zcx
z>~9T?!!P<${886G6}9BP?35qV^&$SP&jGJ_l-;&OA&D39v+B8Qpf%6WqE72DgOL-Q
zKt-9EG*_~ZCyhsb-97482O~Em!l-W|DRGBfiXG$7BDZ(0ayaL^G#-h9=wYn;4w!gN
zCy;;RBb{8Za9iWkc4LXa9ncr88U}FfH#|@dzdX9)g0etn--K6efDdw(x+~0$;hNs@
zxmdKTiZ_-o2zR0e=fJo4ptr#-kah?zR_+#-obj^LkiX+p8VO=2haz3{9-e*q7Qo#^
z!b`Bd601km06O!n&GMt;k^QwyEl>-aWu|e@0{PUgA9?kl{^3dT5Om?N)9a?7byn2i
zj&wm}2i4YgX2T68S4(@3q@Ie*)U#gNVaTeiJ&g<_?NDm@_3B1k6|4d-!;}6q{k|bX
zC$mRnE530;g;{q0NhOdNaFqH6)z&D|3BA4ZMh)-ZQ5ylEByo}7h#pY_kF;K#>=(sO
zgAVrv(BmiKs_X9@b>Bh$>9cRseQ{6D$o4)Z(#Oeh{IT1pw*ZJ+<Dj^_YvDZATP)7~
zi<WTBpXkHC3_40@Lgog&pj7x3a3Jq->S_aH`7|3e`d&Si`S&uS1~*`Zi6y00Q>eo9
z<=>VshePzmWT>va#10KXtR)|*cD`k|t_LNl+XChb0NVgywxzyER-~PlhdA(`9;2gJ
zo@D~kuSmGe47KfibDX3BOmC7#)7&NuIts}k4pkBoiz);IeT}(&oXh}FK(D_qW2c2T
z^4I;0n#^U)6$#LH_o}<zV4kC=icst``|4Bj`bjnM)!>gd%*fC*N!OMA-mJa5uijtH
z-9(|v+7Rz}S!ui5@#~c+u<RkuLUEpKss@?pIOOA?JhI9tHd1}8U^nVoD5S<t2*!@K
zY3$Ok;rGPlZZ(Jp1T?C20rgL!Q%Gq}gjTW=(i7^ZROggvqT=rL-e>nPaYU^0Zqm99
zq1sKFi~<z*ge2l&f`tsfggy9ub})VS4>k=y;^Vhd#*}KI-XpYPs~+kBr--xx%ul_{
z3=-usqwA+;<7cpRdGT%#iHYIy)I_tYT(PgF*jT6h?-)nQ20@8|owF@-&b}Z+vU>z7
zQtE|j%C(>bk{trM$O8C=99%xwXVgITt8j4D^{`L12tC02wz_^TX&-58TBn{1qIPM$
zaXr1Ax8_e+m$_mTiF!Tnp%n7+*LwZb{HbDRE*I{>RpYbl7>CG@R<9aI)%?s3yc8&w
z-9*o%t@~JZ3`5RHTGjp_0PWe!j#)6RtY;nRC#HSe7(N(j-*-LPAAPZK*+if+#pWyR
zMbPN<@XoT6#8}!UE^hTGJIM#D10eUrNupzEI-K||Q4g3!H=?Kk_~&=^xo8uLM}z<-
z%@oLDUtOcp+3ex1?eOC$N9L2#P<o}BfZZeRR)a$z)*=H73E}ss8<%JEnZW${>T*gy
z5#fT#$>k(mzr+OrZW%f`W#_PnsyGpLtI_Wu&Wn8&&H7jN$owgB0y|1Ql<)bJnBm+4
z1Xr_plXL@<q&Fj+7|b+_9L!6NoMrZp_QlRZL;kEuqFKMwSy^r5q5<k^x}HqmNxf+7
zP}Dp{9w?**#sahdV*Y@pD#gg(n&*Q#4WEQ^_4v$iCtbY(z_PgpAOL&PMFgEm^F}Yb
zN!O#RcXyqfJYo$3V!(dkKE<rL`$;5~M!rkZf^}QpxJ9$)`~Mp@>#puFtV~|EH($PY
z^tOS8YIPWtO{eUBDz@-2z!udj@GVSYj5^aTglY#^ai80Zgu|k&zsv4Eksn?#&0Whp
zff^64G!Dt0LeksIPJENc)LK~p#BToi8Sj8rlU1XWwVe?WKjkyZzLr|ahtLoS6~qop
znQ3@*XCp+U5!El!IQ%3ti{aSOn~U9G9?Len34&`f9eq+Qxn*UaE#@O7(e*tRDXp#P
z!(n^AVNeHKf8v1I*}gKtZ;3fut~g|0uj~f7UmW<>Gv`GMS{Jdba2h?Ro@Bm?&hh`(
z$u4`$Mn!?uikEe2>L4(qLgBCdeJ?(g()cQ;-S2sP4QM6`e+o&8eFV-<l&cA+1=sA`
zv?i`zVcVa5Di6U*?c&YD%sz5P<{;@$%B4zjU4lVu<p{ukW{*lv=d1MFXOGvUmD6Gr
zy!1lfVc*DkvDX5y%;R;70unPG&&v1@bNIi|y1h!dOsTq_6No~5kWqvC!{IH#)aN@v
zeg%NJ_EMISP`%d~r}$htl&Y&HGHKGP(rG<`N_^Zp=|RuEmAOsWL-HivYLOG*s!I-I
z)+C<?HZSp6qgktGvZdoXN&c-SLkIJVJOgJ}?zNy^p41?<sC)fPI9G5^>9z#M>3-1#
z38@Ne#&mF(8TaM!-oZmeZE1#X{d@T{_0RysKB3K+kw|YXTg|}ht56o_{M@-t0b4cH
z8d(HDj9;4RzJBq+GuvI0*>g&*QDh3=AReao6Wf6M8*A1aA<KR875ZM8ro@cY@QUhS
z6X$E%j|8S*b3z}w?n$CRb7>*~zIE3JTqiRUV}v%h?3Df@$J$PR-{ro;^q6G(e|X~C
zDc8coz5QKoxP*}g*rKT|d+@iZQwJuEGH>-Ou{*4hP;kuee3fiE1!X3Zni7aq#plgj
z!0aq1Vi1~nZdM(pN*pm6vSX*5F;pB>;1(MPEt_S&!zz^VGdkc7pkRUd#%)Q-&Ch<N
z`z{y(JZUfOv#C+731B_ab)h$l#KoWZE!WyL%JdUP8z+noGt;TmB|1z)Q5vxql&G>F
z^%&}Uh&DdEjg=elf*!aYc0&Zv!|{JsM=Mm>TYD{JT-|29<Kr9F3P;M1?NwK+=P7o5
zNF7(bdf)4~E0juhYA=9f5F8uNdPjEUpvHN&1ocS!pQJFxPKSrv#cO8SbA=fazesPj
z-bOwtItxpRk<0~nT`&9YQjDYq>-$OWlAxuQ1TyK!c3<Vz^?j;~M(0QB@?oYIgFLXB
zEBZIO&q#Mwv#!brVMiKDc~M-+&xhbOt<_^6gHO+DHh!>47*fbpd*4&7-6Q~s)99}7
zsq2x`JhAV!wUcXLDwo0dB8f|yXqlfJkgg7cf>BP5S;Jc!*+fp4TfZ6}{+#uj4M-si
zJRs%NQNttR*&_@?0O{2%JsOeBV^edI^|G^5CEghR=vm|YNeEs^+-u|JPXz)hi2Esj
zLj|$Re#E}UmZ&C*G=uQvu2eLvuTz(!0cHwtHeCg2Rd;LT96^KRFC}%OuGKd|iI;VG
z^qG+5JCiD6D#+{#_9ZfIX@w*&FtdJOophC`8}9_<PtK#7HitwaS3wci!#_~RI{cB1
zS~Uo9egc8%;ms|TR=g)kMO@}NGwHX;NJ0YkC#u(%)FTj5_yb>sHZZ%y_Jf&bm8$$^
zI~e8-r_`VLzbZ`yKR{EjWJLc!Ps@Hpme@s<^m;rgc0VTk!xFXe$UO<%5)BXGLV$gW
zMT&ii9n*9gWs@T2yEHmw!Bt7&MWg`F2aAYKR;$Z>1CvOk2r4MLryX|Qm8p^bvK>Ua
zszP*1zO2xAOL7g;SfmsOy+%E&f8?l>!SOnKXwD2Zu=!RP{tWSv)iCc)NiVYdbO@N!
zj&te5AhpqZ=$iQkPVYbDvWyNK<uku?GcDP!&}(7$KV6xZ)Qx7@*`y=>f(?9qcHi^e
zM>45dCD@$46Otd{EclO|jAPjS5%0?E&`Pog8qraFQ-t9MKqeLz3IRv5j^b2UJ$yRs
z8&MTG3x3wcIgFY9mX(<~u?&aZ4<@NM;(3mqNEYliq$a5X>&{FbZIYe#0#=!SHY?m<
z%zoRUT>wSs`I&EEvcMnS>}Nt38xRY$@u6du#vMS}NGoKj1p6h&VUm1b-{bp_D^<s?
zlxCiW*tcLhCd6s7HZ9r&aZuIqzQd_l6rrBR09@d!G73r9eODcsv#}7K!wlfQ>8th!
z?>X-wUg|{7LLV0X)`CKR<>jyk_ZkK7RH;z9U0#-A)#CD`BiU8U%*r@+qp<%cM)8vf
zL0W3b-u9@0_4C~nLE;gOK)rqjl0!=(i9HZmxfJQgyX2lM1dqan#%x*Z5;sL{RNxt?
z07d=}zP4U=AM+}CG64|Q6TwT<GN^%AwuT|c<a5+_ku*qa0<Nx`f{<&UeZtW97i_v&
zg2fCKSOgKsd*n_68U8%?WhENrg0%a4B~N3~0+x?TgZbQ<x}>r_>#qJk%CX`N?b-FY
zL~Gd+{dyR^W;4gs=){akP==Let(S$2su^OxN~N2csWJ%w7r-C`6Dt5t!q2`{8deK!
ziXUL!Xg0t2<W9EvVD@ccMAgH8m-*dv!<iY-EBB;#uSr5_>DNmr%+}8g!X_?<Qf1aK
zH|pB*Sw~M?@Dj<0ged_Dbd+n;4p=Y8A+WwR4iem?O96=WbO;xHeipaYcb`NG*kP;A
zZK@Y}CiXM9t(cid5?-vO^74+*h%YrirlksE_=)5^^Jk@*Mu&wSYL40<P$QCVc?)*h
z+#krz4k=T%ppDNXb-@8CsY!(eBRP>4EAH)trc%)EHO^2AsXjBp)0*Uf`aK#Ke800%
zm$Uyx7Qa(l8D*ZkC9#FO!P$L|U1H`z0GxPr2se6k>=9z#+3KEvA)itwuov`t0h>H|
z*B9RdNT2rgBxHwgLR(oqDLLyI0T2y?-#F~EBy~YHKKouIy6P_?9oTkGO`Tlc4zB{y
zgGk4s2>ZzQaE@Q}K6;ux0Fn+QFR~!!OSv&l7rZe4hI<V{;%{fwah!THu-Z?_E{fx;
z&D_So)J&?&CYW7rom<el84wNNf4RNW#Agy?WMB4k<QwE1X-qbW^<-Jl?H6^q$!7q`
zM5GTfchelj*LAate92Ei0+N$y>`aQn0_6|#W)_@YA9YthMxwq*zw%1zIy&W>3g9A>
zA6W=kde=SCnFDd?nk1d9EudVu6#3q{7u)2Y1RVi}mw8I4Jqjl$k=WS%mn73mvbLC3
z;pSLkX;Idrr}~-fcf@SSL|XDtV0SmZCHlid!(P|>lV~=`nkFsw2>Pv<jRaKIH~hFy
za*>>r$h_RNBXBVlElqT6cbVUYn%9InZ#9=CQ8`JUWoHk3Wi-WxYeIKwuLnhi_$evT
zm3vI{0kI>G1#EAt8~WZ(@kJC^MFR&Jm4_`)?hrJZR#&_4I3rPkgy8p1;7IbZ5k`c*
z?=uI3yO@VguQ5nj_xO0<v6(KP1h?R=k`mcSP!^Z7V0&&;H!)&dzSivFnUEDb^;)3o
zPP{7QkP+0wmdAYM;23SSqs;e8hmw#>9$wo0Yt(FT(<-k4hhC?zOP~_I^}1RjC4uf}
z=n0>)!`*dNd3nF7Od5CsH|7UXLV^(e^9c5(R|T0`je?$LSNdXDCEOQ{cJTMa=`$+h
z3>OrA(Yh(xrg=P&Q<ydSOMK>1dPZZz<!0q3;WHkxm*T^=nalH%L*O~#(x0`4)R{H=
ztQcX9vilY<*+t>YXjLma!T=s8Q4=W70feEntC$H~jPX+FuQLr!rx)7NqwG^uC&+re
zvEOIrQw=X>(F=Ui-cbQiPMPxT`OGGrMWtxknfY-Myo`7x0e@TipnR+YO+>;T4w4!M
znuZf$4EfCO>LH{=4_4bGcf{_#$_zx?$!3d1+2n!0mD;OiE(0^%wHBMPB}@u4gy_*u
z`Di|_nZp;uDE!Khrk5G#JkV*YSV}Qd@sBJFQl~4;WXa+&?@r-onQ^}o;&u|ax0H6@
zF^GMj6U;-ZYD8uD=ygkv*k2}y{ZzDi`YV_m)e{=;YYZUXNBg0ll=pkcH&-I|{9H?f
z1Lmg;)Dbs!4Q#5vO2khxZ)x?7uLsPYBJ1~)nZY|E54$Jb$}-_4l8nLw%*^jB3!vK$
zmjDF!NV^+sWz>*DH!|<l1AXiVR@Tum(~q<%ryM8`XPO7xw&AuwUb*U~15Sp>2K7~8
z067Q5X9R12&$gz*va?J&@18uj*!}D2+HOIz<?}<MKQ;Ikhx4tu__i*MNj$Op*3Z6C
zSJBVb5?Hi|!VhDkP>vWu-4exw6&144%xVmp){=|^I9W~ncF1Y?;gC;SH9<T-(}uO&
z&2IJdG{_!Bd+_Gf9sT!ykq879f9j4<_4o@SQDa3Sg!yt7>+0QVd!zZ!Bo?XaSP?xh
z;8T>IG(p4-Yl0p~D@CXXXl8@LZlr;_6}0n(0Eg@-EMEQk)O~K6J5|28t1sZ3yW0o-
zhe~U>=h7mf2K%MQY(4kd+{WM{okMz%(0T2a^(gF6uPReT{7^I5G;GKPrKdBN#1=j8
zNtiHB>Z{Ao5}&l<&eWm9Oc}px1LCx3HOhlcBlatb1G6rN4z@?g<qI{pAk@#`f>})Q
zwJLaSx<Qx34vDRxN4RC*q)K#n>_l#Fpinf<YG?BEI?_m(_f4F{pv|Ff>3&<cz_aw?
z%7+lN(;tmsYsfgK6}As;W3ca>8hWB7qYaCfc57%PW0uT()!}Br#JpQP4c}>zFK?hu
z5;x`dx$o62;k8=ftUVAT_3Zww5s#@{_)=j3GlDOjh1EI^Vw&AOrd9yL(_i4O8T#MR
zIHy!Z3+ZH;@6kQQE;Et_TKdzIax<E#C%aX!f}9$W0h|AsA1t)7W?u_uYSXSJu1xL0
z><o2zpVH|FrHaJnU_R1t)DmBo-FX|3(~!U+9-n`Qqpo-BtPD1dx3}ybfUI!8wuf4+
zuAk^o=(g4pO^5ZO<HE~@0w}U5n~tz2JwCAcv8{!j{MORvIXx`<E<b*s$S`k;fGf?H
z_O}SY{KU<GcuTDte7ZD6_dGR2Ixwvg3q{Defj)JP4V{t&)WGZudS$&z64+(;Wp{<~
zok;l9dolXHqXh}j$f-BL7J0+H>!*_P*nMS^qYb{BaARQqBOG1=>>%-T(@a-x0$(YA
z<OfauaKTx@Y4$8JP6(d@$6Sh7qlS{^4LB&yXZ2Et+1KJp%r2bxxim<#EN+UFxJb;?
z?4G-3E=|TfDltQZup;xi)~LakyFG!5J2025G$Wds>04aB&ny%?+JR?hM=P@F0sdYB
z&$rJC^SNR&plPRuIO0BeboW@6xY13VGZ*<-nf=FlnnVr-r<%#{zrn-IePd_o)KuyP
z`a5PH>e7Qso2?@~KwN~mYJ;a04sLDzG^mBd`@kVSU=L7Od_`Kg)sags>PfseVPKeP
zVv5b_jbt0$qHIE&kdKW5%`Eerl1hYq`Zu%(ZcrvaB`Iq|0K7FcKLW{{=^o7IFcDN-
z$C;nmUB;Hw5SxVfC_Bk_c@B+2O5Jk>Ar_}Q{!ZN!iNwku`<Ak5bpN&Uayol*k;uIt
zEik(Nycz7E6Aw~VDJIZst>)tI>L$$t>NvTpz>41`c^D<&bP+eG5%;fjjw91K>gBqN
zByB*D$EcjdY{1Qcy8n}OZ3OQaC)!lZoOEB_60Fp+%zV3j<);QW?0SVRhH|3pkUcUs
zM1?cU7MOR}=&&4(l`EgGm;97aQS;X8f&x<gw1S28mn3mWW=*Qp`i_Cg)Zh!YWqIrH
z4O?AJ&wXHIpMBdQl|%W9>>bpmYU22A!{W?pzvth`y{x~G;Z2sW&5hu9=i>x+|7zm!
z1}3uI|CjYJw}a!^+6*bXMLrJmY&k>2=c-XIY8~1O$@k`W2yAOI!$*MAlf(`^4>*u>
zV}fzDMtM?713@<Xs&@hx!rqw!P5cIC^aNC9Yv+B}W^NH$W=jfVc-lj0t%=tsA&#)O
z&cifV`KQ40m@^Yem(ZvmfhN!cvi(UIVt!-Kg>UsI(eQ!?VfINs?J0Oit;Tk*onxCA
z02i<ZS-u%Q?N`3$?{`Z?o?65gB!q4a4pqg}<#Z251_9NKW*S`YY}LhQA`P#obrebQ
z4zkXztdeB3cA3dO2jm7mJgXUkWNJ~GKUN)pNI#6hNogv8w{I#?hwS9@o3+=_pGT97
zVjaECG<&FH>DJCu%{Zv*koz6_>S5EKW%iGKE5&oroUW^3%=8!igyB08qhf|xXJ#t?
zBpu_HI)YZ4QamHyEg6i?V&NFv0*{B8Z)fM#FX2C!AtH$i2+-nNbD@uKYx8Kd1k~Ug
zA?2*ITb4F4pUYH8+e4ONWrkfL>*@Pi?)%iCmLBv1*+1VXmlSs7ePJmbHOI@rt!=`r
z{JIOe963IvRcNUybC7wnU*SvFR1E(#I}^8M$hc{FI~s+`e$OQA2Ia(BQ`M;@i#ygQ
zi3GA|?Q`oXt<IH5V`Nq`dU?v!M8)NfQDZ=<6QEoQ9N(opLB>`er@7>lAe2{T;gd9x
z=6Uq*@uF(0roWEPk4HzlCEf$;1wKKHJI|^C7Nu5&P^_UmGT@Lq5w;Mee&cT1Ikd|l
z1^d#9%)^tf#0F9W`=(C>qm<eFuI2lD@$s#elu?+0&Fjmi5HM<ti%z885z<{&4+4#a
z($SO)PB!ms5!ta}OGSAm#vU`}khq*`+s_VhnbOPt?%wx{8zdHoCETi!2<sl2A?n~K
zG1Qpvdx0!QX*NA)12@bJpYTpCACm#$tz`W+yuDRs_Zp|7L3j9j?#GZ<Lu18HK0gY6
zlzm7bCCwFu^crP=IU9USAV=GGxxx7+PgZU1X`UMjGrg2HPkq<oNnmv8;ubZx#Ke{c
z2Rib`t|`J~Baax)w(o>7;lsLihxdy!g7D*IbOX0kP2Il$DEuB(#g;^6n|?g^v$XZ>
zwX+9l3meSffZRkISmrGfoQD}+xikZIj;l8-vaU9REP=O0tHr397)8Aq+0~wMbO-Pk
z*l78uwlYP486=6Ef_t-yC`o2o-ZkyL(hWHfM$ucXJA$g8Wrhn%5uAs9xep+WbZDK8
zQ_5e%s>>JYB*kR0deeGTK#wlM7n=rJY>jNSc4J_`^)%?FM2+u)YOF!tk=Bqh1eQ2m
zNtQ~~l*|(t*_xo?_%ZZG1h+0mp1UjblEWSsXLaaANv+@^nAza;JrKOcNsO7gbR6if
z2#gTw3vV2^Z>1=x%zpeGef0>*Sy@$QS85evmooDjB3HVR&8hjJ7~>Op#Op1CQ8&XD
zccPZ<`(zx6m@lBx05rS=C?8;2fIjunuj0JLjOeU(e~$0`A*l^7DI5w&4pO)5{>7L+
zW8cDFJ>*99;l`F;{n$q-n-}R%Alcu%A>lvTIsE;l94BU~sRMJb4wEBv%S}FH^@G;f
z7y8nhshe@WSLgv3gYydeN$2c<(n44;5E1C2*01&`)rO<wA^NGiGL;i|B{5dd3F#D>
z<*DUe7uyjYLNfIu{UOXymqq4=EivGfXsG(#yS#4>lb_l5?WM?zwt#%$Ly}@WdKbtR
z<X6H+Fm+Qlj-9EM1FC_tEHClY(Svo12wN-p){>>k??7-un+h*wl=-v*r|ezw6K;oE
z%cAoL7VKS9T=13!yzrFNrbLp-iYs-|Uu_*QyOVKNveegGq(6@|)oE>K;xpa&!{(xT
zd(dH>+iMWmyjGxX{4{(uLmyhV6sjgGHkc7aR^^DMH>+D-+T&wzL{Ptp`(8!u2>Dv1
za;?lJ4T?vNHGvR^VlEkGYUaMRMLGUZoP|k~&xBmhE3116%0L5k2>%%C+fqX0#FPj`
zn0<Xmo)lb9g#Tv>QateNCLqb0#6;Ly?icdOxn<JrNB^8EH5Y2RG%t=6g-noytQurd
zwTY%Bk~(!qh^<Up6;xfL2HCSq{7=1mCs}x~Zz^x>RUOP~JvzXY7{yA&=7_^rGr?^s
z7!77X>FT_s_v6*j#QV62{**5{b+InK5G@@4*}mW90;E(K-rrRE?gzBjVjd6D2^Ce+
ze0g{s+J~}nntc<d`&W6kx4&B&O_C#fG}rHHf0rw|xhY44|8NMhFG7G%`ll(+eAGa@
z;<r3jBm#$}IEg?^$;Q-X!OSgnH6nW9(;qNb0;rm}0us%G*@H-vM=(mj1FJi1qridt
zo1ZT_z<efQj3gh%JSB?A$%WhTnLTifp*LWknuzbVsP?fnqMmfid|OTgYN}pl_<3*k
zPs}-@8sql6)tZc#_$p72nWnqgk?7uDHH@zZ+6YXVFo|D06g`3F3k~CIFqbTJ@E~Nh
zz)!VOVK(3pf3K;$9w9W++e0l10&IX^9BEhA$ojiSMXFwAn%{hsi%VB*p;ByrG^)mC
zpKp9id^FM_r@RnQVql<HZ13C=;e^DODbbUT73)`7iE8CL9fdM!h6H@5PE3lkhJd8z
zeuQyZ5krvG-E3AAu!db1pU`Dz^T)5OYp<UWO;-a;`koJxhRnwL73l>ZpjQ&_!}lDP
zjl-A^8d8eq#?0Ob%<oG34tElIva)voNeWb;P}`rGo_uX4lqNqbzVDF}0)F77dA25<
zIlfBuPL+L6^yJ<U$om@`p2a;PzG?~5!u)Pe;Ii!mXdjd)k3{LDiw6uN?Y`>aj08&L
z>wMKcQA!yB&?n>GKRe>~TJ{F(S-nL14Ff7n<M{r}lkB8G3~X{W7mtXLfO~w#<(wQz
z=?r-rbR)OAqdR2MoeSCbY+k|cPX3L+W0r41Czfd<Px!iA#<cGpCPHXUmfy;J#UL*P
zlx2$wX0sJTKaPf(W`JMq%OaNx1}8dNFOd-N-+Pz&k($?mix*q>gg@^J8RP#sX*@Bz
zJ4RlVB=SpCE@m4u!M*$cDMENO<eMN*ziM!lZv1kZS?_ligRUB=&9My8Bu=X@U{n6;
z!HZdUJe!1%Mm{40SIi=qkv~~ouLo!qK7msIWLN8r;PFLT-M{UOaxtF`gT2Zp_WcO-
zttEzZjs!U!UOnpR5g4_ke8h~vW}E;bElR6zaMAJ5#>`6f0fx0Sy{=yORgXxSJcxd-
z?Syz9&{;a5y2NfHz{TpH`PN1?TdCJ!_H)0f+y$F)u1Ym4!U!uF^S171K8ur(9di=*
zr+F;lJE&7bYFbzOWZ{Q-uu|+SF&{<gh{#dJVVTv{z9VQ5kq)<0SNp4G5S98z`J<4W
z-w}!8m0YVB;pP{{WgzOb(#%_!ii-BAgg@phsIN(`BRkJA8yXA@;vPzpPDwW$Ve1T*
zYVm=bf)Mual0G^WicqhgE9f0SjC?;0Cl+I~r=4tz#_S&W470mg`Q_23cUD|YqB}Ba
znC+-QVCsa(^MV8gUn(Eu@;xJeNx&pVd4kH@%9`zgbI4|*jH#OEHX5#5po}5t`o{KX
zoIxTlTW&rK<<=uzixlno%z$ubfE|m~`M^((lC8XPe7==t_-nU92-we8-ST*V>sMd)
z4qUD@`x@co#D{w);ra(W(C(25KG55inMdDQR36DOgc(r?Z7=u}G4^f1L<5bwLj@wO
z)OUnep`I5<#Z^_7E*t#7Tln=W#*xS&lC-#=71~HaO^*9&;f-H<JIM-O;A7RwhV{Ub
zMBeIEQt(PMcqLDSdr4VYy(9||Ut?MN!5B$>tVodu$Yxh3CuRX_`FtTcDl0u3GjhKD
zC4c@EdnGb(`kEl(Df8h|ke3QWSPdfBa6}d$xh%~T&)$jRc2915%m{}nCH?yR?g+{y
z(xkqRWW0tMnaW^|gV4R+6DW+XJMLA%<R#oWOvH$APx30?E<twa7pA%rBS~{{&*Tsa
z_xn}?%r0Dfhm0{x+cNaQidd9(618Yrnf-<V*roS6;8hol*nqb5rM_fcy-*XHf)4kc
zbRw6wJ4*EWeiO@gDO~&cJOfedyFtwk4uK~>;7|{%k$smNfxo5G;eCUkobaM0zGtoN
zN;0EImX#Ke?~PdZr-`GQi0vEPppMsf3UBB9%pN2rM+5J>?n!dtS%9l7*}O08*-|B8
zmAtY+EBT_*@(SXCnVNcXIlYwSx)j}B@TEe1zmsAfJ8_Ks5&v=L{w}vJ9k!-;pxv^A
z+VWv8g|3p_z5)mZHU)6l`!0r~R_8AcU*yK!Oq*~dNMY2yX~l^>JHq#wm^#cHyRku4
zisJ%JYB@5UgspCfI!Y9ozU<MZEO8n6seV>F5e(XJ<8hpNOnvJ~U^5L!uOEfC)eW!W
zw<|)-jL7Z~&7qHc!Jh%HDl-gt2eY3MW7X9LcboE^bce<1iy6+f*!$YIK~aP~=&^Cj
zzH23@-<`2v=qTmSJ0WKv73V8^;}2MwP1NJmRlquoG>f(RMkJZ-PgwDihKk5m)TeHy
z>%7M&?I%t+=9nRg8Bwk}$#`~2OrIVZ)(aH4CbAJ!iM~l|=`>uhL_^)}B;?oq#y<Wo
z&y6x~>78EI{(xngFOU9)diF<)yy6KDSMRA?TF~TE4w@^0)tHaqLQzx6HA$P>V+=A5
zG}o;(_ojwD-nM^&`L=0ilT;N!?WoO}Zaq2lY*3=EF7xZUA(U=C6QGv_^a3}`c`&sS
z(@e9T!t-`+?Q$S~I!Y$`!u}hCjxkbw91|(%jMczc3-1w{HevQRjEWlXY2AbyW4?+9
zm7wk;BjZbl(q$KWO>&2$28Us@M~Bp%JxOQol8nY$f&Hy!1<by2c8d_9gr_c`p(Uj{
zZaORNbtP%ep7jJE>EElR&r;qZ%=@!0rZy5#OenR>=-U}1fMc>uPhVhPqzJAmwMh|r
zK_=ZK5n^S9V~$y5AwCKBY3sF8E)mVg6LV=wXkN)QNWu(BGC17oa^T33?*p0|Jzbc=
z;K{p(Ej~X!r84o3iFeRbt3hVSHUk%X-+2J(uK_K1;v>tp$qPgs1q(&<=OtD4g%kVm
zvyzhYm3BWf+E0<ae5Kru*~W{qx9^LirtLrCii`dD$bM+dCW;zW^mcmY7dGsz^e-A2
zBOxXiGVF)ilyHVEYd9A!Y#32xMseY#8H)I;9fm%nz^Y>=9X=U*5X<y5gBiawnhsyK
zhGKpwu4*e$%`wf0RFJ=ewxsV_S1+)EiV^NofD&d@61W=Ty_Q6$7kJG&G4@MjGe#o(
zIPXUr-^dbBUxBXJ$s&uHfL6zEt78mkP1kWsj^Ney2Ucc{q(NxC_V^&nT1p+<wliW^
z((e)p3oT(Qe}YTwlhdP>e5JMb3}#<d8!;YaydEN}J$Z#L$buU|@=CQqY)W1}WqjtY
zzn^%=OzYbYz}PgPMMSG#7X#)Ct-ymKZzOIF7o$zh0FtBZxx~tHmis^g?7Q368LxJ;
zTP9MP3cD9dWpy*uDz5Tx@Q?Kx0O0vh00EQLs8@s42f5-re^!G)CwPV10)X7=mJc23
zt_``3erwJ|jW<9eLp;D~hJ#8oIpv>2f~y{$r-r^dL75bZNEIf4jNLNxHR`<Wu!)qJ
zn#@)VOyfokO~E`;iKsDUjP&TjMmz)8za>e>1z5AWq0;`#?q`0mR4cz~D@mj;2pTwK
zf<)tFlRLXDulUCLzS&Q8b%@<aUh-R{e@;!~>6XE;V&5RX#G6nP{$<oJ&bBHt;ulU?
ztEbsl8FpT(FUDL<%EGY(KyL!_t6S1?&DBJTcb>W$X`dbB2jF#UCe$u&5OG*b=ngrI
z&Nt2J;$KC%hfgHwF%vpHR{aX7pD=F@?UT}+m@i#^f+it%5rfcw1#^nd3mB(&nbEio
z$JoF0d`dPKvca0*YsnUf8AZNTenLJDoI}VRN6Q@SC}qA9?t0?;xS!emJfNm>zgBjz
zW8SpW^9CPO5}*0Cz>9R`zRJqy(e$Nnd)e@HrD^c=T0fLHMi>ONt*FzxUdT}sC8xqk
zwd2cnElH4QUi-Zc8usIMyxjhoP3ROrwY9K~0J!T!pfUE=%-{!Bmed`Hq&(zW8a4%V
zV70}u{MeQU&3X^5q?>KR`WJFIIw!0p85ZU{a*#`aQiHJ9%{cO>eEl@<5pr5xRT(Qg
zjWJU@4hYj*qp;!Z#En5%lV*f*=7>V!NWO8*behI7PD{^CGY#A%??5*|L|f(yDp`*1
z+P^_yA$lX=mi>K$kXdRaEito41N#*MN-@>T9Rr7<iK&4N&8qB<84Y`FU37NDInI)s
zuSK&IG9mR&Xl{8k=H1UXw4omog@;{T=9q6Usi_9~DJBT4rndlLzUqg3B7p#>3WUpF
z+OCT|QkKpXZZX>Fv+T2JD%ziuz6moq(j=HJS~iv*%G>&CK$GRfNnbbJ6P)_m-_`R{
z@{SVF%Qgje{7utjM&k*+!+fulr_?HT(&u^PJQQy08Akh=J$7goVd=~Au3l)PZ{$fT
zH$R>ydSv9TgghJxOvUWee$h;PPu8U|yI0*5TKHa=T2~M3B?>8Woq%oaGOvkB=Tl`U
zrAgSyh_ZGx{;kwXh&>g&55Cn($vhIogY2)i;#<j*NGFf{FkdRvVSVIQc_S>UI6c`6
zo5i_If-3fBtiI(<UEF7aN3cQyv?EzBTzZ(hYLkS!=!vYf4s|H<J5$NMb*($pRZ#6&
zNPmJ*>3*aa`~Jr8dy{K?65Y-OFCUAqzNak0Yqbl&x<sg|eaa*^m*&TTq;{uwQpQJB
zh|wbHcv|uEV@5GUQ3a&s<GUIu3QYuJjaM%ZL}ri>aHO=F-&s__$R|OCck88siJ8iB
z_<FN?blem!Rv9lj2r-X0EAwl=`4I(7`7a$=h#DHrJ0d}9pO&03m{Et;4ha4HPAehW
z(;nr2l?iCfZ0}JS0W(q0XpnM;Yt&u}j8e6B2d9sOza=V;S%HkyL>`0!tISU<zfZZq
z)FXQ8wLBT@nh5&B3`u09E+Ku&<sPnC9A<Qr<TnkVrg2Q-XVXox)9jJH1_3MGvM1?D
z+yi|q2610;*DcwvW^}CYV2?}tks<|7Fl3+gJV5`ZN|i*UubxNz+S&pnHC5K3_A_y%
zStga3Ez`E}?oSqyF{=m|YicL_i;cER@aRSF{AWEUZ+O7P*BJgbp)65<ckw98VLL4V
zm<SX_?4ivAZ7o%;tiE5!S(vxd+X?YvuNphK3-~h2crCE29l*Gy)x-7{P4f-@tGd|d
zqhVoHEeaq$!o}WARO-uV*M#X$44|g_37L;q>;=!QwzAW*;un!U#|T>vNX;4=N}6cL
z2Uu<Od=@xI<s4Vm05DI;Y7`muD%N2(Wrega-(<7M-C$`Z3;y?4W-5090=*@&3pCY#
zfPg!q9x-c>(M0%Q+6yuzk9mV8=Q9PMdp<Kf5lU#1sOo7&ck3u50cG*(Vj_|VZK3h|
z?ugvZQQ~O$V*lp1q%A7;5RR6*8obrfARgt7h{mWk&~rG$>z%)`&ulvYmgBco0&JN#
zde_goDPI0W%SDn#HUF9YHkhZ-Yf(Duxy3`9t?mRO#G4Hl)USZ4JOvZCXOHevZ<+2W
zqO~lPll|I|O$U(hjBgO=Le`A3>qzEF)^1avM8z53W|w&=YF2P*C%1(|Q-EedIc&^a
z4&uO?k)zzeUV51kz_w_am|2Z@%3JXj@&tXP3Kp~Vg=7MEx4zn{u_wyPQ3KsG@r!EB
z7(2$^^f0HDT;)(1m7ee#`AX#^%5ue-f_V}9k}_fEsoA|oza)<`N)iO?-Vv|wl^Wfo
zl_oo^Mu2a=#wGE7BjAGbsjTi$n(~kot8!l;z!rP3wz{NX#OmI%ov0kWWb9<=kB%}G
zJ;_Mw&N93A8+k=klhfw_;-X9?2g|F-gQN^O4neF5Z&#UNvyKGkwv<}I49ixJ3UuYW
z`O~?W2z8K_Onq40Z;vsg&RgbnB(DfaEjS_5>>rou>UL5o2dgcqV@Wi0e6xm_*nO-k
z9fdkEZ-`!xKnqp@1+#zk(MUhOGACwQoP7{AV5B~(Ftq6CDdqb_l$vJWdb95y0-hS=
zzN?v#>R<P}nuNp^yqrwps9)?Im@M>44&bLUo78&*jehT248A>)W?~RJ%;IX=uz;-U
z_XWZGnLUWSifQ~IGm|0=3p(tq%5yEndf?%Qn)<P|O!`wkv(1(YS3PPQVTYn&$wdKH
zoZB$-5+}=(-gMHwsYl0l&;`)6oXhyB#<UcRZL35*=4&$`k;uB`$;8`No6`$~	{c
znnXLldvEnEypV9Q@W0h|7-n#+S78*Ip|{@EgW*s93@PqcPfud#t|%~^Opb`ZX0O;<
z)C``+t>UE=(=u1_g2Lh|aM8yZlQw#q`4?rmEJ?jEEBWz6xSCQhSUvKMob4M0NG8d4
zdvLoOTGOYjOE761K&x|_b}4eDO1i#h-^fE8Lr>hj*<nf3B7{m9AzxZ$483XX{p*gp
z-4wS|gd$ZQqsgPw)E~*Fo>@@L?kf2;R|~joz7<FB{HE99yVMs|1ZJ>h?s{lH&2J1W
zrOSGs8+@b&c`wuA)bV!H?U0t6*;uOb-ePiUDjvx|h51TlIWl3%atT~vk{l4;(Idg1
zdE+t*zp9c;ncr$bQw#UB)zkPdNJ9;dvA?j$qTmM6`Wx=a9yvj!YYr{<`~+y3^fY%G
zNY#^1H&l4W!dN54$4@}vHQYY+H=uE+tV2(-DdC<b8ddU`ENNjeAN7o?WYtrifPj|1
ztQU2bo-Vq|y5jl5wP&lSrvcX1w6pjzpXy9<UCDgvaO^9CoFKjJF~}sYt_f~P7$UuT
z+jsHt6R&9l0si}HXf#cC6nvmqE;HboX!UOtd`_7?`iQG>19rXb3!q+^2zQ&;s)r;X
zi*9@n+OgS+JV3O$lq~KJ6aX&C^n+g>4zz^lGoJy}4>K)0E6uQba5c@x6$4-fk7+@w
zGBTO1Cq`sIsY6c6n`Q*vqtSm$tG&z%Hh!Hyz{7f(;Ym~0>z4AF!pnXFkGX<cD9wWs
zp@kOGt{YB}MZp~8zfAuwmgt$P*@VPt1yItom9;5_z%U~cqmzp2s?~$TX`9Jr$lGU{
z{aJ7H6+#P9LtJ5-)mu&0r>sx%FC?Rjb(i-z=@HlNglAfldk;c&T|92`psP&`MmQa>
z2)CXx8=Of2N?w`1!!aLhyXqsODcZ{Hn^>fU>CK3$m4&8t5CzZ@9?1_v+$P>!4!x65
z6TFfmxH*U&cc){Mle2OplO=e4s`gSaJHlGV_n${Gr2^lW1B{8spI;J6dEhCOJeR;^
znGNQCenv!nzgyB=(}HrygN$2OZ{OLreB(hHudcqN(4su>5<x*0j@-snZ0o<q`^*H$
zQ7}1lFsGRWc3e%bU#?VuGh;+cy#ZAiWsdLX36+0;=En}CR3u7dFU?|^PWb1^VvJ~j
zHk|s(cDzg@vVlYQr9`Ya<4>7!poGOl=ja=R7M&R#WV>a8LVQQ$*yR@Vn?19nWB_X|
z?YCnk{XjH;GKfj3EKhID(gDmgKpx2^YC({rre0C|nGI?QWuQ({`w8MLDU2tD&xAm!
zVf#uei~fA%+|ul)C}F$m#UtG$wG&8!OkJEkk4Vu|+{`50Oorpkd#vGGFAyXsFK3$F
z^~b-mkQfUu4l@k_CpiPN-GytQlJAk<K|3kUTX>J~Cr{D^)2A@R214wH-e3*lp|_A9
z`an`97?~Kq9Q+AUX}kS1d#pEuE}5$CO0&EELaUJSKUwc!wgW7OPVl?xlz_Q;YwRly
zf&3Kc>QVmd3dWc4G+{>xu9fnKGd!&w=EA9+90j0TW_RHHVgXuuA|=!{w^x88LDRe%
z+I<v@oFNjn`6ZARgNa_|R`LMqAbpw$NbQki&Y1l?dQp<Nl{19XCO{UH6>cfsV}@H%
zC(x1aRA%6NI<NdqWiteQzv8J?@Pa2?5AR0?xnRLXTgom`+Czt;KGpIi;LV8inK<QA
zPlbb<TDxFZ9NP0ULx-kY!pZlIZFGX@!=|qw{8O3_mJT~*57x_lie91{hcbMo=>p;8
z181a?YiH!Chs=9}fDhmb6au5nd*;Ye$NbEX@Zzd(-mIoE511CcK9J*O_Mht{Sh4N$
z8Q=%Q@1P1dXO<6DBgX^!6NcRqW{BBTQ76l{O5c)Z404Y;B@gW=^)S?W<7;o|W7pM4
zWP?=%K=zW;0`o<FBNBw!x`QX(#5xb4rN8X*!t9a9i6mpc;~N8yC3dv-)P@*l52;(U
zyhCoV>mi@-Dv`iEpJbVM@hu;6Lva&VLzd(3Rj1@&v;54s;{m|%N6%F$LRhLn0(tB@
znBj^smViQ-qAJDwL<yigL{ef!DGlN<oy$FCGBCT7u_8BRs6Ce%fzqhT?#d=g8hmIX
zq#I{XFN2f~-Pi$_D&jne480q9&3R|$4Ml2#l7?2voyjsP=nKR{DvzA%$;V9RQ<(cs
zG`~f|)6p?b=l`i!0$$YO(0`VY0O>}72v!tnJ+n)gO$A-;Qm01tj}TUK!6MH0i2l7Y
z4Hc+Xb~=)S@z|mVS&r>w_fmcT($B%s1SYj+hX&SC7v+RoD;|+RiFTGmE4Os%K;!fl
zAV_z!v%)Mrfr}E9aA#ItE69kJw=~F<dYO^NlFyIdp8kR4p}8eb4&~V`^H#55Xdz`|
zA7(%y_7_mgq~)Tmel});r}C;ef|>3!b%;$upPJxAay|E`_uY}G)RYTAyd-y4WaR0#
zR7nAKzp~yoB3bGCPyLKE5go?35=r~sDmv;4X4@UVM*+hTgrnpeGZdg#$V}-yc*IxT
z4-fGw%RpRFR_Lz+M%jDuyJGeuU<H_({-Jt609KLi^i`9kBB_}h*gry&eli`48HGyi
za<~%d0D-76-IDP7uKXO~O$Id>aI(aTVnyFj5dt9mS$Fhvj)_p$pvi2Md2Xsi6kFX-
zj9M>da`d{|?#Mm?%+K^jU=ardo@RV|i-dx|_Fgjq-5GgMcKT7%-muImYszEg?hHHf
z%d2T#T{SlAaAa0uB{Wj}>l<M$Hm(JmQJHDUj1r5-p5Gwyu!Rwm=U(sil4*J>Gmc#@
z8a&7dp6V=IFPqg&Apkr`PudniW<SN%vm|fD%1C--%w*Xan<HpVot!c+R*eqSBU!94
zk99Q&(RL3=$_(8Il02#w^w9B|!A?RF;7K6#SPe3t#8#d@iK#3BX8&P97+*Oo^VHS;
z+n`CMMK#tpL3ASN=|+;{&QrI9ZInT%l%BvHVMedn<X*w9UDNh6cSlpV?I<hC>TXnI
z17}iz%+L(cF96T__kiyJwW8qU3%J#Tke?k24=#c}H7-0WJ6IOZ_-5ui0e8Df{zi|~
zl(Z)zF#$e#pB3S`fRg&5`f}tB>CWDHN{759Rv(Uff!H2W8$xE9UG<UrC-pN=dgqi4
z+)YXjX@=9P2Qg+3vLnn35~R!>j>v0Z_D6uLlw$FbpO)r57@G`64<>o62bWCgPCB)}
zq0bbElrb{#B`XXx0!>W+&~Huex@Xfw$bgPZw=(;3yQ)=OG&AdtoOa_T4bX281UI&V
z7x(W1W-2qB7K(sSfq1XX;2CbzdiVN%8jl`kW+jyul$njXGq%<k{7YqiQk^1e0+w#4
zxg*PvK^HmqH2Vud;D^m`zhI7$c`X#8UUUrYTV|h1n~{4>?+o*#6Iqdwe;Aez4=$!@
zY=KDUs6k+NB3nty;mx`upuVjNYbOgk%j|K{ec_$$+HruzP@B@T6FSu(C&cL!`f%z^
z9SYal75r0dR<Fb(+Z~~VyLj_{gV^nnmZTxu(&|1I846l;2(#*89tv05hv*voh9(us
z)lcOmtD)goh!6w9vR8xnuBkgt3HDmORs+Z)vwv(+UXq!1B5h_&KgO7lf$7mHMnaZa
z=BRE3q#-!ZFn4jn)5u}9pz!O+I!0H~41_pl4+G(<SNBN|-WZwXzg?9PU#n@r+Qz;m
zLW!|z8n8(0(8KN7a3izQdhBNQ^0C!N8Q$ma!e)h)%^-l83;C^@obP*(k+6FbM%2}p
zd}a`BRSpjG14|vBdY|*vq*EvU%C)ptV-t1m6ZCRH5-uwskp8EJ88iT8UZL$1TMk>I
zmD$(I8_}PVlco;}NqKGVCq*=5_5*8XP4b-FpQt}052V?7ESet-FLLqP)wgT?yBa=<
z==m0HQ%w(8Yt#O;A)cLSUha8E_VJQ-*md;<iK^ZvVA6F<dKBW8@2s=ymh1JFSJM?f
z((F+u<#b_i=eacqL9{zrqzq_2uL%OM-MLw`;FrOS8hO)7J;WDv^+k_BFo+>Fh&21>
zqd{;jZnL`O;|NVa#@@3AfeP|z%bAGu)WgGh5j44%xTMBLaeV6A9doQa=tb{_N)*Mb
z8xq)9`)GtQI2@-34oBv7sG-PD-`2jXeT$eH13vCN+sIHJNYzwZU)O4ou_t#F<;-gN
z>?C&#KjC(1j9g1I+AUN20{X_^v?qn%ptE^r)E$8{w?IRYcCI_RdnrjO+LFL$mieHu
z!b2iBVV~JZ<_pI^%GaXl2m%H5-Hdi>8r$^YO?7|onlttB7dUg<t!8okwNLFsuSW<b
z>sF0mMW1zz016{o-3>DKdPjcTNRA5J;t8J~v%!oP9j0`m)#Qv-rrC{=jo>mH+lW5X
zv7%#nB7Da0qQHfkf9@>@!hFb0_{m{hsjE9!B_(Ow{=R}kO@9RnRI0_6X0*x~M0tdj
zzA+7&%1p5NwHAt%GDD9MyWF^eD3ICL#~G0Vxa~8$PfKb0$%|a+jzh}b$t+&p<#%vU
zPP<4*31)X{2p;@pu)6lHXY^-*`9lcyWBCzC32H_{cfI7HmKm5<@TBi+?K4CAsEUU)
zh--Ra=1+9wj7*Bij1IS7En(pDBH(Yi+2w$O#I6@*K)_V;wB*Z`b;H4~b#_-RG`<t4
zi`7k|<5*^?l|vVF#o>GNr-7kDxh^$A%(*_Z!GI#ik7{dAGue-+Na?_fT317ZbW~00
zC*@8t`*zALUzEB}T}@-a(b&A2b<#XT>iI>ruh$^xL6GMaAidlO4s(e;-4&t%yUb1b
zmq;Id69EeqY2LYZy{RiYz`%9xc7nS^W;;EAq-J-059xlf27y6$wQnM{GaP}@BzjP{
zd<dt@4fD^id`d`nwg44$AR!A{<_DK{jGd)q%zRB2Xr9*ZXFd2JSC}%}_2d?im#Y?_
zm##YfV~-muK07_#AJBoyHL(Zmo#7|#W3>LN`FmJyM$~<z$}6wKTCX9-d%>zWIs0~6
zC!dhg4)9^5=M9htQPXhW@2rtM+?L9cB+BNfIqy)Cy+UnRO^#lN9yM9Gf@>OQ>H`<q
zilhh9j9P}qpkk7rqqJ)!;3S=qmL3$z10Yg?9bPqOT2GFU+O$pdn)D;XrgR2+?yu%y
z>M+QBe^eM`M(#ceq2uoF5!vE0J?m#ue*e&jL9!oxI@w7i^FwG^hsVk8b<Cg~BO^?k
zbFV?%>2Cpx^-uKW%QoR6BH0Wo$;v!noQx(D|Ke|rJ)jqQI$=*M@0Qjvd~0qAALy=I
zqTOdRO5JnhQ-bC}DpVnSh+C!kot<fAKk{?+;YLH<raH<?{dJj9<cqE7q|Um`HkFQ8
zK|@xzJRNDmKsW3?>uN~9d-!;JuJ2;fpMi8sPCQu+x*=uus3w(}UoTRtyC<un%vXAA
z>c3aQGR*W{oTx+fe!d?8@}8wk&KYOEnuz)OG7SRM`McVVIv5CJX?3Z$&Dp=A02KXJ
zvjYhl60KU=NL5*q+5galMzJs6$4nfvqW;|j&W5=U(OlqWFN;e2lsYF3NJ_qSdrFea
zJVp7>zL39pCrR4Ti6VQ9lv9liZJlBR`USrGO=SOcF>*+!p2gn=&i)k3`>IEwjVrp+
zW~9euL=Sjt&+2vB!TCuGtvoY6(XyVlqQ{J6#;zg!*U{5$Awzo2jrOpiK+czZdX1QU
zCm=Zs(Q*qg5N7!IGTjQIyv*(vTJ+%e-oAHa;MKDIg;gapk@?cl0NzLLl9)G~!9fW!
z(qYC7L5t{r{J0EOZOpW+j}aT6@Kb3vTEK-EnJ1_NFpK%?Mn~hk-_<b4Mk7CDYAdZ%
zj1a-|JYHS9ns6USr|8NR1`{(Bi*hW~q>yHs>oOVaang4~fSZA=g>w1S9eqthJEzW!
z5elg?Uns!z{-*ZPF#E1SDN`=2weObrwN{m%s1##{r*yse*|X17W{AtQfq;8>pV=Mq
zp<M{+)d+<<;Kgl$tc6T3Hf+pmXeJ!^FAh?yeliBx=z@Ty#vrG-^wAkeic#i;XrPwP
zxe-mHq5(g_3gEOwqZ?*go|gq&q~*EhX0T>A-JzDQOb4_BOQ;@ZCzmtk!{N>^^Rh2h
zM*pRUq8a;66U?B1B0*un=!^Bj<EBimdj0ANFS@6~fcyD->Lr0B82WVm^bdTG%p@zH
z#I)A;Lj9lVIOX|>qiS*>1_I0K-$kuG#=F$jqPHz*z-4~X$~Q%>7AJQHud@K@!a4*U
z$V;<tZad5a5kM>RrT<06BU0Ei&2aEV0u?3lTpw_wq57$`%4xxRg!!N?7au9~>z~;`
zWb3mdQ(m{k3EBY-M{uTY>9GO@H&PC$QP#XK4P^S1HvOJG&2I~85pBm~);m!n;EjcI
z#`jJr1_Wf(ovoQm6{=wW9U}R;+RfdTwxI)tN}F(|>986lCS~51$k4;=+vkk_eVXI*
zFhLU@Tvv4axAy9e{^;mKvHV``%}Z44k!=NG@{IwN<OuJ?G8v4O0&_$ssNB}bi=D>w
zC4mj8XYut3!bbcz*^e)?3An-W3NLF70<2_MYzy;!SL3%v>JwGnT-y9jdj#|n`t4H_
zq$MU&wy4Rj8swlpy#g7&&i6R%C=3F4)Y4l3LO{L0?I4G^&(&MK_F1p6a$tZ*>t&@{
zy?4|<&dpUx%n3gsjuzx^R#sFPgI00Ad=&ClYc!qQmi%y7Ct7&%{d-I>YhW-V0pDyp
zxKU)aD<P#39C$m;Yfy~Tk2t@r3_y}PQ@A0{VGT{w^Ne%XklreD=$Wzszbq1AMOf+-
z)~S^lSkLSc9^PIzJfz7_RHbgZYMA}+T7(L)CH6Gau{DJ?F?Cy|nI@;09z9F1X8q7+
ziKGQc#Cjt}UsV)P{G-cHmli9I<H#-cOcro8T|n9v?3Vk_)pSt@36J$WHvq;6><_2e
zNx>=1$VSZ(neAP+!)6;N;5xaf!e0zXMGp7dA^w1QK(AlnP)2K=467uqRq$3jQ<#^8
zX6663r1Y2h@Sa$werT*b{(*x^GE0rzYwUU-yhs*m6MEy+BOYjqXF$JtUv;&A@)|^}
z_nENmpnyazDc&{<E^3+mz?cF2@XdA9*br3AxO&c7eY6rHof)80io|J#GT6sH_`vx2
zX`Ys71Ui%QFcyEIdv`~g*;%c{dRNK<iO6-OS#abXRs9!YbTY$Ip|-?Q`oZd(@=#jv
z_~PrI6%r(c^K{p>&&)g3X}n)>!(+XW@kH!Ddq4Gn;95?8Ude7KGlVw+E7NmkHM<AF
zi*l;e*JeGUpJXdCjv#M(jg8~*bTwyWt5<DT?hY|<v#i~B=zF0qW-XAwjvyWLQ{`J3
zDYI2q_fY>8KTf;(IX;mG`Jcj!3O078nYN%OkH`jn)@h!QzFs9cXXV{7UkT)>3bSf+
zfvDF;fJlKm<mp4tI-60F1<p|u^xv_?EujfHbu}_(3_?YFmlK8xi(sEu>OA|Dne^4s
z?$XD+R!>P`tt+fLbJfT+MFh@s5;)ha?^b^>*`#qSR}_TB-6D0T0!F<kIIrIes`fX(
zH)8buL_5V=XVx=%RI9N8BDDKifvbey?Z;1=31Z$%R?t%kGt6|DWCSqEOBec76QMvc
zXTA{nqaIza;{3GwOckI?fJngkp3iQg@rzdL`}x5}v#~|8a6uJNV~|J=yeZh?UK8<`
zfD<S>(F#_9%V~2HX~f%B4>L`@dT<(2sjHzELYYxkflsNf-o6!fIT7fR*3bvJ#Uv0M
zz1J8AjpL<I8k|XuaeG#M@XTJlCMYyk!0Gx-e*=ikM3jgv2jQL?8O;F=ObPjW6km8i
zwy(-OJJ&Bq_6#Zy*jlV!wA~Xsu*Y*5We!e&NsOStvGym-6Ydb@JgS(4Rl9XC)g;(r
zG<CvuZ3^(fh0ih|5I!rCkb{L+BYRAXmJoxq8xTOutBI2gEbM(j<R>$Q%cFcxvQA=!
zMLEEuzFNcI;96R9RXoAPi>1WH6txuRyC?ZQ;0}>V2$#W6e&%N*(4a8R&zi=e93n6O
z(fU;H<C1fa(yUJ?HaN^=z@z(>@3RKEel}X5J`fJBCx6r=w87nNzd_txrb4ZP*@>2J
z%p5I8y(rL$fA0y_&_EGd4}h`%ChQNqN)DedKpbeI3o?ZVJYRaRm_zajBrMReQD*ct
z<wU^Fl)oeWPLB;j92+AmFqmK0EvdlVG3sjW6Inhwsb-B0NMOy>0M27(hzgw)kbbf=
za{1A%bfvk2?|j?$lpJ$Q9|!y)w;t#C3O<(HuA!A7^WZfrfpTnr7o(aazy^D6u3;cZ
zcLhDC5kuBQF68j~5Kw22Eo8*zvlNHid80WW^P{gjSt`bx^-2Ksb>Z9FSXq0=%sw9I
zGpgzJrVLSQi$qvYFb(D<3JF2#ZzgvK#ha#?G<AgM$Bu|32rPMzzCDExi}}&?EMQy3
z!2=AJuO3xJ&0aKX$;^8V?AMwh%7S@xX$pz6*3LIHO}~NL?(faddPI!E0ei3I+lKd;
zb~+L2Bti+J!E_`p*BQDvSpA%wF*y*{-@xv|*o{ttl_4oszegZE=>~oEB)$w076PPW
zJ4l>mM(w>v@8wKi5HnxC$bX?s&94J&(YVoCFByq?9X#Nbx8<uT6Q!6<`t*F%BmnA`
z7e*q*W}wHv3Azu)E%Cf<wHlhng3!kFeJiW+nEn3(uj;2zW&j5+!k&@plWzJ<GaEp>
zYr0s&@gB5(F3Y^7pTKeYj45V>JEA4?W%X&0@wkvv@vWcl1(NRRBe+Aszvr`SQ*Hqs
zr7!%{EpJ<#l%XA95xQWW%_6%fmG#fOFDE$<?@@es^#j^;!M$*xA|_6=k*AmM2#0T4
z#m~9|5&i-fEwf4Rl*7A=N&qu@<X2KYJirK^TFf5A-pC9R(tTIE#|yqxt&=m$RRdH`
z)7YG+I}dxz^v;^<(zFWBGK2Wm0ls0@udzKcx-rNEA6$a~C5vR8wBo<3b?7xB4f~7;
zd+7VOHpcfXi#FCB(7P3ci<mv4WMDCSdbGFG9G?ML))$@AGCs>8X60H3NV%hE8~{n~
zk|GI&azK#7lSwAb2MmU2cl(<)g1;Wg_-qMeeby@)Ts$fOZZ!l#BSSd^f1vLt@i4y!
zkTY>E*xn#(AlP2-xs0TtZowex-t5ZT(bzSZ1o842fijo*y`05umG>OLcJ>UEDcwD-
z#)kGF5}Y`8xl`k~ZH;{-E!gYoNq3z@ZT*Xj-dCCTMtBG@j8_?E%&*@_D^AjOWhN=K
z6~}bnydr+UJgH4ZR69dwP3F^4vw_|I%{}UBn#8xz=LUgRJ>`brLtYCTTzI6Isoq4|
zVpXaIv#*Y!tHBG`#qK)K0H02!@XGA__0`j?y}#)}=-hNl`DQp;{Wby#hiBH47*pmU
z{WIVVFiL;s*EKz|{A>1_$b*(d)lIP9R<p((BL6z+s2ueha1PvH=D4?f7SKXS;|hP-
zNf4LW=N+B=#Hr~}lD}-aBfsyc*9R<UYHa-ScXP)&l86t0Es4I|RZhv@AnD@oDOUp|
z%~K*zI83J&Ag~~M)|5RtRgW920kA6iXzs}hBIZT8R^EvR3+X|SQm_~K+BsH+q~oaT
zsRalgdAk~$Log7w{!inZ37mF295xQe8k=ab$%O!bf9fgk5fimqUi$Z@{G(WMVMm+m
zd#!$+_sEkXq`E(|@A5mFX?h4eVTMGn?av{PpNu}b&FeWT_2s5adKBC$Be#`lZLI#{
zW6lH8*OzA0L&FTaQ|^*bIFBBpLmrxDiUYH~8yV3<hhfcQk-inY3-|EW)tiJ;(DuF2
z(i5bn{_^SuPf$%8QKTKwPfBrP#oq`_DIAIKyVwxp(osU^bFZtn0@SPai81PF?j<CS
zDzlAm+P)|yMl@9pS~b77?~5LF|5D0J2VS!@Hb;98%0hR`c!m)astEt(hYFCUy||iY
zmHgVl%!mwB#L2hk?@Q~~g%$Dy2k)*yXsC!hCta6O^uoO1gpf;QW8_Y(ULo@ES5^G+
z$}DvW?G0eqzN-n9_6uRgwsPN=t_Gq$S6Sjjpbwg^6p8DE-Tl_%i1Tv=xGkY01FXl;
z7{qUdOT~QP!;tqbx%x7rcM>dmz4ve0{$ww55$KGMdQ%49YZO0%9u%|DA5A6!aUMX0
zv&;m1fqA{}<&j|OdXEWu{Fd030k<-aF~0)Z%ikb_I8_YezRYhvCBM|js~7r6ugvQu
zxU!4sFs*l^J4AllRz;O?C@TvjfwPrGCCtDnqKO{#Lp>zA7IC=KD^j!G9DzH?|4-}n
zjKBnYe3OwPNwdGekRyT_VAj>ZEPKyXY@+XnsCn=rqSg^~rZSJGv8Q@*a+ilH19fWU
z^?(W2)kZp3IagFVzo!AS<5!=c!H5%N0!$H0L%{vQRhe|9!gsDD2p#2RR!w`zd9)t8
zHdY87Z{lE9xXo?`GH^zmWG};P%=FS}X!tdInIY0VaxxGui7$1kLa(ub-n>3*Y`+WL
zJ4|G}{1PU=2vrw_$h}YvlNtwA!JxH^u3uR}AZZiuZ7Ctf|EM~%EX#2mhHj;npn2H;
z#!43;{USfQPdCyb%YL>@ks#2E+$HR(ju@U7?oPW9PS(o66X)h2xYZpk0&|C&6!4`Q
z@51Pd*S`CQ$%J!;5|8vLVU{$$Jy6`=+pz*gn&VuP7g}r$I#iLSF|gTrcm5{KF4Il%
zJzSIola|-_x4uvD&9UERvbwV#P1%{KbKPvI-*Z-TdIQ^8ZFER6ezTUTki&HfnK?Ad
zkbc*rGhE=2K>*9howkGqh8%dKIv6xj(1n@<Z_<(Pl2P`hR`Y;`n^Q)~rcT!kZ#3`$
zt5J`8M3FLQu`&D6>%ZTZ{Nya?<;$Jf&a0zay6Qu{-=5;TpAdhSzu)NW<>W5DEXxY-
zGg8<crU|jLZpN*7pBtRmHoVCwH-fp&&RQf8(hgR`G|YbER+Bll^=owW#1T<9<Xb&~
zX;0Rw^<I$Z4d|o~(8N`$i8inge#_6*PKB1ck%O>Vs4Ivm?r{gESTnTS=hgkebUm6h
zgk1XvUy<O&bl}ngrU`X4H^kA1o-jAUt@+m~%%BT-?9DGmJ!a&!Z)BepG8{x7PQr*H
zi<<VZ*c$<}Dp_FXZEnBYjnZ$`)?>~R5n4CXf4FYj+IOSQq31#Y5DEOe7~0|87=`p`
z%cGG@a-)*MmG9^m(_C{rH+_MEp~)y0%;Ru|bzcP|BC0{&9=dru(UnxyOi1ya6)O&=
z;j38|lvvW6v@RbrPKk+`W%mQ-!(ctGcA~$kqh~xdc1ON(#@a&mRFRf8M|b+O)af#7
z!^uv1=5v1do2|=tMtytV!RW9lbiva0<QXUGVosveq!R!i$XHRY56(*OQovqX1U9zu
zAB^IaGPmm#YhepUJ}Q>8d%ja{6h>|+;YCmlJW!Bu{(MCc0Q}Zz=a4VW!Y5!-+@=E1
zMV8mje&}l70JEd#7_15IJ$S!Z&lAGo9wuqLo%zUvX>!~l@D`bv&hClNAc4zUIp#Nn
zluZfcdh}6o=&B=fsK~g_fUBGZL>ZzvK2dZ(rh=SKmL%XN8%^71)xfYrXj+G$2Hu1r
zZG|~I@mdZ%$O&laW6yB|Mf|}@(RVANy4WSgx=pR@Y86g+ZFquPdatwP8IkK8%9X-m
zIeEtYwnB#AGC$%CkP%upFA^=HoiJ(p>qYBUD>w<ZnL67xisj$YS8W#XuxRG=yxGuS
zo<fB<<HWlN+o!XHxyPf&)B7Sow0D$sLD_pG+(Z=C*hn9ogl$Zy2G>|`&{ST{=}a_s
zcOLt`>=-9)kw&6}0(cSg=}CvXxDvin;ATyzyT)gpt}TKpZ@zcX1A$RT^EPMg^357)
z$rcL7D}ySUxo)#Msk+K?Y)E%L<@;YH59a%HRsWzOd-5XL&0lRUP@tN5H5VI5OZU-D
zAyftM>f9%~UJ-mQDIdIpmqc|V)eI~hADF*t8aGHwp!?pDg!A?U*jG*YECRSCdpft2
zSwb;TEDYy=cIUP8A@HF3?8*np=N2$(YI`~e4EAHEZ;9`<ev~<a)!Sjs6xkx-)o@qI
zqu&l&wn(zcTs5!c(0gH0^iE0yaMw-W@_4X7A25z)+Z&uTQ4-v-C3j5FVp9)%OE@vn
zwHa|cAGofuktOJ{(*h>_wj2Pj=$Yy2HI7r_j2EvLx;&*g!#FXJ3zUSQ1(@m6j@ybb
zDuRM8m5eAmcJY`Ip1|RJXXU+%&9ltqWNQ$iAbdqgfZE`^3mADco)vAo$zjy*lv7P3
zZWAf_&81=f$}mZEseGR3>sZCP#CN-+m#aeq;;XBbeS}rC;1^33bFN-o*7KGiPSd<+
zd0)MmgX|*}+AXiHD1dghvolPM_C^3V;rEGtyC6tS<<G5p8xoz8t@O^l0vw*k*h~IZ
zQN;&l(a|gY-h`+hb;vr%sal742h$dqAo}{0JxV{Fl3<10#l<6*k*CB$*bVOFV#jwh
z<?b8t=gu?sR?*bMdivZR0ULi=IipHEGiXh~r)uaSQ_=5gCVV4rW+vX|2_K5^dLG`2
zvU2SATNg7X7e<mfZM+tAThReIdB**2X3-cb@#N4*0XhBK)?ImtZ%1652t<AA<Oyj*
zk2QFfl|zvJELvdh$o>rAvrVVL%`LvXaB@m_!}$vjg1y*fXjw|U?S01^dDWromRg!w
zX2LJ@K$37tkTb~ez;DGm(VZIBpn^BOs*F7dwsR(p6sP+upf~4C+Pi`&svY#el4HM(
zye6nrPXES6GO24hqM>Ew(S0hU<CJWVb8H-8i>My-aPNd=yr|?y{iGX!-_X8QttWjR
z_&r^g9Yxl6<<YaMTit0_J8uXcOHN8dJvS>)sqL@UG;ArImFq|yH#sSZwH8G0u7q;;
z(db&-9OON@R2Id}8Tn}<hkcu~odc@%cQo`bEo!)Gk38kOC7g0auSXuu-R#Zcz`=GN
zVg{3DS<2dkj0rGbI3-JnZM>#E7#)wJUNKU-eyxEK&34D>sL?+=XjDRd!%5myP657H
z#8@4sL|{4SU^*I`UX)=sNWm<senol~sRp_YwEMX;VO#uN@Y`zi5aBr?Mls+C4TwkP
zJf57$oTs>*eT$QpF!VXmNDnlK9AkBIoVN2zO%dRwc!PH3Hs^OsHgtMr8Ibh*Zs6OX
zi0nmKHU!QDUv`;V0N0E};J{r?UGFM&g^M}xEkhSchQHewr4O`@RrDm=*rUJ?O23;5
zvJm|I_C2^l`%`H1UXV~A03I@HWPdp<e=c+vxjad0CYrBCm2QJmqxg_}Edidz^T1v9
zn_WW~QbY~53h|7fblk)5v0iIw3bmLI(R__s=n#2Sjzsj6-e}JU#xt<;vTuwXSHifD
zFMq^y5_8etsqH7i&w09qNRSp6%%OH3y-^|83WZw4&hzcl_KIMvSoazDJuEVA63e}x
zyubq!*{m+?7IW;&Jy3{&j@-)oe~IT6G+<0za%|cvmRn9s>Z1&#j$3zbb&eS%2;gyx
zJLu`4i+S_5K{Zjzm83_aa5N<<gf;pHzeikWlyhlpkAaVB{Yz3gcP0GhNbj@6&`Olb
z2IQQ|hm+*gTSm?Y4+&A)!fz&uOWYA*j6}@Mi=lD5DgvtM%ni(j%0pk5s{>xOyE&h<
ziK9Z`BY#Bk$lAF%2yIc_jCPuHKA4_HeeW;*_A}wNj|sF0W0ky92ZMTZH{ORm7FOu(
zqnlk?5M|Cm-Y-K#h_+XFD<u3j0c-df?v{}61JX{}0gq;TV8JhHKG*rq_kqi!-?FZ;
zxx<iy625AYA$c)Z&Re1t(;QW{A~nZ`fYi0-yf`_k48UwxoT2%W)`JXS_xvy%@&*L-
zZI|3vPhY_|2JDfgum+VC$NdBLka)j50WG>*LEril30Z`7N5X6Qgbk1IoHU(=THfp2
zu@?XxvTcCKfaOchjy<}Tck0RUKvBKB+nL_--kOn<+j7n|f2Ms`Ley=Y)EAOPWqlye
zw3^8_aOGPLeW*gX!IZAA92ysBn5PV4S16%t75Ce}Ey#5;VUS?L&O_lvU7;8U%)ZcP
ztaOAV{GR=xqu<SQDsf<>KKbR;MP98oC>ZYt6H-bx=MP$<=W^DXfBuGkU*_iQ?>=|s
z^xjW_>hH2k(9JPAbZl>=N^b3k$hzCyRjluB!h@203wq7^&?^BHr+-Sxp<l#NeK35(
znm6RdJKGuLw><3}1L^V`t8I_CeheokR92$p>zg%ALX5Haas)#M_{b$edF38Ej3E^;
zSmIO}R*GK$BmZV4XY>s7H-JbG>D(DTdjlR*ZRY1TYTr3D!q67c>Yp0GfXFm-Mq;`a
zw21*}X~fX9xx_~T6|gODDqJ<Dl^q2ii)+DGTv22k*c@*94wsCZXFQc`VG*njzELoU
zhU5}jBxp-E^A1*yg2a_%pYLxb@AFrO#(u9_um=b1BxF#7G8UD8;7;44qnu)qXu)W=
zT2zEDXC4h5%4z5qSC#(3-Krg36~dmUM2xk%36Y+q7Vx`d%|s0G^Iau7sTQ33$mb52
zMRaibMn$Fu6tQtFeQGPpgHMr^4U5j($h7>)AlAJxWH&bQHX#-394%4*cZEkPe@j*O
z{IZd=QM!XzfNE%>M_sX?b;Va2#q4HjLV9`QX{XB(ele%3^eWzjhg{}V4sW6R159&-
zHi{(CoRS|XX}9UF{c5R-fKTi0+DE8j=0x82h(XwgHB52>Mt%GZvgA{hKdm|pH3*W1
zH<AW}O8V3Uhg!B*Ue2NbyImaULE!>^F>Y|1OD30q-+D1zaN;)9gx}`hN#p|GItPK?
zceOBBv0c~^{Z!BThy=)hmNa4-mwHd0@4RblS|`!@*g5d~jCbdR7uv)hJWR+YbGt$<
z4|t0oO=%zAzQ=~HBuho<fu)WNF!rg|Ji0S<oq!1$-6{AUvxqo0A3xbmvClqHAsjOA
z+h+J~ItlZXl`b@`eQ%u%z9I0ceEn1xX`yoid}ufkeTX$AC&(vaYYU1o#>$yAI%^aU
zBKm#vl!#yWXoP2XPVoJQnaC#Z)bHY<m~MCCQtZl$A8#9b%^IU6DCdpcfGNnbhRv~O
zSK!X)YGnc)_^nmf>(&j&1i{GRXqBU5b>_TZ+;V*M7&VwR!0)wa!{v%Ta*T}vv0_S!
z+%0b)8Iu;F%ghjYONx4ExqYj5ejy=R_2#NL+K(rXM)HV3mORYNm;=ULN~fea^ZoF5
zs_j}rWVgmUNw`O+sRC2n;%CB(t_r0M-nPNH<e`_y6~Z@EgdI{f!q0>cIJDZ4Fe0m#
zlZ>(<Af>tmb@B#lZ=heiIJ4Rn@)w7Rbeog&c*Wj4ShC^TlkVt$9f)KH7yG7UEZ|!S
zya!(7qw^Zx>w*gvZ*B|Gxp71VH)|KB$H;3#YsK4Ii9+Sku$7zAQYZ_2x98Z3izofo
z(Aom-2PZ4J&5-&_cxivVB;}l21(t%!Dfu43)4TyA=@Oc@G`!~6_nvCqg!gPj5AuMb
zrB0J&UFV2pSn4ZZj-@nQEUAcVi%_lysRVy}z~`zO8tLO3?HhyEJZgFhJ`rvCU=A|b
zxb%!==*{Eh)<*s9L`zO&Pz^P|uYDVf3z1qbSyh4@JJC%Wt%A?Bq6KuItU&LFcOhkc
zSA!XA<mLMfuGoNxxphWpF{I=YUKD{8g1zbA9DCiOQzGRuzM*#|lVHoc6Z&_+Lw$~X
zN=J)!W#AV-dm=A52vN|%?z~jz_SI~h6Csec2E<+85acuc#-WMf@9QKHEQd&?mG}1a
zfU)^;+jI|`Qo-*e&g{SY9~ui2)$S!9D_)NkpY*AIUqa%g_0JDHD8(2zE3Jrur;;q1
z$5m=NK^RijdAj!-pclE^up(3CUI{nJ3&Z7lACmULZ%O95F=DDg&zLM|*FQKecOwse
zJ8n9;w|szu`U>p3Kjorf^3&9(lRf;BkM;ZxUKS-LL3CorGI{Q4G}9)POluyz-|p5r
z5Af6T3X#TeVE}pWk_W@Qx9WEp_JIM#?034ua)gAw;I}m?A*E6y`kGb2pA+wIO$_xG
z;5}}Mv{sGn=tLG&Oy8we)+pfvpQ^F>eD@W%ntnGzS_>Q08doD~He;{l4y-)o3*<+M
zZpt2|$BCbJA;+aJ!wrRnDKaFL#bsWsS9V}*suZie3y8!bFf<Ks1V5g_R8#3F5eH^l
zEpV5Mk%h`i&7;YAlpV$=p&K@#iqYSzwc_sRZ3L!6SwYf!G3s}f_np4jtdBl;{|xlt
zNlR2>uecK=+`Gm`XEIfmC#Lnvl&QScLt6)!<gHJMqVxxopzh8a%^aC3&~)`3b>|y!
zbJA41Za!uSV-W3r;7|M#_+nbF--edRJo^3TI%ke2-pF}S^_^jWaYI%Z_-HpPZr_d^
z(`+*6vQ<~>GRzJ5fkmez#B@Lfd=Zb+`%6kUFK0odRU9*s63an01Rdqw#OLY>?_+&L
z$~P?!K5AH2WvWF_2I!Fq&X9z<*Y?>-=?rIUhNfw>zGv^46-kfi4wy}YKE}iLF@YYe
zT}+bc@%^8Y7bkD9<SJ|vMd03157$?%NbNshx~Z#038Eo>iq+CA82cs{1*HiSVX$@P
z%UwoLAc4t}`HV<f=f%(tbQGV!Kl;uHBdYp_v=m62u(L+<)sTnoIc+3@t8Nr^p7tG$
z7nS#XMA12uZ9xcjh|pIOT_gp933R^sv9>I4%jnha0n_w!b`;Zsk_3FW=u%aJo+p0<
zn!n9;=3YDBIEVDv&2=8T6@R3GG1`pXQ_2rcSv;WTtczTE11ZvX^rJ<;Yte-}$y}G1
z$*56Z1Vv+b`-fb)o{6|i0hys`!uVi_0NOlJ0ce3@^LMmBy)S-soHU1w#jA2Yd?-&I
zjcZ3$41vvNNq8dpWwEFF&6%uvSA;_*GuBVYSZp?w1gDVv9gTP5i289<h^FKW;db-+
z#_};w_(mK~RFkOsegfntySW6A>wCgfAR^-E!NCQliCQl@;mdcQ=m)7IzE}I?1nK*6
zmU*ON0qAPgL+`cnfufLp_)EU_NZ}ZU{xyd)=FkD-9h-V-os*`{?M*Q7m5vo}&D&JQ
zi;5OD`e4>->i6W>4b=mM0kkCM!FaBCzgKGXC4nLYpe@pPFHO9ru);>Th?N$$=7Ql`
zySa%()d+Z?o69AZGiEG(llZ}bBV^(a9O{ly8nTAQ>Vw}0XE?2WUdYz+cF_1YT&2VH
zG6#99qwK#ZJ+|n~1;)cb3-ex5i86eM@3e@ZVH`^$?5`3_GbEn!!RS3K%Wag<x(9w&
zSQRj&=E0<eZnGd0))Lz1v2-9hQw<~0*`2AKO%1(=8g%-JLJm6$eK&BXp9$|y?RHrz
zH`WRo>(+Fp8miU$V9cRCND0Sl!TZ9$5RYJ;MCIGtqwWzRtyBFC%W|=>GxAzUANt^h
zS$j+PZH#UGh2O=>OQfAL^25A~J>lJ!^TX89zFBV~ik`%Uns*ffz~~NrMsyhgzXxZz
zF|p<fN4uSDxl~B|LJO8V=~yVZMJrqIRu`q&zGbGQ>y+#;WL_nck|GQI!t|l%dastj
z0^fJ(Td=6E5jj<0lt;DV4D5yS=r%<Z-l79XMw+89)e5C+;bH7N`t5hDu|TB0a*)n+
zoicyRSzJ5>wTWrjmg#oL!%qlmy2(qPRl%+XCL!Qu0i*i7cq}Fq+g2+Y1iGtriD@Nm
z-?8XHPdD+jpxRGl!$Qd9+TSKnDPa<U?M(B}txiYJf;4;0OF6f==JWPX`8;iX^@mLr
z?+gw<Ff0h=lM!K$IV&ZE=@!X9;>Mlu+nSpS+~Km;fH8t}hPXuLW4h+&L@<q`@!YWY
z;^DGD>J6o0VD{@uh`d;#$2JG0YzgP{B+N6vqhE|m2Mjfzno1tPFUB11Oop4B(?cU_
zH>c8R84mbu%puYeh-Arks-ml*=|d^efj8Acsi7&f!uL-2qFAlXPfh)yi7NIgp2%AR
z*T`v;|Gn0Ff>7#H;%+Y<6SogiaG;eZ&gxQcA7Y*pFl`8?o0SY<k+YuD@~u2far@wS
zQ8;qyzV!0J@cLAa@9#E!M|T{PJ1>$HmR7=(+wIKE4xZiA7M~n4KTQj7QNc@*tE!p#
zt)jQ@3};{IXO6dy-oDf9+4X+hnOFX97(6<aue0(r#am%yk~=ebY!h-;WoS`)N)-Bm
z+bNzNHZb3%Z^V3tbN6Yq0(@jj&(UZp<k6=~z_BZxTJmBDL(I<{x$sf*t4>=WPwNL^
zL7gtwRONQU!CUZPEs@OVwN<iEOyDA$(L(oU6>)wPKh!giOBSF9OI3$ABh=0oy@6XH
zBO;$C2X3<Qp_=k`Q(oAYEcvRl!VRKTj?nR((#s5KYvwFE4bk=|c3#x*z-OqMy>der
z0(h+JZ0Pjk1-6k#5Q}rADCB~k(KWCHUrNne^nM49UIahuo5lOA$1JGt+`JjI8M%hm
z_;P$Oz6e@3aeiJo=Yu<d_2b^0zoT!HG$X05oiWc6NgXi{HsUWv@tU`wbv2!m9K?-O
zBQQjD)uM2jccFGExY8-vS+!^Y0!GPEV;>3;E8vD4>U+enQ6AAr1N<i2yiUl@H{dSG
zCMOD0RTuaLjoTgZGmHF-zF3IuZoa)C*aM8fuQ@?06atr8|Da;<h|H1C9z@rp1RLeK
zARB<LSuKvhy4rJOh<GtFNs7E9Y&N%hDBk_3-_huJ>Wt&4o5JvJs<&`7y%l+*-Yc$f
zc}k*h=Ct2?!e(NL!W^0~!0&n3*Ywk%f-ZKU7~|S)IJ=R7FI#XOysE?0)zrJ+v`&y^
z_ooDy)VX&zZ;kSr0DKj_EY|x`{TWzD`jogBBwcR~Qn&PY$pfn_8*mfDSCyU;twEp?
z3ld9r7k^1Wk<9a!pzrAXUbG#`JaEZPH-sL6<$-Js{~N9IOFOaQ4-C#SiQ*3IRO?_V
z$QvTdTc<tf^3O{YrO8K^WZ5r^83zG4{XhkGOt4*TJLQ`Qs)fO4hJ~1Sm3>oWTgwWy
zlGUV2ITf-Xa1=9vHq`JG)2BE^E*C9yklvjPo+Yjl7WnN>Pr_ewVtcV&pU0ZBZ~qX&
z3&3zow=-ikwA2?`+3f#=@sacBaKChHFX2Eccq<1&3GBrePOvYGmF5f~h4-<JmzRdT
zqghYsosWc3ws(eeu@sDi>DB2DCVM!wEn4iPVZ5fpl$_wZPu+=6^oCYm48nP43%{>g
zyfoT7lbh%Etr7-*z!0dTS~IHmhroAC>lCotba1^OKGkYrSQx}N$9}I+T4KF$J|+Sm
z3U_<EF_hpCFZfVWZ(K5p4XOZevz#Il_U~D10lx*Ev3Y-IGE9J>1v>i2WFapHf7GHE
zReY|T;D91Ff`SztWQ+E>U9I5QlA5{bOx%%q?<IuBL1Dmn13`v1FNRpwl!`MV$OJv{
zcC?ZyIzA0mv8<G9W|+V&+1r)nt<w$pMo_&!Te8)`-e2UEu0);mA&Oc)@a}R5<EKEm
zD`BdWjwRE%ZmFF@E?j&bTd17^qbKMMMCb-D_LYkx7~=g+d|^phy}hedYgx-$zzs1@
zRX$c9A*cb|<;h3IcbC{H0JC?F#UdM;7aH8Rn?sB;cgaCI`RE$olk-h*kn?qd7_s=B
zFd0t8A{Q@+oZzJ6A+`!TWUiKUKDfo?aNag*<{d$h*C|Wn#N%y5So;u?Q8y-yk~4X+
z^9bn&L}|XGuWZ<8Hy<tZ8MN`G)^2L-I6DVHj?wvAEIz(dq7mc?LaY96?3p!YLOV#J
z@AlCi$LH|{1~tc^+UA3qP+a>WZ;kgAPjE>`9N-srIv@H1bent&CertYqvdUta$GTo
z7=rB1F{tBvj9x}aR{<kYtf*tvv_jRD(O026@;<#g=f~RHwa8Iej@Lu_R?7mv_awCt
z7<0((;I`)T3L&a+N`;jpBUWDUt!oeejt_o|lC@|Xt+9EP?{k1R7hjss8HQ<9OqcZ|
z%-%=dsX63n=2b2@qWaPmxp3)I81x5+p5oFj$$Zp2`u$GViQEO3Ew7S4w3aVKB=30H
zgOJ+NMy{%r+~`{mx{1<brt&OsU)S{HMOWKScy$eJnaJ;!kbQE-7HNi-)4tLPJu;ro
zr4m|IksBEA%_!eff+vLCxlrEpCDI3ZO_Qvo*lM86`TF<TksJdhC9b?%$)82p>0&Z8
z-RwKiW~tIn+g$&^L-kqY+PD732MG(1Xr+hxQSxMOkAw4=0-_VNU0_uh^nq#Nr0As^
zb1#YQ&L~Z)I&kIU)uq8V^O`3;V#{^eJAOw`>Q|;FpGaAX2SC}#rTY@aHJ*q*mnA4(
zydHA#>=9m&W->I6jWsX*9%Z^DGj@8s&+v?-XG|RTUbFSVhXjlYpC^5t2?1g82FD~j
zotGNpF@spL6wXJTDEx%c7zx@~u_2sP2qy|1*WNiuLl@R-@C#~s4)T7}X`3w4mnUcP
z9)J^E;X9-B4%{WwC1Pw6`he*^O=pNYbL@#|M08b*-1Qz2Wooo+$(u3YAvTv-Qu8-&
z1nZV&@_yo=s`K3TerhyV)@Wh@UhGFqRBb`31U|}VJD-%hB2r~t=2xQ8<|7E_sL$Z}
zy1WNW9lj4MF2MJT=a`PfwVbhik9USdv>vMTPd)DITjCKzFV*kkC&8%ABt4Oe!bTF|
zRg9AHg*y8pV?%5x3tZlcolSP2i>^5}gN?|Zg0d42qT31vtG;KDGW7XO*DncJt!jma
zUV7+o>y=|Gq^b%9ltj5czrQQ(CCP3yIL6xY-u~UBw>RvOv{XBc(S`fA&i59-@U2&H
z(@dXuXUk6`<fIlRW02unPHrRSxGX!;+IK>Ft*@7Wcf@zy@5`d2`FNIKIH~h=cISL?
zMfPG%augjhH*C$c%2hw~kv$h%K>1R{(a0d1aZEOAB}^ZLPq}t1t9+hl>9^YcJ9_2O
z9cN;5V^P?j0S~m4P3Yw{&Qre6=P~bAo@Y1VTZ?nB)Wb)@OT~QGU}Y~F@VjOcgE4LK
zUWs?M4a5rx4d2tg^OP@=Jk1?S?<re3HsAdnJ=)0uP#D;6WMOaipmLWnT6RFgrZ;DT
zdhvBBceN-NNK05oYWiNU5DwWABP!N~po$njgIf)+cWrCMMobcPeVQ12X+<0XBgSY-
zDrtYmi7s@*xLabK_nUC{9y2~=H^)|i`I_lCqlw%#bjDV&LIpqJx9{1*+NWuPLx@E4
zZYW}{{he|lC9Iw4j3laT3PO$gJ-3lXQQw{y_{bI;JXL?EK&r;J9{9R7n5AOo0A#<M
zo5i^zo&*9{7aX)F%qgI(v)w;XdVK4}<3Bnbnd}nvP-#a@uR`xvo&JJpPuDPb&{sc&
zEA{a+N#gycJrAvDIJgNLI@C;<1xs{(>aAWO#Tz;_C98{J1=~P7n~6?<r6R&l6qrJw
z){5RE2;@9)l-+{)zD74E^6u?0eHHIle*($WVh~|Bu=o7;ycEm@eKd6ihiY8e8)KWz
zPfK#1l6{SJZr8dI?-6R#nZ2E2`!kdCHayPdiu(M)N9xl`!L0bg5c2d<Jjj4)k=KrP
zN9gmS>+&7_wr?H;12{)HQkCu3wA=f2Pe=k!SytOwMeD9Y*$pRwy%9ES30HYo>Ia@Z
zC@H|Cm=#+bRK)X?NIx`ZTWCF;3BSFUbEsUc*LNkm`*aQ!o5_|FY^OBmQ1{n6>PGFD
zNg)KxqdUFr4p3GJ!j_x|m;B1u*Pk$UwxwCOFfI-=JI8)I1tzF`&<(*ksW-3bS_!Lw
z<8jGa_2!Ge^d96X(Q>m-7^2EO2YG7@YAGzAFAn=|EUP1_SePhff2X8T+kzAQ+-OYK
zQ;k@F83VzKl=fLWPUk{7&fEHS!Q1?f_wokaT8m(tX3g$(qx3GviFs!>7DTYF!Z0l=
zx~*Fh9RZJdJZr=vzEj@!5NvlAyfrnxVM7LCZtr;i96$SDKX$r<svY}dX+*(=cMouh
zeUfNh%~VOM<nepaE_m=DW9LmFawbd~Zki-sPJA|Cs7PWJ3PUR`Cnq6rWf69XoTwY<
zN92iHO68DYF5(I6G2FrW28I&p{e>M{4b%N)itH>kYM;Pw)$DG5TC06yZ_4e<yr8+h
znY;j7W3w!t(esphqtEKB_X>X*3TDJxI^|n5>v!~d0g?x)7uU#uVM_EME@rE*UHO`Y
zBCtA*Cm7LDv4RPj+lqDsDQRpS4d-A*nL<pLGG@>SLN!qiGP^kS`<|cBujZqQW%-W2
zip0=8X})`7);yG~jwjb@iZB4<)ZllNBXV*RjUq=Ur2AUJ_yOA|6t%aU-5MAB2r`@N
zMj1fD>+HUw5jmLmz+};mjW~J~nyCSs!!6Ou^W6@1(Z|)`Z&83-)Mn&&Pn0wjc1`qk
zmL8R$EtpB(oKc4O5^oHgqXu<QX_6iF5BUq;Vlm*iU5ykf=`Tt(E^L46IAY#&TA8==
zZsgD%12I*8>P&LGHjJ}J!WEU0Jmc8WU{^n5XvyzSx-r?doEBn-z^B@|;(cRcAIzJ1
z6uU1;r=2`@FWEtGDJxkOz;e)F#lrV5C--8u>ntaxXgd!^w7?lu6pL@(j{CyLZj$aN
z-n>nY3~>h9iVSaMLam*P&t$y*+)3F3Ea}xB7q`S5khZIwKWC7LwsdNmDXi=*YQi`r
zSj89FKPR_=hjOyKm3N0vsEl?CqCncz?|O44==Cq>GvR+FjIZahV@cal3<K_vgv{6+
zA^Zi*cB`4(zUl`f&mD}?w}d#22)q{Yuo8;xyHgV~r3tS{`{zX7S5jw0BY*5~jQ22i
zXL#Se2}5;YByK|?cxGsYTw`xe+P*b)mK+<MaI0Iq2R|RQEpNXQ%``hQfD+Tg7eJ3T
z>cqY!FoxQ8uC%DylrWlo$K6Ob2(nuQDGh>V$c(a=9_g!7(hfI+?uW`zZQ()2do^<q
z)GdOT$}*j&eCs%LROOFnBmvZ>K6~Ud$I6|lDvmqzux)*(oX8A?#bFEQU0}N)Z+V@b
zZ{7>N9y3De0smmcb6m-)X)6MZl#O$!Z$T(|%5~HSzNBIr`rHz4AJ@#(x0puzf|}#1
z*RFRb!&*#Y;-{Gt`Yx*f;34^NI9I7Ns#a*#!pM=OOr4UzeEpV1@=wl!rX%Yx+}m@1
zjHdW@1rOb(*nM8ASHn}D=&jd)-#Lo+jdh0K+g8Iej%o@9Q-MaBPf_hFr-bsNwE9MK
zX0+fMvIBlYkSpgK(rVbU@>!i*M?+kl8MyGZ6&0WC<wq|@_pvIGtmN^xnS;QBc1lpZ
z@(L#nXv-zQnISe5rP2()TNh?IB489z@?aN<E^heHYocCXgp2nF(<#w{<{V~-NVw4H
z(1LFr9Z52rz1w05#NA%ZCE+5?JN6ea*FM-fu1boO;0<64`QFWNM5EX^6Mj2_+r3?S
zF;NbJpwe=NqMT^$0dAMYgRuF1q5OydzxDn)Tz#?SH}>0;;!mb%dWO!Jei7stqtq6y
zgiyO+6Q0nn-{0jR>uVasTN7i83w$N{q?G2a-_ax+hj(ab^5;wzB(3fZ@eUs+U`Rzb
zMBs)$S;!H?@Nt(MZaq0cHYQeLSP;H*kWEo)L~s?&qQFp5oB-u#&3PY~A;TZf6$=CK
zz?a@1cln(Mj25NJ8hE|!oXHEy<XdJl+LK39e(_9NXZs?!(w(lEAUN6M17bbm5-Ja~
zhoFno4$Op9SFO>UcVv0itsntct>@U<o;a&sw_`M!Jr1RF5Jqj05CUG+d6(VgePNXB
z{ti?0$g_cx{Fv^O2?sDxv6J;%j417)@7CB|E4$qCfv@zXGYKi{-w=lc-oZA-5XrkR
z%2Div6o}EpeDdHyL||?dCFw+|1K8XE;gq~{Ly1m32Yr3_6VU^Yj3m>_E8#=6LlG*r
zIdTRlkq4uu-SRHK(4sk;OJ(ln>}Usdkk?lYF<Ri;<%dQrDk&+$Z{axMyE|w9jy{-^
zd622^xZ%r|+!{2uggoz&4ul3-ZmnGbSN1uv9jVQ!4qPXRLX~KLADWv7zffp8^@}`9
zyrZ<HDM=D%jF++T5yn8wiG=%|aW#Ek#Bblx0u4MduW(7D#dZl@be?K$WH)hkj*XwZ
z)3Yxwh~f$`_e43DpIpa*zlS`(W9k)F97ksA#Y|hQMZcq2vYQvEF+)YkMt+FrlhSBm
zEr^ROS#N)2-<3PPCtJsHoXbW3G~Nwtohvm!DR8B{v4rDJaf(+ioC(Vj*G)35pPGZb
zUFtM17ZJCeGod2nu?dl9&SbO^(<WSxH=)Z%YkS_=dkjheM!(DnWeUXRdvUj(vhXL-
za2!pI4SN4c>ZV{zFsOm42UtMztoXh8PL|-Bt!NQP9=p<j7=i)O@B%zY9!fNHMgRed
z)Uso*TEid-n5SQzU^m>JWFFbmH;q*=kL3k%GV?@18nCl}wOZ4--;KoUXI3ZkYyQ&I
zDoc&rY`Ia)1YUG7C}V-bcu)A;yF)l86zJveh=5fp3+zN?j^qB76+C)ix-m5C7wbo^
z0dv8(H7d@9%+y`R<Ml;W+nn5)Id_iUt9$H@5Kj&Q{m2a7+IB>!!1QZ|)?fPd945PD
z2qqoqnPa>S(zZTu<(JPf$e21s+|CTEc3>JTOr^nj=_GuSeNfEvwr4X3S(Jy|EBFbi
z<Dd+A%Vncr%mB6%PHp=$tpGj}Y*rR+M+HZ2zvo!Kc=QudF)-Q)Zii|+qbayqQ|4w%
zHWVZ0aiKSx1;aite*+0uST?Gql_KU$<Aehl-1vA0A!kXlPwbgY=;4h6<H2p!7N-gL
z$oNT|nH%iF;~cG9^0C91c{IyU^RjtvFh7B>8pD)DXkkNxuTrCp$hUO_2LKo<q=f|P
z$*(p%toILYIaoWSeGk$>H7wH55Sky{_O^7v3;zGWZy_|X_x;Wp8KZ=@V}kcRQzP%-
zLJ`aD%(r~kP7k{q^609J1&i>YGc+4%(yMvxXK1mfI9h_7b%<}Jx54X6&CLh17gO&b
zJ2+O99WYfgCtDFJJAu9@q?+7{NO-)zP)wk6XpO5kwxbe<m07Q7z%z7qQMRtUE`nzZ
zS_v?<Oq&FdR?CSjMEjQOFj7Gb_<bacC8e-T%By+r%9*hkWwI4`O0Zec@`}_f@V@dl
zLvdvU`THQ&GKmOzTwjQ@a%^_u$Gw5@n!cmYTro&JkxD;sjwtu}y*{!12j*4W!Jx_j
z9d=Qxk;tXF$*&poB5mW<lrW<m>=5<Vusy4Yi|>Tr;%*^eZeZLh?y`)tRGXTy?*=~P
zP<5+n>VC`qXdI_!WyX<X(^ptrug0JJBs|2tcj+Kl8^>9y64d+|>A@ooEcFE?y;`K-
z@U82Wv02u}&E*x+n6wrM9Y-?254&PMO4<$fx@+E(eT$sT5Q06CS9$PZNwkpN8o@He
z>D3kzTOGm01ee1mwIAyjNC37aOxL{ogKrB%pMw_qx|SZm;x0bSFo*iG6mX&bJ_4pE
z)r#9R7WG8$icn(A%qbm2&$~b4Z*>g;=mT%4%M48^o#?3tKEZN@(9v_`DMLJQG15RC
z<Hfp~z0i95A_0~rK8|+5E_qXFPP^R#0h@I@uMk~NCq~~lYJFp$BR5p{T$gVU3@sQ$
ziqjD{vYTzqso0L^QcL*GVOlMH(QZXKun?>4!&n7QzLNBtQ5AGns)mj-@SOBz%0s6`
zzu7_i;K)veRCfxvANOUNR{MV25FZ9c9#>y?G>c-*<pV(m&#gb=b_YCa%%p@`$Dy28
z!@mUvQb`yQM~mvd#XJbs9L-~)m;G&DLx7W`Wn*JH_WTV3UCKJSh%0{Ah*;a|u_p>T
z(UH;hwP6|Yf8K0_SZnvqo{=oY541rUbeVU~JmHJ2s|5+La`m%tq3d*wEYF_!d4b<d
zBRxyCt_MuK*1au2ZQvyQ_Ihr>hh66#FEE~`R>|%7dIyYB@f+mG!!*8zz-!$Tsf7~?
zcfdTt-bBE8%t1QKgxH_&MTj;W;EfcCxfC;cTrX{Bi&LsLG9#7(eplwq>KvgQ4!W-<
za%5b-^G>lx3~AX~qgesCcij>>7Q_m5VHbB<oY6EkgI^9TuWD5!dsJwzT6(P$M}zas
z6Ouoy%6yQ!{O;-=%iB9R0XgR_>J3qpzeGmy+BA!FIh_;mzTFhrhsLHEL*O#q(}>r!
zSGdbpu>}Q)X!%1sbUsHT^ZN0<<gL%W;S|*Jr={31*=AXVChi<ES~D!tpoBS+1)eej
zA=5XAM-twLFpGYQ_Z-(88p{SBjZ@ypvC(J0u^Wtc@ygd+Lu*|QVLN<vuS9_f80a|4
zGs+x#TU1|GgofV_bgzW~_6c*`3zev95IWJarwbUClhys(%pb9e4OznrU`2cNg{3#T
z$*ibO>>L|<qlvQg`S^Q`)8^i?36rSpaAa^{0~MOR;!SrPab*LYHeqyX-hkPLT4tcY
z8fXFcA@zAt)^Z1Ur80Y+{Jg9eE8xK=LV1hT%W}AWWsTh;or!P933%c$;^l%i|D51E
z0_QI=G<vkI0iWK3P5u7)<|!pKbKRwSLiPL7GPX1}Z7DT$;efIA5o=Wm<Ae{^F@EOh
zrpr6HUrfJlbmvCXIY)0fei4kkz_R33Uc6S!wncX{?;r{3JKfE<%MU(no9$Iw8qnS@
ztz=6qzdeqF(<?4*8~LWjhU_8{KK&uoQrZ(8Z8%Ap)qyMTx7w`X-N(oQNNjB@?dA3l
zbkORmPiA^sU)s}-)#-0*`AFh{c(#*w(A45Ll>hoQpEDFb4bMtQsg&qDpyoxtM`GM)
z`%S|*;1}39MJW~SoXB!2Uq%@#9|ZKlkpMA~Dh-i<kr#7Pf<EXx<wnED_VI67o;mOL
zM#N0067Ga?HEm8|lo?fg;67wph8Xh-HWo0AI!;Dn8RrQiw^XYo0_vJQ9vehew-Una
zYFljy*M=#xL%HaD2c4oVi>-Yl+`4qL?GdTpiH={wOCnZV-m0N&0(`bi)%c){Tmj^k
zjVYK}()op@_iZ<Oa*r3z)I6I02`5YI-mw=09-Hiu8ka_NbX<5mrJMPR7H_PI$bPSD
z8tQvVW5;=ier_wyA{TxVJ$${ae$^vI@L+^!oD`j(zBll-vo|U2w8pu#zugn+uj7R3
z5Aay<Ow?j0Rn~ytRGVlTZw1>qy^9CLu$E%>LSIP?Hk+KjwK}-o9Q{@iYZ&a3p7k6}
z$m^ob%tq!1zei-7JbPg#-h`K=I=4H#0>rA_^O$Ml47%lr${QK1hD&g+%rzP3(f`0s
zI~)n$1V(|{assUUD~BCynB=dREdvjXg5x23oCjjb;#}*wSxdg^4{4eJE*J5tAjVyf
z$*u3$-x%~wT1|GM&5~1lJNgnD1NFq>NW0qUA-LK^MUR8&_gK~`==Zzi!{)&zX3568
z<UvjiJ=lA`%OjsTtDAsl$XH%7Ye&<VcIN8|!xwSV;tk?l`Bj%sZ6U+fL#z&XtI3az
zE<K1j_Is;(WJSe4N%(zd%s{2TFvr0C)2?Z(3pEe!RX%o0PTEm)fAD(^N+d=1$fX(<
zQIWG5;YAHahzHN6-W7F&w4ks6zs1rDue8<0$AOA&mw%fY+<Okay5eSO-nPhS0w2ko
z)BC!@WAw5Dl(0LyzraaCQq!!|_qs$~@-E%^>7p!BvmZ>F+A`DF6Xw{5CP{CFO3lRp
zGE&zm#a7D5lh&~{ZKXy^Ij#{h5}r>%`3Qv^V7`VuG#YlS2GnzUar$$8y6+e6rCcHv
zOz!9&0bZ69ozfrj=Jk?rTe4~Ea%PJcpQ!bA3k1-2+Hxj5T^0zKbO+}kXiiV7=JyeD
zDHlbRTP?4;$m_0zVXT{(R&o3ZSpC2DjHiV&Px;mg8M0}qjEw6gNpob#HYqFjkz>Cn
zzFELzZjrbROOe(Zk0Znp?p8*}S^yM1Y936q=}$x1AtiKitF81gsxP1f?GRWsHIqt1
zlK;{wc8Z&;ky^chl3ot9DPJ7$sj@+%G9edc!~F%u?yl~ifIlz>!_g);qZ#HN<((}_
zz^Whnp~$W~tT;RLaDfr<vpX76206%VN=HkAYDcaVc)N95DTcjrY+TX1g`4lp%_Q}@
z9eqQr!sUZ!Kh`yPv0sL?a^&S3ZU$-))-Ig+T_Y)a{><bph1*F{ght}aL37$JRzX-Z
z2_Ba~pl2@@Qe`7y(tFOVVn>=Yd3XK<kG0auoiI;{#aGBg=EXxxiSDS9>-IG#Vf&}d
zO(_wbI1?h9=H3K)QO8NmM3j)<1AecSoz&oM#I^AAt!@e`f(h~L@C=~Tp_^T^RRQ=d
zGL}1_?PTZJji~DcjbP-#?=Ppbs#19x2~&lzyxM!&zl(Vv*%bzGgpWCq0o&_@JW2pu
zK%>7HtzW(+ZALcu?l3V1o;PgQU{}WW7&C*$`nx2E8*~les}@w^*PG3=mjuMwPUIp+
zb93yyn|NZZ{WYo@55|)GB;h60voGYO*@CVHBj<w=^}5mJ9-{8R#~d`i_|wbcdo?7A
z>IvR$r3ay1orU}e?*Y|MCU{Lp7%#x^4jZ(p_XQ3ZL*%iQ)SaUgI97SSDj`C3#-kh`
zC}M3Z3_on5B6QYnKezYrS~CF{#$J#8-tOPgv~8H$i)xKNdcXsDO-F2%%6!6o+mrUK
z+?K-v#Zkwq39SzIgbK#A$egzkJ2T)HxmGLAE<6Hq=y#K3n#N0dez$4a7+ndhFbDyr
zG`f_sd7s(ygYgZse*W96f2U*@=N$HXZp^U}{9^89gAi2wX258$9XDx~>7CCgu!Tx-
zGMMI&I}*n7#Y(iW+it?Bj+<zQ%lirV1)ekH5%9yQ3BQGNX({T^Tk@2|m`jD>zGBx0
z%*J5_d#sLckT=!vSZisW4lD~}9o>-nZ3!<VLv+d$CWeAhG}HTwyTZ()4><yR&lhM#
z0pBTtt(IrE4G29bs|7>Y^~ARVQ|(&zL3=NIZ#!Ws1;hR5&0K216s8qoAWC~zz!eFF
zVbMOP(}lkH??i%;41U7M2v`j1J@n@UjNA1I^Ng?p0q#w?D;S@MZ{$QMid*o2wXJ`!
z4z?Vln>0X&^k>D8on~o(PD~}HsmI%iFkra9Yx<jK$xcvuJ}r#^*)GAeTZ*HWJ@a&L
zQ9W*o$N7Bc+%7Ad$rStM?*geu#%$_7#Q0M%U1Ve~w0_>@%C}%R5#AX=BwPWrX>gDK
z+b_>}*;Q_oel`NYB&`BQCuW5vOQ@0yexZc3w)B$58b!O<T2^m`>iiwuY4LM|FK$vT
z;TO@HhT2~I^IZ;1+8Vm<%?(AsuoadYq_Z7ZFmI<`vG)*`xPx6zbKZ!)hj4v^@Yp(U
zdtQXG6D3V;^=6j(9`Hs))j6A6XUmJ^yJ^u`+`@Lk?}4i$1QOlWDHM~}D3^-E6fm_C
zN1<oAqge1|HFS&6_D2j@wCoT!8RHupqQ-I-rbh$VDCk^InTJVG$-5)Loy9#mxb2%Q
zcp|{<F8ol-1*UKuS_7r+cxm9boS4Z%-J&IYQM<CxA5NzML!h?c7$WN5Aefsu3Ek6!
zS@1n!2lj%~xFthgEM~*1^fwcpby;L}qZ^oK>~?8TpJmBUD%_tMc4d{O!#tX%E(;T&
zvNiISz;y8-mG?7`hPQl!vNqdyyH}MSj`U5pSyF<jZ$<N9@87%^ssT4d$nVYxvfFCZ
zQD0P4XI}fp%Esc3=s@Hzd9e9X-{1T9?;4TZ3R5wvO*xONO%*K`@2b9geY+Tu(!x6X
zo89HzH(|fuk-zSHD4B*ng+a)8dq!3N+(M65$4>NJ)NmoY4)YO#s|YBPr5G*(2;KD1
zviG}UNE_w%bY@%r%ZcI?;+`T@(|f}BXqbrt?KffQWy5q9jFfVa&f1Ye{AK1m`bAtl
z3c~rIPHDM2mrH+2$)iyN@J!0vnS2*RZ-obW|Gq}f1SXVi)~<Em1AdFLbI??KXTl55
zRfS9Bk!Q|(;k8uVV4;x0Fjgf_2ounZ{0*>q2(I&6QCbcH`6`Y^WH*n-CUEF^A|5^8
z(RYTw8IndlPkH%Ky;&UI+zX}AI-Rt8JR(b$_GFcZR7K)3<0IO_J!~NH-V&y4Yf(yQ
zek<VqZcj6*9t*M@yQ9BjUz{G5Ip7z+$USVRKenEP=PXWSuTf{sL6AlZ3ND>Xc{EZD
zman~U{i!)NblvL>R&G=TFIg3^BSHQruX19qc5E~=N6)K7xBo^Ny0n&qyuB<rX0=wG
z6F()inSHF7`+l&Qw**-jdx~KYF>;JkBw7)D@;A?T*bfbpwYHg4g69YevN6_(c$f-1
zY;@hg6`fszS2;fVJs7p*{MgaC<cy$PME#KIM;)=#I~fl{dMew7WvEKUDYvYHoSuI1
z*@}30!`up=H6%^%2^dzO{lk#e%%K~)AlA0*=Q|ditUWEH7i4t7lKO<T)bcr_oo^R2
zY00UKoLpZ@^IPKNM9e^{#D&L7MmaD0{j_AX;lscf%(-Fw=5KwNvI4FqK?i>zcFQ5?
zU<oSjm@_A_$@$n4jEYAdys1<>hcv^>2p9$8QiHve$G&@5ow&XvFY+c6!t9{z-Fp*;
zW9j5~P35^!sL{7+K5Dr^x)w|enX5;{QX}9O_6_@LmiL&{nJ^Fjc9erfal&*fh2zF9
zBgdvRP~3p{w(KpBes@nZfgrhV{49{JTFm5d1v!&G#G}@>{I+<{X>@MId2Y17W9dUG
ze1?xE@zcq9!z8sJFV>1BXMz=~KZGdXeuI#YXd%9G^I7MsNTpBP_|w+Nqn9I0_0;@^
zes5G*f-#SkW1KdY;TOuqV(Y}TB^(A{ksO&sh8@dJS!LuHc%3v^bl-V%+e^#dmhBmp
zsc38xe8#Y4*__Ze1z*xA#hhEn|M4cFLN2c&yF=$j3<8+i-l1uXkW>OD72`DWvAsgT
zV^pgzAhcD!qhAnhi?hYfB*$(|cVgLrdE_8C%!VOa87l&6deFAUZhlHm!uYqF3bb0E
zIS3vcR?}TpBnX(|zQVy5Sd~Pv1h5>!p5f6U_-)bVbP>7DoW~1Hte5c~xO?&vmvZuv
z>m3mzUR5E@u!J$6lj9~htxgWU<oRB3$u!ir)>lsMt!sX68qMuFFf8Wq2@V?*p=fx`
zkmo8lZ}eH+(j-f|{fQnqpi5e7mT%-Mc{NpD)~tD(FWe=|#t?<+N@|fgh12Ji){T`g
zvMWvrK~y>k?<IS(qpyb_j|vN|g(DjAi}zc=@3}hFVO5Ch3!XE~%&sX=0!chtr3#JA
zJQlUY8BVvB2@^{Gt%grP8##>^q75hD+X@p6;J36hJ2jHe$Y%gAr4J;HbF730OznRL
zGZI((gpuXcdef2~2s}}0Y3|>?=Aa3F?_Y6Y4W>zs4Hvz)FjQGPVZ6?r#qYRm<yF!N
z+@ndo*_&MOn{CQV+;~ud30ll+60;=W_eqi+npj$Hb`d;>tk&3uN0bEmzF}4-!KKV!
zluC~wWMKEt9V^TUMvkVtpDhHV1VeEPQJc(bR`!4tSyWGbZ;w9)!06t%@`t8(4;Y;%
zGn+Sk5;@3=k?Kt4&g+y=Md+$v!KOsuj+Y6TW=8M#N9cA5eoL|w;!^C+qiG3nR-K5m
zelUxs<G3|?&hLFgt~tqzA1}~NckpnEu|y<Nt2(Zuy{Va6_Lif!jPtqB*PDcCptXsZ
zw?qVr*RkoBuC8qE2=psVu}JWsiEwj-UYi`j=${Jn6~VW`NE78R+ya2#OCN4%sAj|i
z3|nxnU^~B`^W@mPZYLUxwyQjX@3~g`g@tr;5NeFhS!4WC@@U+sS9L8lI=FyIkyg`&
z-o0yi%J=(EPf*rbE#SAwAXDvPuM1*=-#T;j;voVw@@UM9m)(EfANosQAq$OdJ~4g9
zUdDl6m|VSqQfF?$^MrdEwVk+C#G<uD;(M7|F0}6(vcMH1+JxU8yK8;~L^H>DTZC;+
z7n{}k!9ymMEKwvqb8H$<ESbw4xgY#~?iIwrZW9Gy(Q&Pfi`GSsEpJ&%QZ<=%2i&E_
zBo#R6N(Mal2qj&j?K~yrYO|y9x9)jg@yc^LeCj>~{NA7K`ZCf}J!e90ruq#->>`hb
z^1c+fP>}iNty<1C>v!(>GO-<9atlNDf_|5TUtD79Fof~%XMkD~xptJSeKZWPUY&^M
z0^*&dQSa)MvWtvoO7`f>8SXp8sR|f_Bol;oxv^M63~O=`n^HU4c}8l6R$8E^B;HDh
zVa>&S^7-+_qEU}L)qmEDG#N*5&&H=N%Y#7QR>z!TMDBKufu~b-wXKEmwBXSkn(-~#
zIr6}&A=)QN|0XOcE@ih)lyoXzy2%My8|rS~{UG}h-CEUxmqOn?s?*X5BdSHfF7dKr
z;8Bq`4-J=F-P>@ja4dMW5*?=YmJIl?bJbU@hnpENJ7^aJaJw0KpI-DOyEQ(n<HpiJ
z_MpmFP3}`W?i2JQy#DlIZ{c_&7~-@G4P!=aq@tI8@^biTTEdHZjCLcv#nsS3K9&>T
zd)L^B0T03`%O}r#a3*?x)kjPFFz45Cf!eMwm--Ddibj+qe(g1<J>EWcg-hf-H>XYS
zg5@sYft*BN#9DW1AGMqat$F5Z<IOm4!nBOK6(4J>JR{Lz-CkPP>UlKXvQ_~`=&OA2
z3+J3BM-px2Ad40?cP^5mPxu8u)z)g#*h~0GKGe-M+a2?Q-@HtmUV8NcrsdIjeAG@>
zUgdRpL2|a&5|o0^lJ}d0FIT(@Bmg9NqQJLjtb<(oTGqy(*DfA48s%@;qP3lMR^YMZ
zf2KeyCIVYc9dr_tY?0}$O5qeM2`}6~R=`vXoF*raUOpJbI~zB!kmMkv%C*R%K9QNP
zvb760H*&!}=Qrtmv6N1`)*JzbWM!6h?Yrg~Uo2gvly>=Y<_Q<Z(XQ~lONqFKZ%MP<
zG?HEMJEB)OP3M2X+jo-=B7OFRF2~|Cz$j+>b4ujZ6U97PaTsC_HLnj}>=DcADDed2
z(IcrH-Wlz41Fl`hX~YZ#U9W`c8gjz<tvM^<_m<AcEJL$733rKrsaW1(XI~+wpkk#j
zI6^ry;W6TrFGec;gJne|s%-1=a!|_Mnuv<NH&#DLW<wLPUC+rgzD@LYLL6DDk@K4`
zklDW3GB`>2-H{twHDS*R%rWLdqhdVslnBXLx}&_GqXxR!`2<z6tcY-FH>5HX5Agn`
zO#!Q!+KK>KUgSl>&8=N?o|5xJ!Z~v7cba)F`0c7AJI(9kYlmf2aGy~uR%-cR?AI;%
z@V2;dU)dhNSJUifvcG{gaxUggr_DGps`*Y%S1f^7z;9(+)&QpSBt8>}mbM*4;de)3
zDI#K7#pzrqZ1LVf#<*eUw&7FHfN=%g-g~@=^BaWh>QJ6%t~>LLsNsfSFbab??M>T_
z&DBQ8eFDGr5k)oCL0aEA^nur;b%D0Fmh-z1VATNqKBxKKAX7ev)p`p8tR4i?SG`W&
zLIRHu0{w|FYDtup{iP+Y(sV&mZr>2>n4BvnDB1~2Ol5|HN$`q**9q4&l*dhW0lxkj
zB}ta2d;c!CvLagEH#vF@EP^puBNB!J8jRueiWAHp$8r-mkB)%R8!k?c?3>>oS*37F
z^#zy6%hfA2FejJE1m^tSa&<kwLk@2yeC3!Sie5gGFuRI;;n3wTbLTuRM&;h@eAO)A
zscuopgdTU?j0Rb&AJd=N@;RLVd{may4|BYnHMmC<{z8DxNmtN;`LlYy-dcJ;7%O3C
zQV@9`0kcnV`WL5#k@H~Z>(VB^=riZhF?4m46z=9Z5%eCN5`$^r+F^s2)#=Njq&O|Q
zE|i@@H;di8Owfrpo5{uJ*d|Bta%OI+F`(qg=sf%Afy{eL7*oS;a$SqxyrEsL#H=*9
zxBJWm?;B&kPe&x%8bya&Lqq{0MBYC(3MC5VB`7cr?eZ3eXQVy6p998NAgr6FT@3^y
zUug|+R_{cy=CVO^uMxL2x95B4^Fk<pmVn=<bw@aJBH{<WC!CS-Q#}d4KvMHljx)C4
zz*8;s(vmMFrVSCSUQ?yt#;C<Dp+oV^87&+;HtdaFs^gdqXWW`Ycl=Uj_NnE_gBEM2
zih~I%bVNVjEoFSO7*~!E=zwfnQ-KmO726Te5c$DdC2w_sY;(bZ=``_DR!a!h2%g(5
zfe+1h0|E6dH#Qf!SKd3+7Upsku_M3TXnfrGO=*xFA4lM;WS7F6IcDBFNUfIB#YFX+
z2(O^|(=;)}%iFwmJ9m3;ER?{+La`G=mMY=N7gT{u4=)v90>cyfrQwky7*XD^qAXE`
zYvfzI9M(qaE*hqD-)Yds_1u65lB5n8oZtwU6xlY}Xa@zM(fD|;x!Kf#Z_ecH{ipSQ
z;Ym9LXw-Mta6%UMkCIopUO;6DUmCIe9c(btnvB;78Fy5dfZ<F!>VS8hV_yk$`D2<@
z31FnK!cSz3SYIHYwc*rjNC^w93WY2M7j+?EN+VwUa1qO+8$CLM4{t>#rziYwe7Llg
zklq9CsA9Q#i|)xu_$|1WphQvV8@p4j8ex6A9CfrV2|dgBQX%!X?ZKo2tp?G@jxRFs
zINcJw_ajdNZxwcRIW+>=rZl!6sfnXa-Y0B|tj-{zoE&=Vkynq*vS146Aw#af<;6AU
z<iYfBTjG{p>OA=R^<Z1lg7DA?Z3=aF<nV4pt4#`+emSTieaB*&(^dIK;VynIWWaZK
zF7?u!(*u@lj0-aJyFgf7WtxoJ(7DM$debJ57_qnDSpw5yR5L@E)9bs->g{S9-g0Pq
z?Q2@B?f6-!3-;+TpnROX8qy8+i^E<g-Xz)$)y^ta>~UKjjWt$yz+L580zNWP)kY$m
z@_-jT0M><~Sd~|U1(sDQWM!VGTxc*irxQ>6W<nTQ9bX{Rm@|2C2^j=FH0ZN@@D|Zd
z68Oix9KfhwI3X*cj+`K-2v!1yjP}h$tt8E)j-<giXkhL5a!8@iJerDd$W_2T&HMd=
z-?BnHXGe~vO;iAuWi#-$i^UQmSevFTV_O!FHqDH;o>1|P{x89FntN$L7dHoKZwwy|
zPd1i6YAdZ;^Wt6B%Au9E435vOJI5G^v}--XRv}(fDQ7Kf62l@XIf&f)y6dUior8jJ
zy`CS*u~;x5ckwEE2h?lHQ=%|hW7Uz)h2eMW!($Cgo|@ObDGNFUyhfA+n(N0t`u(Os
zy=|goYiyIZzZ<Bmw{@iGS;?ahv6eSXbHx@2c&XE!`^u7#6U3Nk4XXz99*>E(aW%5e
zEppMIS8sq=J1xJ<oJYwjp4W1FtTEc%b?Z&QS?PIn?I8|InmKM1Tc2T-QYzCek8LC|
zrwPq#M(j)<<aX!%!Grlj$=E6R1JyBZD%qXloD~}KCC}H{j7N1o9IyqGZ?b0X$qymx
z#5R|dy41?a4T~O%EF9eU24k#r7ts!#WC<^}erkpPZ2y4wx-1^51w^Nq;48f=LW35f
z4Z!$aIOqHxt2Xn(FA<P~D-wP=HulPPWkZTNuaITQ*;qRJ;+3JT@rH4ldec_Id`&}o
z^tK2fp>MTg!3y@g-rhE%hs6~yc?1T-5qO9K&b>yo^(2fGaj>epJwXLurPx?pRz`DB
zqi%Idm+e5!p=p2e7@hoY4&9Z)(i2i!`v&Rl#{Bp-`Uhjhq5|FCAruQd%F&&gRt?Fq
z-!)Y00}pFp2YgQ0z%A^4^S*NHue-d60$ac|TZUK%lB#)1hBnz+6{Eb${T>*zs`s~x
zZzQ+XgSpg;92q;{Hxb6>`-?_Jz_*U@O?%t2Zz6BJ6127}_D#SLgxsKo$exep%ROeO
z2A?35HP~H4PI9y~(086N-%IbVh)jFJrxuUFHYy$|G1n1nq1A2&o2Nm}9reamYzChU
z7;O-123KhZ1bjz{RA9d&(Sfs}?3YUazR5h$p$pfjDCG$!{VX#it7^%EsR+fV7aPbx
zF$D4hOh0^H_d=n>48*qwoO#_#1!VnL#hu=uyLJcrxc5d3%crG;yBtK$^TA67&da8n
zIpwz(^(>H5tU2SiJTFTL(@HS%d^DMFH&780<;eI_-yj6+9ZM>-u3pS!X8X22m}Vfi
za*brEH@YQnj7{(p>yRjavX_`ntnGjsyPt4DxjSx?QUcfFAbsB`ay4rElidwpdSpdw
z!d<d<3RrLO4S=V_YxI5H!R7^iYv~n(ki2NT1JuW8*VOZiTwl&)vJdl;b4~_rFu3UX
zbPr3`?Hl3i(F7sBw^DL!0+8KYjX8UNFce0+I@F+GHYi-PTN?zin3~tlTkeGJLBTtJ
z1vHng_IL%2`1}=K(9*3KiKWzsE*Y-CO~Bnyv+hu>#xc_xmtDX9UBSw-p`8l3C*fJU
z)6lX?>;dE4Y!Q$~OMH&~B82dZ+j|D-ql@8RoY2Mh+#g8b&bLr+>Cl=U@GaLV<j#h*
z1&mFzjp5!RXjfIi@AK&NbWP4Y=-tD5SOXJ~8SreQSiX2jBO*{7=cV8A9-ZiO>PU4d
zZK-bEq~ij=2mam&V(T|+dfS?oPhOOhXM~p7)PsSj*>|@Q<8TfcwTF^pALT?`wZB2`
z=n$fMH))wg(2FoI<Sn;QEctyIuA~QC<-&3&Ve)bz7F;qe?Wm`<;{@=YVMrr*i<~ca
zAh;j_zcJ{nSrU}&J!dk@m316xw-D76jIWT*9f_Pz!UrW#FXi_edetdMIe0v^c&P(^
z0kAW#?-l!94gR-g9izW^aw7B$I(_?ovxw$8q}V3wIQa<fAKcC^5i?jmS?K$Ond<Af
z$td!amb4(>Xu4~MRW-vn4ogJ8d%&Y^OAGH6T7ZJzQ(|WIq+f9-%)9S&_7u0=aD^Ed
z0osjdze~SH&UP%WE#eE{ckNC(<B{YKWT&QChn?1-*WWz)JsLcAl4Us&UjV-tskuAj
zeF?oTKKMOQ*BRs9Pq5~I`<SM-&SQZap$mr}Rh{Vm%sbWDX&s9JmS{QX5W+^d%ne9x
zIW*o<rYOJL8^n2FUQqL6;aB9HnsiF0Q$v)afz>vqpB?dvyd_JGR@RX`R)+vaY19<u
zGFohifqTDJ(U7A?4)WfQ>9jaIosu&lM-);VZ-pK?lM8LrQPe2ETSdq<wKLjp?ea&f
z!xJmbgl3cr;tj#mU>)bKU)T_2<gQ+bVMDq1wTYbr@F~SAKW{aNom+<}*3D7W_ogK}
zuZ-N}a%sy&!Dl&lsCyG28DSg8L5_|a(nVA#VOM+NBMCf~^Ne^px4go)tY41-hr$sd
zu*||#>z>xi2Yt&=GgbQi66@x5BQ7w#D&pLh(|e1#RUNRFT{*jVA9Hp)c|l#ouxVy5
zSuSV4fctIE*j17vIW%U}RwtfNx$|N+I?dHB{x#PC0{0`EIb_&T#Si$Xprd^@y^c9S
zEDlXxhDSK^BH=e%Km-D9ypT}DXlii~xuc%a>6M`sRhzWhf=&;Nzrr?8yhJ~HhYW@c
zg_bmOQ3)fK8}iApqjth%dzH{S^ZrgLS#f8@=;~Pk52}n>yO5n&ITP9fom|H-<Qx0#
z^mIAoTE=a8G{n&^@w&BZbaVpmT1*=r*`5M=cbLRcuPQk-_M|7Aqi(QPI}h||$uSO7
zZoy~7(6@x`ATYg%*7I1UwC}$1#8^F<R7SitgUXT~V`ZfxOWj2VXm-e>Dd^<9-#1NG
zcx<Y6hG{Ur@DWmN{9~1MC|-S<HB7=V!Zf7V-<QSxsc&70tAsFzUqpnWu_NeoS*~=O
zqi4J(jHRpv%kZAau^Xa*-bH>iqE3UPOt*=H$hzkQp}ZVNAN13I@LbV0xmxu1fF9+R
zmGo$3<g}OjAYonK_|i^S6GF?hMSi$jOx_&)+iPAR4SGjQ0}?d-J9b@nkzzbh^G;I^
z@6!bG9ZL~KfTq)_3Ht7w+LbGZrGG<@p)W)e@L3>a&;~r=K8JJH?*+T%gU<<bJ7{P%
z2f_2p^kwW}yiWrjvWAw+;fyU7F!Hu8YE6;?0l$6HvC79luXZPlhP1bDD+>1=iN+H$
z)yVC&JmY%>V(7O*Me8;&-nrJ$pwT<eII(!u4r{zC_{PS&+%#pxG+Q3609FgYn@>h5
z+7}cvqw!<Dc{Cy~PF;6ayqqBWaVMkC7u>JpS6K6)-@J6#1l>xjEtBQI$X9xo_BOvO
zFA~CxQ@|I6TaPWr4C?3Bm~63Js5`b;%mcDcTn(z~Yo1gUwKAb#{95!y-fwL_dGy51
z#3B{=;k1BZb=eFK8NrBIscg1u&TzvaCI>;zwpu&&{WLr0O&-LV;rEE-x<I(A%AC5{
zPR2Ai5Y20;i;A65&V=^j;1ZM83-s<AZg3}mFj~Rpwq!96k%@P}y}3_;<<%Q}&@;WR
zppoA5SR|pHaBh8mBh+-{MD6|L>6%gz%AJ)c)O5txXR|hO#gr>ghkim`-pZaazoThZ
zcU2qP!j?zBMZt+zX+h3_XC>p}a^ReC=aA|g$4TAyAe@<l;MHRpj>bndPr0Z|N`tiV
zIOQOG<a|oDd-1uWA8o61gS1taNIS)x8rF8CFi%d9<dJ(fFm(Fv--~mHDT)g=X5K)$
z$!;`nw%_kXFPPcB7aGh>c#Uv_qa3_*CKvzHrYhvwk%jND@b-w*E4H08VG6yb>Fu2{
z3Vnm(qT1ObR53cbEoC|zT~cWY6g4R`npaeHIwY=s%<6@kfAWN6kF6<!sO`+@wFzH7
z1pOnu1dI&PvChhfzO^n^LQg{O2r42l-e@*Var0!rx5%t+$~3;@#SDaJw786Oa(j=W
zSl8HaeM-hkbQ3Na;qe77k#X-d4&GxAJN5aDfHy^`<CfEYdXBy65|zFwWZ&~0jW?}x
z8_2uo(S%c0Nv3|+^Parzr%c~2LYui6Bgebaes0#Ft98i)N9|<QurC!%9o;fN?|bIF
z$tb8|$sccv>){n@$hUl0LWA6?cj1H1KaxZCbIfN7m|jo!mY*3NV*{b9Im7o2X4nUh
zW`8lLFwRPVt|QkrMd-nw0KRjCWaSGKb9%hwruOnFsyPpgw(ZY0;Zo)a5y`Mc5M+my
zD6IhQVuZ_?2XBb)j%Y>-FacvH8&(b}z%<~I)NUDO_UPXw!Z5zp(t3~ccQIHWI|19i
z{KKGWZ>O7EyRqg1Q&~3*rO_7&nWBW78&Tx_U^=(0|C7bW0pF^`(5+i7QRt94uGmkj
z{a^@&77!^jviSp@cs%~J?UKl2!t8?W`#>6^C47n;;UvHtO%Q`Fyj@J;o!$)@e!+{(
z=7z+x#s|ae+4e5E(Oy5`T$N^eFm{GDC-|bgV=0aoyzHIuVqpo^Og0rhc+(E36$T!}
z{b1zsx>>JT%N+FS+_bDK8M*;ensd`QVMWfH4$`59?{8ZOdT$I3{oAFTfqeY!)>Xz9
z-Vw$?DRqVq8_hdjrGQD(i+^7aF@C}n@^94LB5ECQINC5h_+Za56JB#$Nn-6A`K3V*
z;kg!PX4W@P*@<DV#fMAuk^|p!tbnl<F&iBuf$WjFd;Vmo$IoV)wL41CdIlt_x}BFr
ziL)n+x<$CFgfX1ZA%jSDa8{dg0ArWp4(`HMKhbT2daAWO!M8WD@e9PdZvd)L7OsFq
z=32{;y+l~fwSoXb!fOocKp<NG!S(%@ebbO1D>(=Pk52a@6XQ?#P|HUQg213-&;j3S
zR4W@7EX;F|x2$SxIn{P=LV<ZG*FW+V1MM#D4%MKx>JR1xTh@MEHP+@XvbM#Waq7G|
zzxPq5Cm(SQdYJMe-^#+`rXFq*B>`{VQfoN8-Q&n<FKAYJO@|xE<LF+h$;7aS9bW-Y
zHk4+RoZqi-u%Yuz$R+0lu|<qLu*7Z?eoM9Wnn+NdocBeP-95{Afcc?B5x&Vet*cjH
zC5m#V8GP?2?Fqj*A161ZK9=x{j@8jG`dy@f2b`-moH%{KSaWv?@VeEB)Ay9DnlL6l
zGmIsZU<6FT$nr2J1tJFQs@9x<d>`uOq%rW@E*8D(V!h=>(obxgH8eWD3DWE1ROt9r
z3xHWdO$F}Al_iW0jSI~P59Yk7FTO!!jX9(AKIo!5)w&F{IYuy{XBrNTBz<IvWVpgY
z(q@HMWhGi^!S)wx2WcAkjuBeLeZ_YI>QW6;wlXEuZWsYN!*ED{@Kl!uPt=Jr80cXm
zq{kBNvz!N&Dk~+RJlgW;_n6DVjV|aBnpeOGd}$YhZCS}n<k9Zcx%gBD8#KD9j46j3
zhDlh-#3-$vZ){R1mRq=NTl46*_%1UQ-*=+tHEyg@OjtovE~gFg_IQIt$3x!L_k`$b
z=m+Xi3mUY)!Iid&Sr)^A7m`ulQQY6OsVZAgI9T(*6y5J}3KA8~_Q8YK+^o=0YZv@p
zYm;qXTp78>VV>!6yTUCYiU9n!WW{E_7RNPAC~L7|oR{M*!-amzs;!at&LxnR`y1|X
zZ=tI-Psx372D{bhl^Hb4U3ssG{TCHi;2}1*6meHHt^i*(z_SYURQ4N7E_BW{(yUqX
zXld?Q!Vewc58fg~NoJ!)quN-~TP-d$@pR_|SpZ#GAknb%PN4}_EYCpdG{=75o-m$!
zpOsvqm%P0_leeZtf5V_5;p8pZGj?`EbSv@-yp)sBJ7LN+mU5Zl6a<W-W8@I%rOgRa
z51hfNJ3H1R?MCD<hra76@l*1epXNkspCuXqf#s=XLA>PW4sQODh5hk1A<-B4Ojl+o
zL$`owV{qmJJ*%5z@TKs}Cp)r0yPN<eN8O_|Gn8)Hj$hu&8eO`BCt70I+Ai<lT>_sW
zFF6@oM6q<>DM7(f-tH|_1BZ;qxAURH`3S?fyjjP1w-vO(!5$^_fE!8q90NTAd&VK3
z``tV!8s_rCy)SvINvYN#%GH~bzFo{V7CK2*2tH~U?~XG1jpZ3$K@b1CsxF^^-z(dk
zX?=l8kS>8)54>2S6O~6dc*!-D3uYcoXSPjvsvE)ro~t{rc$}Ix{C<IDCNIiB?Ss!9
z&Z&|X|Hj6Q&g9|#p3aE7)Wx&S`>jy8dct>xr5#dkIT2O~JKCY39v{7<Z?O-4Z&-~$
z1$p>et#9#8UGiW8lJ0iHR`M++h&^F2H;@QO7`NB(ooge~-2?uSJenm<p4q=aDuErB
zE0K{?81&~Ijp25m^}!3*N`F;O=l~9O$W?A{c!7k{H7L8Cn=Rt8<iICFKwi_kSs((Y
zsmZ*zS|U9(IM3cb`W>1<E|phVo0ta`4Xgu(Qkb{QqVADExjB(Iy}_0r<m-C)tdV(4
z^oU2NS7zn1j}ZFS5bFI4?O<SZ*zFq6#x^Gn#VA;-C$?x46*t*!lkcxF3sAW4(>7}}
zn2{46P^sqMz5}ehE_f{5*C5HgJ*J;X;43*fr5|Vz510>>Gq8M1?+4G`g3O90Qwtc!
zFv~5xZ9{MofZ2wI(}ENr0!CwdiR)OmT`Ga!I;2x7nETCSpyL$HNi}%ulSIOeQ$hv)
z!Mnv(B3taO`IMYow<PlmzM_YCqLDtYNv<yfw?e<*eVa3xWV=gFa8I>p@Gg;gO!Ua7
zG&&i*LOqzh^m0`dPp%yL{b0-7SA>a14t;KmGrq{)-lV|f@WQ`t^*ZyEqzfWVdP<L9
z0X9*+rrW+x>HXmM;crfSa-Yrzzwhb_l>>g`gb8=8L$`za2KrtrvsT~N{eEM-m}fha
zsb2b(@Wm3!eTc_aZNM<%Oq4A*0&r03aA9!Jf)G3P&92`V&EL{V@n9?`N#A?C{UHQK
zio@cK?``Wj_Iv@*x}oi92N2-(zOvy0H*BdO6yAyVFLe8S2fwJ%2}>NWzH58~wBXMY
zJLee>nn^Z+YIS|n-dBb!_P~gyRF0l9Ueixg>|ZYUtuM<ed5DwtfM4u>ZH&h^_6JWH
zKxkvwq6ZuJNc+-z<xuUN(49)$sr&Vk0Qa38CYcGD%|u;$<D#d&6W&Fhc1o4l`~(=y
zZVM1YU2qa@LOr6oDEcw7lZnpdkL=sv+tmt`rP+eV``*Z00cR5ps&yxapbzv~^!$bD
zE6>Lk+)@dsvy_}PTuLWY(FN{#G`iVNzpi&fdLE5lNTf)~vg9c@dqgQH9BjS5K~7uF
z%2w=1bK1y+n2AeDspq{uOQP$3#pZWcX=e|0<5hCSAdf%8scy0{%(3B=TUnvMUp-y~
z`uxZtS^U3;e9@FtsuI{sV&C!p%Dk~#T49}Tt2$tq^G=fPP^|^*vyDjLjA)($CX;91
z>=$czpE7ANKrPi>QYT+X)!b$6MxuS~l}FQKZR#o#z4e3nOUbsp1uCCxxIII*aa|?!
zZ-NU~Hmw)~?o1AXX|DON@ZeisBScY43eyeo&3n^Q<u$7BXrG*9g$utntQWS*9}Lka
zoq;2fIi)HaK5iTP$GqDwfT$_Ld-r$BOXtCLXEjYSg6Rz0Jt(*Dv53ooH-uMRxVh#B
zjvOf7appVK2<Rr3N?TTSJjM}wfWButbRtT$wM?#T-H1JUa$zQ&f$Ml4{qFn`X!cvC
zs8**E1LcRZ6gu_P93$!nqenPBY3M_fSf%uuD%ZeQrCL9F5gH76do%}28H-2{s980l
zBSe>|v}Bmp3YN+oD$n@uA<%#K0V?OTBHZlXJ-b-CXVE15A|^Y%Dc3g=ZQ)F$qZi8j
z*19QC@!;sUB<)bL+tFM<^?0-8L?Y%r0WlHS=NP?68E1^})d1e-ST+JKaSn}vOc7)M
z%4Z;9%*RXx8QPK<ez&t&HR3H8QTqo*i@_<=Z=Ibo4Kkhf=FNj9gkr6!I4xKs!1r|X
zW+4nmv@x|aH*dD7J0fA4pKQc^%kO?r^)XYo;j){PUf6xw1ga$!8~BW{s)La-bJp9v
zF;EuX8C%Ze#c)WEs2nqbrZ@RNMSg8WS3Nkxz_2~TLUU(0#Ck4^X0=DgJ`)sk5ZiVm
zYW~5uEOT!MUM~s5TP(_y;LQ<%qqzW0g<aaXTAmTt2TSCxF{4N@sR68Th~w)KUubrQ
zSnlZ&=_xe_{2ogSFjS~OITOB<L1JzBcA_2@Ry1taPTr@3?2k2-*4cTRNTE3|3SaHJ
z7ILy?z!mN1Kw%&+I<vNDCy{pK;6sllcd5+Ktmfc>KsTY<TS8nP0HtDP9@q=M<FYrr
z>2S|ECBtQPVV4q4Nb|&RNGAyWsa|)(!x;EI2gijTq=~*WzDG!AatP7Y4F0ti@C9Gu
z*wpV5JC2*nmEj%zl5;Dc8p+anLvI#%v{K?7tr;iKopxkBH)ZkE5}tI=x)qw5&PaIA
z+QZkNMEZR~Vm_R5A94A=Q@$%Mz8v~CzIkr~cDuLi79EzreS}vV8p<nwn~i2^IJ*U5
zVTy7eR&8+JSDx`uZf((LIP>6(F_3*Ic4mGjbR#KWs2+P(tRhfFYH?TCqA%wL{9f4R
z&Z~oK&rW!i<Am@k;dBj5tB8%(NEXgKWxJ!8I9j}!Fx+WB?%s#&+#((+3|I`ILh;X;
zbV_g5XnrrFiAz&<^V5n7w8l9n0*nlrYrQ2O5(r8?9O?Gg8Tw!Xi6C6l<(y}v0^+1C
zY9#Z6hgkkwo7wSAd)?kokTP}|c}uoRb*@y@^UdTvxy{K>Hc#;-M5?^YO^#wJ#jk+h
zD%YLne0z!Y-I9xcaRdr0ICH>w%7*+VQk;V6QoGX(R(^xHfmbOD8^LuM)F=GD+SB$j
zh`Q&|c$J-?c;1XMZC})t7@4;PlL1&J8%i~m6K{f;TJmv2rDcb{CGfm<lN&gk=g9rY
zsqDz`{<a+X#dTKUUu4B+qG(EM7Tm~_=#|BW1=Hs@jaiS2sV^7}n-<V9b%7A!_K5Ac
zjOd%QqVSDwp74EVsznme4Uexng2;87e>(apJI97q9n!pHsB`SM?Ox8qmK$Xr(6`H2
zCevZymluu`$njPN#@Pp7taJ6e*IT|H_f|FVs&p*{kAO+fI|bG2{C?c`Yjlqd#m(#7
zDr7lYQgRV630SgZvdE<r<0l)qD;*$>%p0!<@OigFFkMjo1}%z<qgnO$ZOh{I1tz}S
zR><4Y<gsXq+hH=dX&WC2?;?{qZsmxEKwuaeV%eg^@LkRe;Yvsc?^qT=-zTVteM`6&
zuA;iClge)k8%jjLnA9FCBF&J=2gY03ekW);CXDFZvEK=HFnKZWbP#jo+kB(`0lXwf
z$xjAJIIoeOj}v_!HTZiA4tVdjuj?48=LG4XTiw^#BQym@H07j#mjj?Btq8KGGWbaS
z6kz(z({w9LAQq&hUMz2bPq3d?V5V&i2a?(~VL}{JOwJ5nyT#Jp<WwhN#HjWzsgzd}
zcv+UbG%V<iD}49pq0FJtQ!`Qb;`T8Trmki0g&BQoi{EdC*<Ih?xM*l79+c~AP6D%J
zXR&*!d9p3d2`9&5ivl{QjPBU1+;0Gi)6NI8LVFcok`MMpKzsfj`~r<NeJ4(|g+CZa
zAL~DyEJm&dmvGu%ZsP&B5{97SEm8QGgH~+1T`1=2Pr`G^{*Lk^jpG3~OP6eUnMJ46
zf@btKdw%&#GAft0@GB^@w}p)Ql2JmlD;~_EnqyBc!Hls_e_&xFfX=yGZ|VtSDBu_T
zTCVnf+83*-E6|DD7kmUGYsskM+@L1*Xs=vqDwvkSeY@TM4RSHN>7knv`l7%`+wvuo
z+(<dDE)(+Q9u~EZ1YB+5&Q+e|A_MchST?Fl#ndo@o6ZJTI0QZ6cV0{3a;xITb_X~1
z8R;iqQ53>gSi?zGJtz3ivz^M7p-2r_K3l^U%B6pU49G?bI(jdWZv~#(rpg1R2BCSO
zf7#>>R}_BuQxAM)DXhMu>@1^Si_eY+={uyGfgv&}=;(Og<^*4?d(2=RWk<*lmlsp`
z>#cJO1{~zlGQ(glQZxZW+OW#rr*;fHPq?HZ_P+CeU!+>5Wf5mSZ1j*pdy*vI%@=AA
zICU%J@<sW6VH9%Rj@}?8<iWd!4P13`-Eh@jQfbb#l(nrK>V_=Au(}g&Wtgi4_tcRA
zs6FBS>0^;3N$k}}R|9pRw^r-emeGUID^H0nn>AXZPbkpIX@=^LSdV_g1IAh2i^E8g
zhI1!up0rml^&4cip{^t8XU&`*z0^iAP2V~6`!Tx7t|U9-#~n1|wNTEbbvS<kcE~qK
zx>9vDj3%#@4A6j`IX~>VTd44_P`I{$|1!z+^7H3R-o?nlO|6PSQ&ax2RBPl`ZFUmI
z<kU(l)EYkcbczsaCyaIwHHWPgV}wvAAe>~EiEeh9Nt*p0OC(SoiRTwJ@gA3WS1Qs-
zkNWPgUPDzY$L`qwbt|jxkeXCWg}iyywI4egX+=n`j87YN`<w{Arq^cLD68f~5d4WT
zNS!4Ng{diT@a})cW>dz=P6mqtoi}9R8Re`7%~}a>^})IzG^3xGw+(?1TOPgGsOAJ;
zG`KA*@cvg+{D9E{@!~=|Z(ik#2ILqAk-G)=S%M<7nWKx&SOk7>n`#)zy+@(MtwIR=
zOxKp_I`B5x?Va0-3dHykD$#Iu!l<o7LCaIhnj*BFMr>t8xL4DcbT?Yaaqy=&KcMP&
z(d!6stPm1q6PB}pf7aZY%kr87%qVRgf}|QCY%!ngg|+=V7Nu_UXzvZhMrF_U^uCI^
zb3`tLvt>KCO3&NO%R4AZX^X<Q`Q{xxa6?U`-?_InWkEh`B5}nAS-WMFo}^K6ONehO
zkSQ8O?{*|^M9q@5g@Wc5HYVb2s#ifRMk_iZ$P_H;m#7@xgN^Sk(k-Duy`o6zz<;I?
zLRf689j#Z7!RxRR#v8JP`drPYiJz&L7z3sD5B|TobJ@9MIj-}(pW>ewI2?$~`(?-j
z%LFvDd|(d<4TwWsC`gno*%tiv&M#I(?W)}^svEF@o{*>4&Oa(Et12@xGGZ;Ls>I7e
zIreh+`0=KvW>wnZ!5Tl}R?;mRME*)r^tP>kMtl?FbQU`kS;L`NvD!^j?@+Axe2d|^
zK)5hmn~UP69(T?PrRXZEs3k7QIl)w24`uup3~x+Twv(TO*~DAAWZlpftkFL!6}H(T
z+or@}J}xKarMUZor)h`r=q@=_+~|dDu>xGpQ<+7^G8I@R@j$hAR?66$OL6T;;Is*u
zEvC{2s>=pgpnti_+MuQ=j38w^9WH)1<Q}!HwB2&vPk&%bRh$-aX~&_i(LWSH)tb}@
zUkIs(TP_ZV5P6R)b~0>~NFNCaMh=-p41tn*3(M0AyY#Bm6}>y#PpO!pPq{CYV;$RI
z^f6-cq+ZQV+gOL<Tm-cFV<BbGZ+lZNYsz2t#66yCMJshEfwu9Go5O|dYvrc2lCRc8
zh-4(CV5zg(urbw|Y+LsbNinJ>OlJAke&lFLFvH2K(4qGi7%q)!@?1o9dhBT+@ZLl5
zQV#WVT#~rHjGP^_2X4iS+okc9uhXAlI?7Y5Rvd4H>Eow1<<-K3d)*B=u2sPW&TO~C
zC=bX6r!UvIZUkpk@H<<+?tL!;sTR3Sl+**8^ShJ6l*b+SFb6QHCrjD&)Aug;vm7;7
z<wX!A2$<?$Wuh*Rw}9^6lq?Fbk6uZOJ&Vl3D$$%}&a$mwe$XsZooSYjFy-y!w&xZ#
zH%nDvicJ7M%oVm98o^cvjM9td9N^lMa(12i+u`$v%UOuJ$26_DshP%?E#vzdlD;id
zMYPf`k78V9D^+KROnt@8&k|+1yrTJnyH(1Xz7~w{0WZo~#Wz%(*TC4_Ta%GSsde1W
zxw=YUI|*ACN{B&ga*wkB;B7KDoVRzcz{hPr->Dk2Z_)S5_Mum)+z@Nf8fM4KQ~ZhK
zXQT^VGkK0Hp}MeK@m9tC-reIcYHW%pY*r018{%KblNxraN%rat6I94557o4TR=^X2
zZSkk2cqhpYszv0ns6+ICPe;3`aF2uP^P~~MTN<xr2o|SvVA4FFuP(WW!bMMDAgUJa
zSm>zifTt0#TTXKGR~3%zZqi@I;8at<nPPWa>yLNrsXCIBN>T)y`0q2|P2P*0MCGU=
z!p}V;F=QJe>u7yuD;ElbpT<X<Y*5?SS<E`9axCe64RNzVwRagIb6vr~!Xv8zQ;VUp
z*OJ`5Ah0lc@>J9|UH-*1kd1cE$gm+E3V3F>4K-g6+mqTDQE&v4x>b>}(D@0U=V6eY
zl$wi;oi)z0O@NgiVq|dL)qsGzg?EfLvpNNqh25-v!N{1Gs}lml>M3nfvH^5AoN>${
zlGyVhHBJ4%=ZkeC2urtskDr>A5h58s!8~`Rs3PAoFSsWcGNZD$TrjhB@G8!$_35nN
zPT?>#;In}D<|n5)dR-CTaZ$11wRFJe9z1fm<KWu7im9p~_-<x^(}i7`w6vGAaNO9=
z1)Iy&V~Q25XVd}iXiHXus9=$(iy`W!Na*R+E83oAt%;=o;^P#C&t6|ePn`4SFfrkh
zSCL+Co`-JxXUa>JqTS1NcdON;fTRhE#r>_Z$M7$A)vF_GlXq>2)Es@wHa<mh{jsSs
zPlt=<R6d4?rZX$0UAEJuG*=A(=0s2WhLD!_fT38{X%j?y#bKz%#2kwV#SYsA-Z4>-
z&E~nomIk|6pv0f-X9gzS(Ooksqmmr`K69{Zhv8f9=*fAfA>lMxt?WBT9J{oJ%Jm0~
z0C7nUiMM7VQY^wqM~SVdPP}`SlvFagBoFnxBRy`1NQkou<e-!UK1-+@Id5_eBg}ou
zam^{1d4Oj#$*yhN{pdE|L#tjqkfVDX6R3tC#>o^HBrnrVu&a#jp&vIaPpqCxw=hlR
zfa1KFXoQSd@kmy@l%SY99h}R+d#|F|<?jwTLU1!Hf|tWr>c*+3NN2*kXaO&K2P{1`
zf??#*o#k7tsli;)k+aiC&5v%8U<mlku#)#mk*sGVF_kOYiUQYpz-R4{`dA?pb+z8+
zd|9si^gYTn@7k}(4lhMV&b5&o87?U0D3ndkFaWpt2-x%)Eq%*$bc9aZHkrTBy<{8o
z*ic=6Vei`=hAeN3fs6(Fg6q~x1|#j&=M~h)-82sz-P9sjdGhe8Ou$)Pr>q5g5niD_
z0L#8{?Kk|nH1s8+-lOwiPzaTsHNUqDM%`+gUttqt*f<;JAVzA%+35&vz{+bO5hOYM
zEr0~On>^UC6C;1z!|D?x1U8I|)#0<UXN5#b()5prxEfJnKdFyFVoq8^EUrCG*Atxn
z&Waqw<f|Q~aAE3mX2TgU^hA4VbC%*2v&7S>NNGaoSYyIJG9q3~NM|qYOEIZK&2_{4
zpwVH%z{|Z|7mpxIKih#X3ums*0AR`c&E&eO)ws;CeV+Lgduk>PDmzSlnv}pY)Zo$j
z?9=mf%^zovAZ_u?Nn5hpD_Bx3n{ir=DO1Rruhy_&%f%_^Iy!vbL%PjypPQM{y3f=j
z!?RyD$O$Wo7o5S#(QC@*R9-@f%XKqos4C3K$;)-c)7PdfR^oR%28uOl+S&^bm~UGk
z9Ai&wui6S?`zhTM+v4OtFKiN>HFtM*Ck&ZnlvUCG?U_6w7qo24KV@g@0VPnAGT1lD
z_XCzXvcjHoCg&9g_EfzEw?cD}<!Y*SX4~nx!(%&xZp*%2kknJ83WQMW3q{jF3aCV3
z!$Px6q-F(ZzQ(b;EhgOsiSWEEbas?Op%9MI96tLD*XZihK*#Vv>nYOJv(NKnoJrNK
z5wkMw_U-)HlZ}pv8Ow`lS4UvCD9WUt!w&_4k4x6&#Z1>00!2`Mk^@5^++P3^daTA-
zNicp`4m`4}bXBzWhQsHRHCZsi#M@~bfg$@<f>-j$&cyM~)%jAl*FAuq|0z;^ggzwT
zS}0()u_g(n1l*fjKJ%)3UZm)xt%7}$?|r?t3^3HVa3sg}oMoapgFHsxKl@4gBpWA<
z=@J#e#Rc3<@ZO?8+oSN)`~@y)XDGvkg{Hj$Sw<+@O3{<EGha$eC3-k6_&AEid0Vqn
zWgPe{`l)-O#qyLRU~-|A8Gjt%tKYEQy3#KCA=VT0^Nvd)y415wX@xOWJ)WO^-)9+<
zTMSIhYKv1c`7!u{lV~p|6rkoO-~!KqSVAb&+yS3C#g)j|kp~QItM)30_x0!rB$bMV
zatNgt4S1L30n4^r4R=90y}d%I1~(XJv!S=J$=S$-jRql8SJis)6R8F1w9z=(CdRj-
z%NRtvQd7|yQTX7%h&83qZGPSQs8;(#%^Otmjvj+{L6cj3^#hF|o&U3)4HML%h)1S7
zJL@v~tnOT{Y3^$@rlk|m<rH;`i=wo(bU8E$v?Kx(FA$H>4xI~N!VX#WHM?G>(aBR1
zYqeP>V5k^|9^zgOlZca-*ba`j;>@MM=XH@R7^jVj2a;8Hc2*)w9hQ6(Fo#99w?XSe
zFR&z&OrVUM$=v)^r7Fz`<}8Tq1xKLM^v3NZan}|$j^m4R7hAIG>j3DP!9-tx%IJuG
zbhsjP-V04Zv?lR>d0y6V&k%S+vY{y_%9a)hJ!y$~Km?GEZhKi2W^i8BMM_UH(;V@{
z_5&EXAsy^cRWcl2P~odE5Dzr2h)`WuCOmR?S8Mcds>f3Hio02e(b)E5AR-t>&Mcio
z?aS_D#!YTSQhd=ZUM&QSzKI8WOw$pV&=iTO<{+JKTajI0#4?mkTdHr|7r8;ngvbb2
z4bM=7?;3>z1&zBWKN<`6-cmR-%7E-J=k`mY@M$N-)>bhBo@Fx;`FAOdaJbqfPqtSJ
z8w1{E30f3i*+9=nAvS^n$!&D{e2*{)^QD(suM0v8uMF1n-n$h0f-R{NYVrWdjD9TW
zVEsHa`Jn|c6sld}=<O~U3@R#;u;_dBd{^MMbY{W1REcgM=c>s@%29^7px-It6Jhq0
zp9WfpJc<~mJF5^d^pkSRT8epudC)WGkJsRFwPoe#nnlhYz*AUht2$gO2JJ31vZ@t0
zE4veoWnT*5pk5{xMjlI93cdnQOEWT^pyZ<P3!2cC#Fe^2yb~0XFN;QVh<bGo?rnvV
zYn{&h3_P;SlffjO!^mKk7V*IGCu($iK<T2I9OxOs_L*0=Y9B2dJZw+cc~o{xp3ew)
z@?n{p0-iy%UWo1429d&^8AEc0#|6hOVpY&(`z3Zab2!1kqQ-3I6>1qy-M4rH>dLN`
zGyNO`7>b7JfyQ_?jlyy(!z49#&8sy+TlCMUu6_Hkno7$Q943gl11@8^aMxL*d_guc
z=_J+wm^GWDFr8<QB%ixo0aMN`@2BQ`!Hw^JML|uAZZ`^h<#6$_MgPu<k`$$cy=7Qj
zJ=Zo|yu~T*?(Po7-QAtR-5rXwXmN)k#ogU0?(XjH&V0~Ip6h=9d<SGFIkPNltvxd^
z*@-B5lNgZ8a%B>8^qhg~{9##}%o=Lso@i8EdItt*#z8wz6jlI>aA$fe1>!pNQjIuz
z*>L&>C8Z_O?w&q^DCN{TPq(LGS!!$?C~7CNOb^{|?%RcxSvL*$5$&0io7mu$)9T>{
zM$7UFxV!0;QjRE8Q9%hD>eut%3@g`7_Zkol7@BiD<?Pg5+<Tw`0-AFI2CDEAE)sgb
z(Ixd)JEwjn6+uF){0d1<T-h*VR7aw^5PWBZkeRW8(?2sMUx*#{`)fxm-|<zJZfo$9
z?#7hy0W<uTcUU;<MvE>3i`1osBt;=u_e^X&=DL_LRPW3PTADvFzt7K#8_p^Jw-J5^
zph&e|!&D7pLE5ByCa_tsVhH?OFzCn+gnL+V)}<P$D$FAvSP-5PjNuqZK?(P03}rPJ
zqM2PT!;YR!<(H9H2K`w>2`g5^9)m!8JKRdvG)mX#PK&;nT`exuNyF~w7@|I6D9ZV+
zB@y=d<n!B^leQ7e`rM?=1%DVKzG=gF)m3k~R&fNX-KCs?`Lq+0O2=gxf5myf8h(i2
zBgx4iBX^03suhB^Tb7`$<*tu=XxRedw8{j2^%oI~<`edk&S%0ON)V?e-xiU3MuS@Q
z4krX?@i@CFv;=?obmf(S^jnimefv_UV#{)<4dc_Gy^i=e@lA)76^TV<MET0bi0M>#
zI7Ov;gVp|HL&$o-aYsX9W&CBpAghc-=TKn^bx1%NPLbNwP~L3KN)b`IJlw<m#t6&Y
z6c@9Ttvxd!DAqS+d=7rdc+N(|=aF>;d9X$r$~Z+-*1srEFY6FvUuXBnhxvz1UTWxu
zW~o*}??%}3fP(p1vuao}?|hRz$<hUhM?WZtxX<@V1A6+Cu9Jv&9m%&kp@?q7MSj#8
zCA904G(Jo+sJ&CwelW&Mb!S{**=t#-qjbFn5ib;U1E{1`wQ9pKLz24r)`*{%y3D-1
z^rz-O2MI?s3k1h@P{m6NfY%q$e$`Vp?CZXIN+caN`J}55Q`yb-_JlNvD0c$mXql)A
zMxEZ`T*o=A>;Wzdy(5Gv(3bgwG$pC1#a>k!jdPscY4lHqOFr%eqhbJq5^InBAW6Xn
zP(TV!Dy4BC|7#oh!tnZh-E<CKB=k@;ql5aYuALhlH48TMm;(zlSWv;}=Y9@YJ6klW
zLla4zlD3fraKEweV~whvlm|*qts&Az(diAkqJH65p~36(IP-PW1H`(g2Gungn`_nS
z1TWo86a(ZV4bey;L@i3D_)Ep<TWQXWM-ZOG0q<mXHm&*H4&}lUc|~nr&a+=I0D+bX
z`ummx$ym}~Wg*s=gmGx0-)15=bX~Nw1<!dmLsMx(DMVauXb;H-3cgQZc1m(?{47p3
z*=jqNnUZy>%R3VI&IwkE96?*D!Ul(bK5htc5(KJ%-E>f>?JXO(eHT!->!7_iGR8}%
zmM3YbA+cpxn_p%;HDk2!fzMX*(xY{ytc=x7bVRR$Kp7^fI_J3YF{2%-<xrUvC=5w#
zv+Z@QViF;wApxHQC~$}Qo(qFlOYO=wcf16-o*tUr-^YtZY$X&mMXFFKLpQpy<=6l5
zx$woJweb6d4dC(9t85({n)k82WU!>6U!kE|gnoO))`w!~%a#j11nZ$AhgKTu<Z!4D
zwBzwB9<?XiY`!$y>8Het8MqNgE+fVVpW!nlM5jTvAqH+%OCEA1EnV>GN}5VR8x8~9
z+MrjgBGn}`%K#Xi82UwFJS~D%q20rCF*TagLM|Z^B*<CE$(XI-0kb$hVQy7uQR3r^
z8MT=rrnJA84w(>vOhvybtXFBa7J6${WpoSRyztViIKQ84H@KTsqH2-lc#FvOzi;h>
z<C?^$ldU|Wz85<5#hNL_0kysp&AUPW?cZh+R$q_LQ5m)mv<5zMHd#kwCME>{#h7u<
z?H5pTd9llO5!&Yc>o(;R&d~QS9C{I9Jj;ApNLXsOO=P+$Y1tO!21xSO2`zX<wyscI
zUHLTnd2mh#CQC>XcWOwMLMyaep<@~v9Nl6~JoKPpI=Wld2)z@J(CmLFsNqMsM@-po
zCo)%?<_kfj3p3RCkm^mbn@doBU-esr24Y=kmD0eSwcadb6N?9wm*}guV+l#A>oBvp
zM5U7CALb!T`k{yCC4W|0Cv8|~JPzHR_H+OS7dWBR@;Sx;Za)aY|Flx?Z|fE}5L4L}
zu_D)oJWUB|9vy1#X1y6N^C-oh5#~%#+qD_ngvgWB20>d6+FxI=q#<2*2`#JYw!7~~
zprZ!A^&`w;%zuz3*%2J`q?@wK(k6C>bzvow%`$2Li#^K$*VawIDOG@;D&{4Lrb=q9
zEzn0_Gq1@A+YKFYF8{O;m#+cpH;rJ_gc4O*<#s`*RI|z)8!c%nGlqw?nx<Pv8a#g`
z-@55ey-eOnWBBljru02{zLGYYIh!EWDj#i81>opbYl;^=i3;7r*ZE^MF09yW=Wyq;
zqhhP>qx7Kt0$G%GjA|g0W-FIEEmb~IUdDE)mFhRun4>x4J#Hw`NQvu*75Gt{jH%oP
zni;B@{X6#(r8wwf61B7yI;2vi=S>z4t?A8~O)TS0?Fhm&c~P1cwo?O^NyS<P&LwO$
z8_1pS8UzrwLfHOrA}+K=*=x*h5Pk+9R69P?ei6IE#3_GSW5k2qDcD3vveQQkks_`D
zIJSH`;ou6<I}1-cn{j1lV>*MZy_;Ap$gIgHjKR*s#w1Uf&=kI8TM(GeS;QVYFt|SB
zftyOQXs?+m<>bNAaI1^Fd0VK1DGy!j9J&<&hhAsmos_M4S%7MHX5A3A!F?|SF@Fd>
zz!pi7vjmW&%&71z`PHy#Zx*Ad@oB>hswRXKl;Q%leqm1h(uZk#JHZ`o`ff?oky6`y
z;N!%jcCU5LxObK%mJ|Zx#I}auQrcn}&4Tw#G-$UI8B<O|x)Rd8Tg6~%|9-Y<2zJdF
zm#M*>L+2j1T^as8E#1A>g9^kQnw8yr=%uoT8jmp8U=IQAur<wO0H@NJA@x@S9?i_5
z4H-@4kDoUD#!M-~(tUJBDJSwRuypFWuJ>0T*isEkXt*gMp$~*6z)Sph6T7inb;(n1
zahT*u6nCJu#}((L(_{-3hv^ZVqC08w{5J##Q$rbDWN&xl(Cz|&{7o=TiRT23!?QS0
zdPg$}tWNWJW^#ZGklHIgYI3ZMt)_^y-yV$8n+{_1{XNQU^op#!kHh|gE(_2YJfI1C
z<~JnXM>S9wBWHw=t;fl_O9kaIih5aID-$U-=nz!)zU2m7cpl<LgaRmuICfp%1wVW)
zY;eB4DU$dp-Q33!%~~F(^EvlrL$E|E6tv`c-2#JfZ8}y%%~54v22_cWkan4KauT%6
z(jy&sp;Ml5oL)a^fdqdi=Z)DLA(;0~<I#x0`Iu-|9i!{{A?6}3Q3-}1tQj=eLju4d
z3t5{OlsVH1ZSVpiCww(5^tFcN(S9}Y2QYh<`4Pmxk-AL#F#9)Dom=8<RoqCTAK^zk
z%8iZumDHHT#`xM4iBFB|voulxf+af-)k~qyGRM%hnLCxOaP_Q>Id!D|&!pTuMW{zq
zwv(2zcU67`;qqW!5o<Vk>?U+VEs0@$YVntq>%<?i!=SHDp8IA1@<fqn#hWXfk?Qco
z;?XwIeL`wksWsJF$Xl9dX-=-+-I-CNChAb8aJHBe<~nY7x`H~m8&5vTExRP4SR#GI
z(VMN-W`B@IQz7=We(=XLtIpmq-wa>1>=zCObv&232gqEc6w<)u<T~$ld0{IxG4YYB
zjVz0H75~7`7Uq7Om7qH%+Qx4bqo2p2)(<c|45f>*iXj&5=j9!Oy?4?Z2Imc09jL=n
z-tiFh&;m=3;i>yXD>m(CSOZbAZ49CAQMJ5RG+$6y;2d)0PNh4twt~)vQRFMb)w}8y
zLe3c{S*cBw2Syj;qexXt=Js{WQ<cY=l`;pTVUxQ_-aWCUxD(K>8(L-Z5)O@V>j)u+
zolnN9A0s0bM{rXiL-iPUd)f)Wz$DQ{{cg@SkqMF+(T4&9qVV=T%f^;GF0cPL6jx64
z;dYO68B&6duxU-o8&lFLOlV9MmBJ%spR&$`J)(MAsTPQ&mH5rFFhqf6IFy3wqHlBq
z(fgltzfH$$!o*9puG_RgssfHI3L(RQQJB^wq+KB0riAbyO|G@0(-He*)x{hjiHr&4
z9c(%5DB7IIHG?lQwac&=riLo7<N6%Sp|~HkPA+1enHl0;rXnpR#OO)&ry1*Ro!U0o
z?!BNT6Ggb%O8gf1IES{z2~p20m{F(`Fir%=lrppBOO*zLetxT>s;-UpMg%j-56;%7
z+G=t%Rts&!*!FpX>Vhxfa9l{6l0j)$2+gh(FMkvh<;8k<HSDS!rVdHHpi9;@iMR_t
zPS-_+69yXA`sJITsFN+LN~1T-w!|et-$eT2Z$JKoem3~DE%FJagTc(b!66YtS<$a*
z0S_&vnNH%THtS1Yq4JNV_1OgBrXQjg8h(LUR`o)G`4MQMqje@8xriXk4esAcd%CI!
zL1C8wC+s3cWypmUBhPs_8uN%o4dgW|b;0BX*jqM{FjkZ537tQMMAp+I=(-9xRTYiE
z_aa78%Xewg_>ANu*^2ui6zwI@*R`kQ;SV3KS!}_SqKML|KGrE>P#$~H`Ue~q9>t_l
zP_%uo>NnYLp(l%A4fk2eBx&TLq!xeM;$j}Y`b5>z0kV9y5BVCtYaCyV#3Scg)mT+&
z9wY7uvv5Ph)gj6TAxZmXy=GMHNCfn!81It}ypnP57cyrsz9|LsIvMA%AK^RT++KD0
zREPZBrFGqGoOy$`tSrICltUrlyH}*sX02nX+oGcsB`~Ig=Y<A$#pIqE5k=l_Cr!Yb
zlDC9SI<wP=@~&?TFZ{4>GOyWh-rf6U%qJF|rcitCp6)$c>D881aIDcFlKDw{TtbEG
z*=M1kPe&1t>c+R4(@Xb5bYF!z?F3xiW0U?}3EsG@L=b*q$pLk&M4L@7qvvX22po7$
zQF{8er2-i-1A8wcHIo)2*4_=P1a+}-3{Pt(U^6CSA#ZdKPPdT@%*sl|4;?EvW36s0
z01&`}EPeLtQEE1VQ}gEjN63SeDJG)`NdK0Tb*8X*Gb0?Dkw}LWbi;67oXTaOcUn_)
z2#1J3NIJ`UJR)v3mv<y)1PrtQv!bh}c6315j*5Tr&Ay9(v~D9V2@MpiOy(5MHIX;|
zS1qO)I?07tXbOKV36(6(s(tU%tuPEelD#O0c|Xmj(6$h-!v>;61t03o?YWifQH>JW
zq_RpfFzGDd7Hz>cnj4~$lFPUL3Ebdqy-#M|wPrjD(u!^nixXzyinaZqkua_F*V_4E
z!}__in;VLn`EO@=GP*lR^Mp=i)>W2?`PA)8nHQ%0c<6%M+OV^yGPTSs|1@Z;luy*y
zXp7HN=oMD;b1vGusj|8{t6#t$-xx}}{w~l*H-gg#*`FnObmG2YLBk2`D^(m!Mz2W~
z*9_qVKQx4hg$Li&sV!*9OPNMH0GHJ2%QIt7ZPne(fow+~b?i|JBSD`S&SOSYSVEUD
ze!{MQtdL7LtXu5*z`(e}p`yHa{fXQ5a%~lwz{U?PUGzAYQ0~eWO>znQE~xdbSgMB5
zaZjg*7RJdgrsRZ*FV5SFru|<qc-Pa4nbe<Ft98!J$O=q~d<V!i3aYh~T#grF5-M><
zzu6ziClZh0y`3I!7!Lqr0OKo9;QA)2`e})@mWVB|S3?v&>E7OjuB*2DqmLh+8YbR;
z#cZ(&M36$0N3E!|^duh%M(_z&(UHSOZ~i{J%0o91ftPO~gmm=uHS?I=V_v_P_1u!Q
zaIXQf))nz`ZCUei0Uss4{%@sBm+<(o=c=?7*LrR`$$r+w-F-HR6e*K2%VQ+blKF+M
z@nZ~tw*6-d`ULQVbOq?7;ou_cQOu5`jS>TJ^WnlmsrGJtvzeAQ_g|k8UcDMMoIYYW
zF4<WtK^<VJ_2No;@fO$o)bu6wWm_~eR^a;qQ3IyCG~GI$W!;vcBvl{QkCkMosV-_f
z|Eq5F>xB-d=OUj-mql{ph=Ji=pjmwP=XAa$(9K1jzPh`QQ{aT6$VNA#%v^>j&Toee
zT)|HPOjw8m1Aq5Vcxd?N1WA1(fEvVfx=VBx$>&_DPLeT_XrkI0;dbFJ=|s^ZNQ
zaY(R@k~zoeRx)Pd(}_*kE+PON(DRL#)HqV=B*^cX5a$?|T%V$xVKho|4l!L#z?_zb
zeyhzy+BeAQz_iZ9B_(eoLP|=?Yd49m@K}+;jk}@_<uXbYe#H>EHq4c3hFV6vm^%Hw
zd<&DOsys7X5l0`w-E%7hdspUealXJqnc*p#>mmjPCFOzxS?cvYRTxx?-Q0elVc@&X
zPd9|O9ntl@lm=#Ms1%LYx4Ky;>THU>=MWeFD)G|rTEXEZJS~kmVLnO3WdHtbMm_qS
z0<WbSpVYx0X+}_ja!{XC`yir7?P%DNWEY^=*WA5_tFh^0y`rQKCu)%=IX$b9>TgLb
zw7UF!s6EwDdi|@3DXxr;7|o;VdXe)5@e_t<qJi*)cZtKzwlu$1;79Gr7!p5)NcnA-
zO(K>MN}Ls$=9!Z#%!)x^+C#O7T*8DhB?|&+Yv;#Lrj{}t6V*0we6g9;jd<ktI;E~{
zH61bh>0sRrg&5Z4<LDF_DSR$(TVUH&UiB5uYa$>EH}lG|#=4+Z3}`PotS55^(|vAj
zxPVKp&zVLDvs?E%l)wgC=c;}F+2DB-uwUq~Ge2nNJ?!Gg)N9gV5Tbh`%_YZ#M8y<f
zZEf$C5hdYXx?SIcn)RxxOAT`wDwT~*&Pg*lOiv8hP3H<-?QC9V1|MhWYX~`UG%+P4
z^Oov^mH)&ATUim1E^Ksj0-{#zQxBT8Yd<KxdQP!P7j*G6_yqdkTYhr>=RE2FFeS01
zU&BL?J=karq=1>eTKPs^lct{)h<RE|Wf~@vat7d&kRncnum!a1Dkb02Sq#UfpmxuT
zIcCIcj{=HU810W}J2A$S3n7#~T1a4YtsAu93ojwTe!_MX8nLP7(H!pGVD-Gwr+KP8
zW+3O><m8Yo+P8+)Rc+}bsnxM*064fd!HAH;G((1;e!C`uMz-#-R6%{J*N#OXpwurm
zb0!i@z@@cLJ}^Cq=QH^6A-b~DeQBP`jG+O;zU4wp8CM6prG#f`aO{b~zL(jxVG^y^
zn>D<Hw~%S|P)aG7Dg~Udi^}+gRt#{tEDH*i2;$_J??7`Nx#`fKtbU+|u2y9*)X;F^
zH>Bsxk)1D?9kF%UU^4k3Pu521ij_C^s}gFm$VQeng;+wKtuv%!9Sj~?O}*ynDW5qR
zb6-t(X;NTSxF4Zebop4}5&XT?Rj&iIZxxT9wEtXEhMQ(7(fo$4LIM+J^97TjW`0eH
zD55j=azv_a(zxXSdN`YZ;zDUwx;2?WUsB01dY~t>K9ft_uiyd;1h=_dJ3Ge2DTB{<
zI(%W!wG}b}h2Nyc7BgIy;U_VPN@U3*cg;e?<S$|ED7n2LyPS%DHXn^-(u=8ZQ~O!$
zjiXC5v{MN5Qp5Y9llvs(M78yNT!P_kqU4+BwOl6QIWh!@NSpQ`?*=+*9<`+fU}3oD
zk~AT5$aS-!dTh$`3jI`0t}gC2L=~gF3|o{1Q%g0K-W&ph{Mo04s01Q_*zT$Xpp}iR
z@U(pYTTxobVQ#aSScW1MiJDG{7uPNbfA&G6ugqY3SWw4)jXBa?`}|_8mrj)<dkLO>
zkR!g;KJOF@mnTzI@i<mZQr9H7-8@AZjl7h1nR#(|LFeO!rg5`K464nEu%c0J|5OGS
z6_)G|!1*wxyN5S2Fssvyh{{Gu#RAepHH`Fd?&@=KKkhxp{-RKpBu;m|qoM`;p2i}R
zacF1Wt*Nr?J-KKG(zVX5$`-0_1R!QvHDd{Yz8U2l=oL+`n2Yx-+e&#;=rtq8SorZX
z4yJ4-=(p44eKZ2(q6sMCq;op~)Xuoh5jLLi*6KEFkKbY>AH+i}*pnw3#1J}SC^m1%
zXU!Y8Treh6iRO|g8Cs~dkjw4x7^LS$vp(BY9{g;8`?d*43ceMO*mEF)T6azeL%Cc(
z!HTHXROl9lO|<ASgDX97$moqJ*WxLb{LDzLOf`XIqct&S-NXl}BwCjs*pQpp%-#F2
zkW%ImB<cMrsYEVFyifZTQ36!g##0~;SCm8cI!@hzZz~3nX_83NvZ9p4_cbM6GbKh$
zulm9QvA#CGOc!mG8{nY%_N6#qCYI2>;x@`;9L&yzC)(kyKJmGG9`s-rR9qJeNTo$D
z8@P)i#(^d2FLea9ObA@sRP!nn$JSsYGb?o_b(gAcl1@psu%tz2k$`S!;|y2VU~{7a
z`O=l7)*cGlDk&7AUhAk++Rr!Pajm!ZS!O>xOKyjY=Js)YC$Gz-O)DRX$)gH>ZL^%)
zOF~P>H$$L#T(xk1?Q?pdd#10K!BeP>{9>qf3C}_Nl;zY^E#63Pad1dh39&bG{!_s;
zE1iyQz_ktM7oMa0rWn?%+sNGy;19}yh*VBjGa^;xNLB<eFm~j;+?e!&;GA6SBVjCu
zT-1n9f-0#?;Xk>hU6IR`*>o@`lE$>5wzlV{8Ag0WH$(+-sYTDMSQ3*yvKgW?N>j+g
zBiMzzEbu|<lt8PTAjv^mk9>?p<7XWEjc2%FEC|~8m<m?ztQU$%9O}x888=^yymYpq
zN4!Ac%|#BnF@$;k(1+!7iuaOtAAuNbLicOMOSC52Qt7bDtqnl?Vaq6t3Z(Kk+WCb`
z6@2p1A+KilX(@@uZWbnPNh!Lu-dKi<o3?=vqsgOs#Vn>}@w({eqA8)WS}?LijZ$tW
zbAwc~{JKp~jI;1=m@Ffi_)$T~@a3Sv>nJ_dGREWz-R}@);pf417w8rPCU_DC;B=xA
z3RQ*I5*DfI@~W4f>4UpW_dM%VsMNVjn=t^FV(5_!T)&VO|FI^@WLgv&$7}Gmm!@YW
zYfWY7@TA{m23~Y3ETfQ0&Xh&SI60?we0p(GTyW(yprYO;EEle7cZAR?rZK76*YtHJ
z9eXE8Q6FkE?R!^``xlays>XUcjzxqi5#(Gua;znE1<2VqO@CiRI?i#&djpllc?1QA
zm3T#-AX~=2sc{B{$zy9mp{D%atvYPbA1pI5axM%6B^I(p#%*pqg5LmQiO!1`dc$D@
zxYJN>cl<=@g$&TB26;t4>moCnEzKf;;;S6OLXNcw!`Sf+l#DnNVh?~Wx~K;jf9np4
zC4o>`{C0%Yj%q3Jm4lK-ZL*-1FTe17LBGybD83Bu4r1l^Z)?=inK2b;mibl+#X%)X
zgRj-SJiMq9K^Ksjm|rHOmzqR?k5z#LNX~Ibqfnn#My->7RT))mI-{u;5a}=_<s`nX
z**N6>Z7yhU$_yrnrv^1fH)*w{jgDXhR=)$SoqPz~zYCUJh?mh(nbPNo3{BJz^IlxH
z`RttlDj)A{c3u^3nozP!u=nKXF=~<Rz`t(6N^iv+yO--12^%}hVy&$w3zN253cKd#
z{AmwUnlb>qR`O<;Pk(UXmkwXiJ(i!8L|L`-<M9D(N1XKRx`?FOk$IY~D1Y>ebYn-5
z_Et>Vg05ZiY$r&!AAbA?dPhC;6`fA71|-aL0;P5?P@O=>NkK%6yeJqoo|(F(vc(OZ
zW~UA-a<m*c&;0CWlxCf~NRF$zVeEo!7oI?7q~u4mHfN>ew$8mv!^j}Llfzp<+$s1x
zsLnI9h`pAbaRCMVI`$nXnWt!Nz%o26WHtu{pnRXWH>|BVpeG+&r5Ng*CSy}2{R}p)
z+Es)iG;syX0-H4llp4=~-z;l>^lew$K16@ZXyf^eXk_T777}NidZv(1;7lnpnak}&
z@KAhZ05uv6!EabPqDP8MT9`h;y9~|1Qt4L<k2(0eO--G+p}7VUFC&?n2jz0PkSMjI
z08K#bCLbY#6Ft{mur$BbSei`t&~pdoODc*7Y$D328c;r6{$pycxE|n}m2vZcF}p@K
zD|b4r7^uK!=fLUWaNnG8$1xBpd53x`NX=;^l*G5E1ZuN;mf@9FpxqWH8j$2k##$*8
za3|xFq-ZsrYw!{U5B1%R?@B{OoDW@SI~28&cAwiRgr#LHl0+p%HLALE*}zW0SyXtF
zb=e0n3zHC3-`*v@)_*d`8k;KNQzS{2Z70E*<<>+PJY;27o*QLMs!lwVp7Id~X@R24
zY_M|)oh1qVc702uLxcjR8@C}YZwId;i|Li2d=(eZtLTJSRdk>&7G15lu8Y|~CThL=
zi*XM%FOP7$NfqZKNG9R+Y+>Qyr}>`D)B^sk;gnf3@46pO7YpBhcH3<*2)8|Z@7Kws
z@kL|cCwBM7RVdH%cV92q{0if&e?m58r0Y)hR%=;S)hQ}na>>&Li~fYx=o}W_*(IuU
z=UTV;xgC6((6j3EPds+Gt<;_-2g*h<tvCTxoU@Mt!8%gXRm_>#b&^y_3YX_Lgc56`
zbx13sJY@3M#MUl?0D)Ba^;&7;u<4RgXq<HsJ2kI{T6Z*sgChG1@u>&-KG>xSTWPgJ
zEKVJU(KsbqSh;pe)m2+)Bdh4t%x`E8*(KS-NpWFOt@*n9A5sv<A+Au1&#O{2^SY^k
zY+dbJNUR|+9`d1;nQ&u~vk-0G5G`ZJ@ny67VKOsmviymVqVy7*IXEM~SCmP9IGqMt
zD&yGX8k<dW!MH@ue*7(c5sjd>19sY<t!^2}kwXmDi8w?51HTefl%-pzy3H{h-^A(=
z--LdV3N;o?B0HZqOlfT*`pQM9E|hQ29<m$&#=W7v$*)x@zQvxEn6F%7SybUcjl5IU
zXnbcD!CW-EUw_?@4mZ`(|I?*`Ikw4#c?5N%2i{F+B)7X+86|=Gv;DDdZKC%TSosdc
zWTkjq*R3CEgVDtb;Un|cJr@Xiuc!)z4Mj9LoGQ+VRu9wGM$(nTNo>d*)?9`$I{Upi
z`?!g88A$6%KT6tqM07=PR<5q-kggSmkbIa7QmZJovSv==gL-&=hcF~el%4OYggY(S
zYYe3BHJI>C`vZaV<EH4>Rf;vbjU;QAqOx>mI^7Zt5k!j<28*^mkinU8DA3OdN6~~m
zScyxb%`|x9oLjPNT0)E#m%oO<<M~v-^wvkG&EaFvy2$}dGiSl-?!HF<Rv3)@Tp`DW
ze^fYHp|I|rAg#fu^Eo6$BsU_-q!0@OqgE_vVPI-zgN{L<r=;aoB?lkX0oR>Rc%Vx(
zhf0w(!N_A1hk-Ir2$hv|i~-KMB^KmvR<tH~`b*_wOyb99xZv+StY8i;r}tEOsAk-8
znDlI&5iG*RVL$8I=apY%5-D>^s>I~-DEzDqbrc<$F!;dLL|JZaLMWk9p-E~OjkM<)
zGg+L!#UHYkg@JmP_tV^9#lmF<O_*9sifqzxqADH9CvfZ0PHrh}!xty+NR7;b)RIH8
z$Ys+NFxbvXS>m$chDa4^Q2s8dc-`o9w`s%@)lI5eYo=r%v0rn***t9<EsLKXR&D!D
zN^Rn~3zuR0TZun3fwK1ZHT;rwvMHuQ%NALLA9t;CXfb-MzDp8Q0clpt<Kdk<L>&h{
z1wryer@%s#I?lYaYzX-dc^xk?k!JRddyx<<QblJS2372t$|a*zgHF|Fy`iN-g4uJq
zh8R8PHrF6`MPt|xcc&@FcJmg5#fS;)GfYBdr6VvD3n4BnzPCL-EqXFdZgg`P2W^a?
zbxh`~L(FsK`(ng7Qq^nWAGz#OLzfGRmIXfYvYs4&<59h$&HwB^Ca=dnuO_RK(eqH|
z=w!1lP~3o|`NJ$C>8qJd#I_eWk2L*?Fkz~g4t0%s4|rS<8;=?a#)8$1MI^@-1lfr?
zNpi)po@l=n;?WEqQ3s@MS|MOS1~#fpIaDX6u(DHlvPB!o?N^?aNyA6)dkVZX6b~&T
zi)fm{PMk$|>%hbl&{%l7s%2wx(#X0v5x(dq@-@HH!vOOHO5}*b*RryMZ^L>Wd(qTH
z@Xkr9;QFIj${w=4Tk!L`j-HwG(Iz2;O30O`$5ZvA9kJcsBW7AXd>s{UPZk8+wdpYc
z?i~`x>Ax=#(ibn9WAe0W7WsdQ{k|4+IDtVC*)QB6ReKu_pM6s4#58Om-Vp}39fSSj
z*{N}INpTjMbq80d_F|s2nT=C{F-fzVwby59{;}5YLNk>jB|&``j&H4@s|G@bON@EO
zr;VNp&l~*IJ?!}wHcpw_AEvk&HF)Mk5)0B1hcKdUhvW*9J}hqD1MVfZTm0vfti^FL
z@gAkMf(jyw`b0yJIB1FpF_z;ckw@Cr{I4%i##AnO&MgW^1sF8I5-q_Ret$zOFfRXP
zYR#@iO4J&`N{waKCYyzs%v3R0DhU$_6_c5L5nSHLz`}HUt_0qJYSm(TwC`Zt1lHm3
z(&Rnu`YEo&>lVL6gix4HEO-y)ckQT`5;G;^H|CiF?v4=Ba;p}5B|rpzjDq<C2BGa!
zv4ouvD=`v{xUut!zoKSKx?<Le8TuJ4MOXCqx5*S%gr1ekGcMe-Y6?m0iH;^FXR{t0
zEDSD=wWN4cg{Ur@8B70W^ty*GZ7|tN>KR5kitjX@@5fD6-8wnP<K!|6pA6Wm5k?)7
zNpABk+s=;93f~Nf#-Pf0vhzSE*QuQiE3vO+Hi<S>Ke7me#6L?OUV;!x`laM!SabND
zvOEbE%YKXVMF}9I#T~LAS+fndN|tXzsvlDXG6GO)68hR({fW9}Gl*I?7MI>Zjv6&6
zjE8Ls8dJswg~2e8YAvw17&1KI+jO7!Na$>zinBu1C)`cg`<CABJDY)G8ox13a$=5<
zi3W730wV{B>hOlmMz0|{n(?8~*{>~R2TjP;2vc5O?ouq>g?MnHYHcFoEn-K^v=Y@(
z5=}b$>omR4YMzk>TykY$rLNPPjr%h$%y|Z)GJDrF?D{cuamoFouHY>(@aiRlZJZX0
zD>{8m-N+MuAqPld_XGpx#deaUYGmAWI%W2~)!Y)HD@52aTcLo<c}eThirF1M1DrZa
zqtlb-z0h(~%Qz+ZhsD_lUKM-GX*bnta1EWaPqxb@fjCUPr2}$h@Wg#Y3G17THXUs`
z-VrhVIP8vXGVB6p9?P)G9aGMT_4&{rGTl`#)<DQN95CQxY4J_eqL`*_++QTW+9*Sd
z-mw|VZF+O!Sx1KW>8n#}&K8rCsu6$Li`j;caheC^i&X!Z|Mt#VItjm!(RR`!1c@&u
zb&O22cDmC^sT(fdLxGXbnv+uUTQ$P=^IX1u=sjr~<VALS;x;5&Lqh=qCGo6G`ZI;~
z<GoL9--*0xnLKCo$dJnrS>iWxkZO_nJinI46sHtsE98w-6{}EQy_Zvln}%VDRaVXF
zbIvJ-;n4+ykC}PjN`7EQWu&=!hF<%R6^hA)|8$l*MV+|z%S!xF99vc%)tjHJsr8FT
zf4bqJ`ip)~jY9q&_y=1UN^I6ld<K=zdiew0sKiBg)G=nWmrkX>eZJD71wdiuh;44f
zX?7>)Ki9cGaYiab_d%m~Plls(yE@x+kE)9w!_^(3n4YNE>!=9_;EHZaf1ldx+(iqf
zMVc*U%E#wY+vJ`hofu^<<U`i5u9e7EIPKNvfK{@tEJumAamDPPqncfu7Ucb2vl)m2
zzaXB&36tbIg_4!*RM<ljN{<xQxLVUsbBNA9HXv9?=K~4!u%9R4$QxP-1d|N+3aY>=
zceu`_4TH)uMz<}e6mcRK|FToX9UhjFMlpb(D?s>FQAW@;#p)|Sj+e)8UMQL6!-HNP
z+OY=7Pl(bd@r*^at)SKBov`VJGCq3nuPT1iJYt_w1~6}I;@ow9Nb&^*P2}QD@WO1-
zK-{Bl7Uha}b`$WE+_-oCVEGQ=Z*_xtexW?5N~(c6hRJYQPN(UzOWaH6hkn`OQS3O(
zw*5Rp2mP^FhzcaiI1%TTuKdfK)n>Oy<G>timyT-XAZTh?xt|vmMzZpHbzf*COIzG-
zA)#}SHHCb-p@DMcckJcj!%Bi94g20JzLYc~F_>{Ou3^e>EBRE;&Cu;8TGH_A`(yD`
zNtaGz2OUO_K8<35j8w;EXyuGSJ>3zbsh`$P8UzuHeOrqt=O&b?%?)Q4UnoQXjuO|j
zJ(hkjUPc-yKYOA<Gk#GQx)nT%*$K&wllFKeE)a~-Ets1q^vuHx^^igY4s1BhV|?_M
zlViMbDiLLAYgV84Tj6iwRaiwW)?DLTVrOgTh48f+!8$bU<5T(G70VU3zfM%G1Lzf3
z?IpcSwrdJYve9$$KXMQ$H(~k;l8MgJ__`JcIlI84z!tgJkq1_~gL7UZk{N^)yik0z
z`zee>cytP6qw7S<n6tqKWUMjG*uMafR3IO%p6stuA{T-^**UoKrL+y4C-!Zn<YFq1
zE13GA>F*H^Uc{;JD{t!Jn}*+QS6~@-$KR)@AL6E}+y{wxIqX_Jx^_#rq>OJ`5cL&D
zQf$cg%QVsSG#rsG?Qa6FcngCRhEJkkRXZzDKy{Ae>D&<1<~3t5KDR^@9gik3s`8|k
z&hv!W4Jm!W`Yw?`Yty1Nb5!C#bXx($?6A-JQ9uC|s9bo+@c2o?DZbZoEoB1CxstvD
z2Tcs-yFi90D2PP|*n`eYA~`>oEsbgK**QljUX&RJ>o02T#PPGXb;BCiy02qHnZ%SL
zz0UO<FH#dnqX%E(!Y7K7L4>H8dwiI0VVC3!IX}@h4s)-oY9Aia1Z4Ei_si3pyG=?o
zcbIOo@g!>Q-GH}!@O`Aue7))e-rvuZAN8%h4v9V7hea{3cs(D9J>HM>mA`K30nhlK
zpXYoZQt|nHUK5#bI!(bUCAPxdKaafjaZ76M#thiB)<m4;T)?%}zgF$gFSyy#Qv>sM
z^j^3ewGNhNAEK^+O_f(`nS5~Es>Dtg4K$<RzP|0dK|u#kYmm3DnS3u15j!_VwaO1e
zKV~SkWXNhtx{HpBPe|X^kCN{o0-xvbfhSv0d**R+`AZL5XD{Z@w_n|M9{L{diyfib
zf%*w=r-vyVzP{#;u6E2OR+~4iE#S6AOq^WgGmDmoF)2blU^zbBdigwe1!a_$#tF(?
z^|kvvUgXQQKgvCPWy%f~jt)BZc}$f91QMLGXBN&Nz({v<Hi>_EJh;pB5pTc0KRDxm
zd5~M{D;S$Cjj6xuV_4a+#k?tJxWl;_qrbL#iO=C%pjT|Lbm-8-xViL(dn(G)md<<d
zPCGrXhka5@@`QidL-(kA9W>Cx0o;7w1CdzX>t1auH&|17(Cou+j+(-KO)vmLJB}Fu
zp9RNe^*E1HOZe1#3b>c_Fm5lr;kYRk+iOU=_XZz3+D+Em2iG0<+wDs0Z}m9ub2GFc
zpueYt&jWJW+o~tVdK<_oLVX`$*`HHIftIjs5wCl59+tq`oXptV{IO{v4B*wo*a4Q$
z(QSSfIIGJ~1i5ucB_4+>aGQ4Rj1IW12TwI9<<(k+<hu&*uWMFoK8;pu6{#l!`>@4O
zz9qld@6DfHhiVg6E1InpK2&S?%(r#g*4ege*V)dm)D1i)t@n4|!ueEtKkZa`ciYsV
z(}VZAj9s*K%PrmMl?9y!uUI1x+wKtD89YI@FSNZlEVLQ8E)5ZBS77p3t*x^!)$7qM
zu^*dyc3TBLjyblRolo7qBwC_2y-}r4-iivI{zv?29Xih<A|$W%e;BcUf9SWJ^h(V>
zy`OsA%AW(Z2gAQzz&w^feVuxR^R4%NIxqo3@*Rgwq{3F^zhJ-G0B@&;Ug5mv0iy?q
z;Oz_T8)vt+{H@1p9nCLuD7q!XcAaWfVVn`^f#6jhg~d+<2*@w$0ffVAzOQS*gT3~r
zGPLofHq0z@hs#6jv6rqk2f(9=<JaM0*Ln1@I~e;5F8=2G_G;!J<!Yt61)6+!ttOVL
z`Cm)@b7w=gjS)&@W5YF9Hd{Vm?e6RNkha9Ynuo2#vKy?YF)ZJgb%3P!tf>xJ{x0`2
z)e4kl%;M_h%P(JpHNIPR$Gfa{t>+6E^d`7xttnd!S0j!SYc{D;)8!{u?$e-vW`W0k
z>vh|8_NeFXb?g01Hs9yTTHUWejjeM07f}1)Lfw)J?^EToudERaF=@F=`|TaJ_%F?!
zPy75<ulK(!S44rW7wqAD$tcQ4d!2=_IsENQ6Xo(4ND(RZ6Yc(QJK`dIt)K8+e7=yH
z^N4r_M6~-{7bCgMYAO!xZFq;+2R!M+Rc(<S5WNxkukk&)e!ZLUoqwlpig@+hvc>SO
zBlz7MKj5N&edo&kI6@H`;ql^9bp_h)e*NZ%oPQkgGKK~GK5<3{QO|KB*4|92Wch3J
z6W24nn7Pf$j&cfX%Bmvi6+Az;rzM;gtmlvVJoY7T^_Zaoxh490gQrz4t9I7b1u~HM
z?LPMZA{hzN5-xAq1oM`{_3%r!GPZRSxhp7)yw>gOAt#<k6)&^br^m?_G69D#2G7C#
zE3KYS48YZU8$FUL;{e{1_VfG~&!?eRt*vJ799hEVlaW^#-^KR&_TzTewk6wvf=7T)
zHLxr5_UwfI$u<K2RzgQq{dDCu=q<%+eHVP$+Lq2xZlkLoO@TrDCu>W3w`*8yC$Z2g
z;WX>!|Apu!jmN5<O<w~tXKhRAx5n~ZmzJhQ_^?x8d=}ek+pPFgSJNd}498o&lG?`e
zi61PmgEQgITq2fOJ9Terp1U&Z-Yht>jX1ldG^Z5xd#L$8qt0#z;upwrul@((Zrc2v
za`Qh>2$DS^>GOXe{8Yp^8t$H0xhMK(Yh(w@xA|Cl)wlOgTjtjB{f}#uVoM7&{|Cwm
z)f}Qi0pUAdJI~MYu*JV!!(KZ<-=ZVf|6e<bcQUg6A3IV}bri5J|6|A5twb!<@446i
zfqtDpmG@2kHyV0oXs}hznYViPQ?IQrCcx5*mDjxGd2x^bLfx;yMk#T_>`&;groh=(
z6X3%A=~7sATtIfd<D1J?Us_*wR@>QMk@MjWtRh}%GI8Hd)q5pZ2KMufbjP$oJ#i8F
z1Kl*q>d4w(?l<Q6fzR!)59s*VMWOb23WHC+_bX$-J#>6u)p52!>cbDDKB@TVS`|
z%cj+t&tn)c5qscqpc-poCZ2wmbyaCfRK(rZ!gjp*A~w-6q_-w90Tm_E-$^ki_;Wt!
z2<;FE{%1{r@U|omAzij&*rIQB6Emm{NcsvjI9+p`ZiNZ0=9EQ7DUpmps5{4#+#ai@
zK3US^5~H-A?_vWc7)9X^A@d^P7T!KmMq`L`rc<VTUL~rETmo@l*gu+h`YPwdijBn(
zT&}1>%r)T`@I&w(lccXvXfR<YIa?x-AK95%yJ@^yJt^-ONmKUHhI^1O4x3D~P30kp
zFg>*39f?mW?vSWo=3xzg&KJQ}ow>mtDhRt7;UC(f?lE_5(iaJ&IS}Suv>;_f=Sq3s
zP5|WhYWmJa-*yC7bm8{0XxtbqnUGj6evHV9CL)<L!*gFclLu&!5|YZhu<^>&`r_^<
zLd;ME(XFd;s2Qm8RUguiB`SkosYj}nqHY~Klmwd7Wk{Z}O$eEAssJ5UB?Xz}Ayr~X
zig$ra5<vG3(VtnVykSmwWQp3%-zsU)eF+;ZrHN{n7IM_7(Tg1IqBP_Be|4idFoz{J
z2zt7{cED5Ee)=#&p&3H~i7s9AE#Yep#E%42bo!)L5^$YD<}(zB0PMXWq!v|&(n9Q#
zU?mRCiO<D|*12>k2U@`^?wAxy`!06%>Ih>2;%)ocgG4ebN<t7rBqi#fx#F4Gz|;3C
zDGE4&L$0Nz7l%^0ngA3ulLVREp!(dvrtlQ!l%w{7Z~I_PQ68p!(yWCyeI!>@!&>VA
zj5-q=dN<M^SX}pY+m4KBCGj$pJNMiGI{$U0`lms)f>d0uAytT~x{tG1Zq+qlyc(+$
zI)sHFV&(l9G<BbBQfrKN9K(p#A`(bxzVE>;h=P$uFKN*UsSp%t{OVw9{07mhwNhXb
zVPQKEeAOV|G|*L>JT(mK6d(`n%X09wc+v|!mNHKr)~owum|*$pUYe)b(t9y3IsI21
zOn%2;(cS12AHpxCgGG6d4DlRHSxHS(=PcgMWHytG<~$m%6N~!**2&|wD(WBAHWI?^
zfLA|-Y>hMo$Y7$*L+f5u(G@tG_&d`+3>#<?`gLNd6ccPE*c)U>{CZ)>QQnKLp4`-J
zMRs~!wgpvYg!t@@)DR}hd2qc?s)F4z>ei+kfQ*=>_Pe^I;f!Eg6t0S^{{Amjn4422
z_yj=Fn@P~fUt!T;vS!qM#gUl@BjO)uqN5}z=vVB~l?a(>96?J<iW#isUd{*6wnG+_
z4a3QhM*~-cex2%U`%d2j6`Mq7`Ov+irUMM?&}-$ijGXLZN7HUcd0*EP$Szw1L##r|
zzbMR3x#CBF{0eV{VUreTenQ~++1X^BH=5nRDvLv#9z~&|lLqR%vO*gYC|0R@7i}`b
ze|CHRR7@A)+%|1`?NPeGst?7Whsghx%YD1gsz0i<RJZcdHLmRX$EdMEkbD$4G7K^M
zMVZjgFAFTEp)2byJ8~#`8%9|?%O0#=#L1toSiw_wm%@I?EOw^w2KRIy12eU`wvC=y
ze)0h?5`8C^VspFqRx>0nYU0*fd9T6S_m2b`ZXxs#F7Q*u@Y{45%jC2YPDgdwAvn#8
zmeX5DnF)3uoNh(y7G=4)+c-T%e~&wOxZZx#pf9LgP2;7gSkKh89nF{kv6^vQOKYu`
zW{i?7Iomx()vC`s)UzGMlZA&4W-DJyZ#6KfGi+|;^t6J8WSKfSI7UrkiMb!OZPP_U
z7p6ZtI5G%!t>V#XHAL;x(qv&*8xy59FKqSnh$%x#onM*CDZ(}Rd9i!o%#%a4cX)b`
ze7@^me>pY7oe7EJWZv@N0VoTmGFt+8-rdBQjo%=?Y3|xRH=bq(Ah<tNiw|CTaC#0C
zlh8zjR0C2@by_&sZ4qiOcaH8?mW0UDh+3X^Pk1j9>qCSeMW?n#mOSU%-92~T9wE~O
zz7f=TIaQ{wwzdX7rpz8SSN93Crx$OW9#@(R2(-TWMB35ho6W)LF(I5!+lPgAPS2>*
z<+N7vSBG19cd<6oH{d&x)g#{N>gpEKYD+8l!bnWg#>!IjQoE;T{YE3V=cm(<T!_uL
zj~;1{6t}x;>AXC_RyDs70B3w$^Z*36j%W3ZQQaM3P9d8|r`wK3(?U7G+4uG|-s3IB
z+zY4M<0j9rf%V;Em8sR>{nLYE-jkLz-gDmkBoy0GUDodx7LmY_IZrlUPS3HSwzO9E
zXNOz7UA?x|>O^|IGxz$p%f##H5$~h>-3{K@(dlt3(2mp73Dv8f-S%vvu(`euVO;+9
z_WE{}&feYA_I~HY|Bj||xVidn_I2l&7r4H2)cVpnH0KH2xV*_o&@EfV_Hp;Ht$)6}
zUiEQv_c-I%&RAVSpqD#)2^pI5e6_o}UiEF@X|2cC%~*XxNX);x-#CH^1xA1_)6dM}
z<z&w7tTJ)goU~<?+?C!GC+#{Gw?8_E32t5wS!fUeEr2)JFHcj+<?Cm~!)mR3RPH~C
zTMWX}8Lw6tl-p>eK<O&T1xFeP`F){j*u|4$;F<LZl-nSKz2Q?Wza0LCdwf{k!5*Eq
zL~3PE``!fQp>yP~yYl!j1?Ric#m}>Tq-#6p2{(Y=cLgXa@qON=PmFg0I*16|1@{dB
zv-vYRYRJd&<p>s#_v^`jc()v3SK-MKSXS(9efrol(URe}Fc!E`My|ioAi%%Y?A48#
z=fLZa-WKsVIdSa)N`E~=Yo&L8g!Z}t5;ce1P<DqELm#<!_38Pb|M3E>JR^%^j=JXo
zXlA*x*HB8c6ULv0Fy7X3e>yr9=|x#{@I%<7R$hh>{9s-@jKfp(I~I3e?a7K9sn26?
zCui*b9up>XA^zqMP<XAl6`%VR{MY*s{@0Tu;Q+0d-IZNya2Lkk^HrH(=K+~_zJ2`7
zFKb>?h%`+2k(B+ac~Kp{FF}m?KdA%X9*@3zLh!wRIeXp1e?5q*7fO-?qIds39e_5m
zF?MuvFfp)xZ`m4JKr=ECFc7>qIN4d4IsZi&{zkP3bpFMd|HT*xn3<UfSQ!}p6h=0-
zzxu4K1WZhff7<Wke^YV1^RY4hMf;93v9rG+Z$Hp~u79=vzP^t$ax(m(d8cD$Wg%dF
zllxvcI5-H{-gNy}2lE@o%E1i%hxTvT@9lqh{{QNC+CTJfR=&ykBkP@qk%5urkFs}#
z>~E8LFYoR5iLk#hz4!kxzB%C?|6|~r0e@)U#@{gLf9U?@`&$Ml3-iCS-{t*d#Cy%c
z_BNrv9r>^E3=Gi!wEyV*+ad3ccrWjPc&|C$^#4QmPyJ7z{^fg<^={)o!F#WnnVAWg
zSULaH?@stDEPo`wY5EfqhWA+it@dBweUATj{u|48@K0g=i;MA{{a?zzR5B5~yWySd
z-8~F{1pTAopZeXX_x3w0JM&vMu>IA4H}YTXT^HvY{x|l9|DP=8w`_TH>O1Y9!uD4-
zalB3VpYk4nKLL6V?*A6(KX(5c`Tq>Q3;l2T6ZJQ*{)zb8*njIc!~V^Lf9c+6|5f&W
zYy9on|F?cm`9G2Rzwtkr@L%2U?f;kWfBJ8Z`E&dIpNzlg|H-Pi*!<s3&iWRR_m=}6
z9s+u0cRLdTdQn>&Ct(vuBL{OkCtHWV`mzSrZzHni)`rfGN(MI4w2CGG=eHp%1Arp|
z3-lizK|xzL0xbesMs^kgS|*P7dy0XPfRmM(K!<={!pXqO+(^I%U}f^AkY2#i$i&9!
z&mjG~uJ_J65A@r@Dm!r#bAXxCJNp}@`ro~Gul!LXYW^mI@hyW06#oVctj(?52`K)u
zj^NEaM*?X|XnG|l6KmBsZxZnF{ri&k&i2o2rA=%APG*b*jErx$ApKuGCIS{F&NtzI
zci;2%?<N5YGdshZeQ)ZWOy0d>W#D8YY+__<Z1S$)?-yCOG^Zw`I5L4cR&CY96OX|<
z>!r6})5N!G!30}JHSytgxYz4$^hNcx|68>j-tjzAqeCePsj_??DoZ6h6KZw6W0py-
z(K=|q#v)cD7d7_cz&KTMN;BvymIj=GxP%Igj+HbJ?iJ=EcQRI!`r^`ejHc&a6&h`j
zs$kg<WRM?-!O+2=(5bi97{RW5ITyL@K~U0YtWR1$K<Clr$+9K5_!1&{8mVb1aig24
zK|x_lDIt?80X)^njg$a%ehX`Zue{V>=2ttX1}2xgAiiW3<;qIP$XwCG5R;RWcl2Mi
zBl>+Stp3%A>vw$CF*5yobpt>7H5ie-mf87Ht7d@T3w0P)M21G^b%M^XtE;!yKP+Rl
zb9nr^JYsYqd~|d=DC5#LE8B&v>$VQ~aUJ#wVN{qVT{;5Dwj%*-cdxFicCEj9Wv@>D
zl|I0)#bodNQ?0hz$%!={<w0(*oh-?BVOU1zLkX-jY#OZe2PHAT!jMXc7k<SeH?E{t
zWaH<Mh4Imzxuwok5Peg#EvrLIbJNE=$gLBreN$*Vdk5B=n^zzx&=In`|06>siwg)w
zatcr<*jEH-$;SE=)mV}h7W)h8OR9gQzdusPdDKz6Sd%aMbkFMO?<>7ihv2$AKVQqo
zb^}7fr}j!S%ms}PHPSdGzNDAzDCE>vK=_+Al4o=fU=0<9<}1J-76?sg4lqSne~#jN
z{d}L<`eA=XIz0_o^!dw$9EcL=t5`rg>0+}wRw^jP>n|)^NY&u>fCo>bXa1RM{&S#`
z5->q{c;@$RX1M$_6tQ1>90!>32OvT3@9~4R7dUHLYaU<}`Rj<%$~Aut5Qr`@**pGw
zQhj`MdAGeEbX<!oOv587g+jdvL+d$2`xRNfmib{%ltrb&weGo36>^-Zis{u>zPD>~
zesOeq;o&U%8JYQwB-8vr<Rtt=hUR1#WRT9+*?iylaxK`m<9C^?tgpu7p5%-7ZTqfi
zg0CeD>kR{IaHTfXXB30vbrBa)mIviJBvC&hu`~l0R8q%O#)?TB6fD1BnuCXfTS*c1
zoCr#V;3JlKJ6|;ZrUzNTR`EqaRRsR-3n82aVTY#V1)yAY6S*19mA5{-4?00|k9T|p
zeFs~LzL!r~MiDy$^Q^Q{XUujrDwoGY>k@j6$_S|4C}xfP{{d1!t-or%gK6}|5kVd#
zf{Q~qs|yO`ttD)=Zr;araNK+=H0Af>dVD627CxIb2XSJ$U{dz5y#C2TJ}3W94<71G
z>)ZZY4qD5IH#Z*%j)Ro)HbN}^E_nU2%TKKtFEg*WaN$P31nWcOxD4SYM1>VrpfH<_
zDoaoUt-dd3zm(4M$%WAw<$!NZ!|86dbSUClxPceH5j@WZrSm&Om=9kVCsz=?VimxY
ziP5)^{bK&>6RQC-W;8b>O%~VaUM>>8**$2xG{u|C9=+m^KbhI}8>vxB;T$xsTqH*w
zjRg%IaPNH2$OoLMMFdFNdA4`H4ZDjL7uZ_E?&92%)T?)@5;l2w_7s@HP*`;mZ}34B
zFX7Nbe4<3#W;4-?5jp98IzPQIful|8WJ<qr2aR1-bLj#x%?IErJe7F}T5%vzme8>t
z`7*c-+?A8ld#moIMp2XkBEnf4aOcSR_QAaw>%P`lU;BMEjb*&k=SI@F09EQ>Cwl7&
zp%5SIc&Pc1+oeS`T}e=R#NlbF;{wbplD%Hnq#wKNWSr-3(ejR16*AbQ>b-#J3}-0w
zKX>%`>jDNK*o>`J?q$|oK_i2B=&G+Ppu|(Z*x}!#{J(@Ruj%iB5eU<l&nwUh2&IhT
zmD2^<$=&SsocKG!k}n3bl@g=-1=;9)!S3RzBg+`l<(d2P7|rg2rq0G*jb#x23~TDc
z^%^17dVkT*Z#e^9(9zb&pKKz3;+fiMrxBkuYo5m61+31YCvTs(1uHu>Qwa2rAJX*&
z^eA@J7k2x#phTxl^z}Cw;L0jM@WnTGuTPIt=&QuS^~GONh~ZGH*%v3Ome&oS3rlmB
z!=1CjP~Bhg!8E)SZ3$=GfR$i>Q-=7RgYw3)V`&#SQg}DLzI)EfjG3w>oG1tVj%>yi
z9P_3n;eV*DP+NW^jG8XYtJ+w=rfG*m2jP}_{>9Kq4z(_A{bHbFHy3T@*o7f7%BqKL
zj&u~Z!?@sXYle(~lJaAE<?D5jcD4p75Y;ySqnrGe&fM-Xi1`2|@N%_90xjBzPRDdK
z;6<6Uf>RLW$HngM{PgXqY&_xw`Mg+|k>sQo0XPGqapk(1LbXPuob+P3rWa_cDXA+_
zK;Fbi{=FJb8B%CLVz=4OCZ^YvCcL$-IUmV!!%d%OPk)jn>}>EQtO<s#)_^FE>g-Xu
z!4=EFhp6};)Si2m)kV8oE)ryQ#kdTJ&zbkt-_4^E#VCRvZtguL3M@GEcq>Pz3&h|D
zC<puQ8yO3<RD1h+#u$ULGhpXIwO@+ts{AyZa`6RKPKVfx>nT;WhJmhp<G(fK{UQ(}
zwOmE#{M6Fx^~(#pAN*_ZrNSEH^AYDBtvDorM78(Wt|M)~4GG#}zdq}4kD}61px~Mm
zBD<zaNaa0B<pdJ0?%hl@4Tn^_m!oA3UnV>qot~^TP#X+wqCagp45qHOG#=C(2Jd5r
z*$)nvXlAFRQMGDcCr@Eb#m_DJmz0zBeOmND-?u2|kil$<9*m<=@IlX`WW#QL=N`B`
zycL~-as7_*y>g0V9L>se@YzWdp=N{hT*~*wi>fSxa1nW5{()#5**`jv=<fss?^<@N
zQKgdCTl+|{DYALaKq!Kj33Mx4X1cz8NFNn(4O!-^3}PU{yR%($liqM$SMqd)%m;g=
zcmIl#h)Cq14BnI6VeaEwkKaTAzNp?hB@pSclvKV3L`gcL^A4!IQe`ZbuONfw!fpKH
zydRY_-q`NNDd_957MZpE2fQ?>c|${Ww}yn45v@&=HWXM1-0wu&ILHt!;ALNk+v03z
z3_f}o#p15t!RfJkXlz8+ABpU)$Lw<;H!z{OSM3@YH#yXipJI`hN&C+04s__3o_nXy
zJZ;?ULX40(r#+XWYPqw27MyQAV@{2M{}<5^*0sqKta=P<&9CzHr8UEN{DXv?(-Fy2
z-RV0sle3S@`NS6m`bw>kO1TVeV%@2byi5*beD2IG!FXM33n$$W+MhOu`WMP0LftS&
z%l4Txmy-o_5<kn&j7`!G9{J>5C=Y(VNSM|)B+nx{uV+^-C4E@G@mQS>(B-65>i>S=
z0=TI)88lBVmTWQu-12q~l01hKLXO&>bDgWysK-c1jsj*NF2zXyB!uKPC-$^mwe2_A
zsJJW7#XGn4fMgsj6VNYI$sh9=kA*5dT<munSLAW0o7fF&E2dv21JR{gMw3_Et(k4Z
z+!TQ;0_Qe={W3Pjq&0U<0U9bZB9-cVD7-I^5q9;@uRCQCHx^R;(7^|KUNB$bYxxwe
zL-ABbV{<Ka0B#ov*ugLt9u#QQvH+v1mb_vqu^i9mJO(R<aR{InY?jjiDyuv_WcPMi
z%#p;aH1B<=33QkaEH7)E?Jpyhr-0pvjGS=AxzC2pBL-tq)TAW<a%pd^B$_~9H5WCU
zvMekv$Kt4emT3=d==|Mps^ya|ZQYdc&~!=l*^z*#v%n4buB=OP70T;_@YX>tK_$Tj
ziOrfCS@HYLuN}#HVFY3rHyS<k;WevR^5_$Z_T-dNQ|(cS-oQH#n{Snv%*|G}bsQlg
z%yW;<<uC}+V-&J+lS}_P*dZ=IT*lE)x2ofCCK%Z+HKL-+viM%rPe4<K37$J5jdR4-
zPzl&aB7TNUnq$<iEc_s*D5|tkuy$=v@j)dpmdi8tU$94A$${`#e!0dD7T73=Dy=Q*
z>I`i9miCtdG9U!s3k4nV(TiPK9Fu<H@WgG*9<9)(K$<gbaujgy{a}odJB&72vePuJ
z2(($YPQ=KHO4>(!mZbz~U(0FReym3pI-K-kY}&#&PC(pCrEYUbI-{}qc;FK_A#At9
z6y1%FB+p+2rh5yO@@8i@hz3HRzDmNdm0MPZ^e0ioDBjs`g<01MCb2d?PX2QZwwd*6
z*lZu=cZs%6gnj~|w%=#brOLS<NQw~n$;PwKQ=S1g7sLOG1|a+3kRhld|5TEYKJjOu
zBE8eH5x|$+6B^J?7F0mv$$0-}Rbv*|Z<T#q!wV~c)IE{GkLvL#fhBFUo$$*@7l1~I
zb?J(00j?A#cClV(Qc!PbsoP7Al-drk({fAs%7Q`dTY%&;mR9bVaxv^cB9Zz9$I;;$
zO?Sxf_ERpw(Woh<hCtst2rtUoHD|py_~VCe47_P<N5CjRQ{XU5x2ux%u)FrzOa@Q)
zmKo_)#y;dSxvZ0`Vj#!>{j34wD7W%xV!TkeG!phy38?+y65K_FcR3jnpQK>G(0C4}
zaeY7hbzg(@(Q_k)Z<&Pb9!MRNrE5#1=DSjQY;ei_=@jmK?_TAXr{(2?+Gk3ds&11H
zG_{M`h9yLq;0^9EaoXgGR@<i^Y7NVM21?h%<!9h3_@RMJN`N{E7TbKjAKsHXwHRmK
z6&b5i)WiK{L{pwOFlSIY2Q@wU?~S771)3A!WRxeXz>HK-^?D2?hu_)HCWgwy3=kO6
znbyO+Mdf{1XD##-dD4Z*Hp&e*QY_l^e)8#{!cN3+)NConl*TX|>{|EH>yk?NnVh#1
zd#lyKIzl2%f(-!WGo!;HTcHd$RgVkF1+ighP@aMwv)uMl8l>d&Wf-`h`Ps%q3RQzw
z)tF@V1HX<2N<V#{xW)3NE1WL+8V#~Vtt2h&k9*Iq3u=e{Y-iw9clGgaxXNvWb>(H+
zRKu0xV|XPbUQGehdw23o;i43ClI1VxG*L;%&>Q86XOOtMN_eb*AMXT5K+(A>1G_91
z8|8Kr0Ig7vs&KxaNu3OTJzHny&!1gKbLMX%v<}&*Ldi{CJARM!oi%vOOWcTJOMJxC
z*5X}aFPD<28l1}?$#K6hxPi8iI8ZlGn#yA0ZEQM>dd#dOL^N&%84}BOj8!miDN|9B
zTC)u0>$+#ZdO<8H()>N3Q7yc4S#d^|DS=`_aY-uqA{`atcqg2K{)Cs4kXKP$u6}{K
z4DezDMV;8LB?~xI)rtNtg%ASekvkE2rQ8Dt@?+8|uM>(W&VE+O26H7oCq0U+a{&gg
z<`u*4jB#fVw_98uJoOwfUfLMKw2hG|(2K*pfg%%L0S}=)um*TYYrpYBvjvMzAR==#
z8qTPf7Q$`oE=Gi1e1??4rO`(cXvXoZv};%R_a-T%`fvkyj7UB_&hQh#7gnFCpaD`>
z^rCooP)w-UM1(W2w5~(mWLz((e;DvKlcOpFwn5bo^)SIPBtXcBLGRJqxWcE`O$0Hh
z%5m_8)fJ9MO`kwM>uEX%;vXgI*dFi+^B`ojP1a&}YPB|y-ure3T)bmGk8zRX#Id)B
zqbrr%tL-5(CP3$7j0dUiVefD0F&x;PqtK|}^e`A?GggaF{aT^N`d{piIyF(@YL;eB
z{Pv+j^wM-cW4*BBX9FVma)^rTcwS)$2tmsiKlPQT!;wJYNP?f$<iV<%a=?=g$$t}a
zu2jr)qsD7(XJV0(6_2pl^6IJCIn4HU<^Vhn(maVjT<Y!XF52;r*sPdlxc%;p$=#~w
zF3oZLgJ3z#fh(L7yYZOV&X+)6Ar-bats7`lkc@<2t<c``yF5tZ{RXP6a4k=v=h*t^
zb}85pht|!DQK1%{mp`32^ty@TF$-1FX``tms}3R@n=|WZii%LAzP~z@HA1%|J9X*0
z_na#7!Z8P~^x0)GYj~*^#v6Vj@uRyvaM~obLQT@2Ws}BP6kK(LxX+$rfb-kWHwsZ+
z?_AdxrkeRtq(aAzsa+PQkgvxUIrpKJS6SV?z;jU0I~~C5W#ZzFhk8IuE9-PMc38x@
zU1^nd-jI@uRsebg?^otA9Q4ejFf!@>_-XK`gUePgzCw(&$$6Zz&BH~)KOLikw&H%y
zC=zbba;=2-F^LW^>F+AQSN~knBW|?L*PcP2Bin;r;U4@o0+d8`qY2c+bfeI!Gr|5S
zCy|(wgL~O>io-K*RO(uzEx=bzkSK5v2un6pv??dx?llhY@#@DQp0^_Tp6C^E!%m#J
ze<JS^e*B<ib4g&xthRcgFLqr5;<&j)WNXPVKYSRA12mViAC6jbg#f!!00P=-Kv3lT
z!T780;o^HCf|FJPH~^Dd5rW08e|}DuC;kALL;;SMpA_ep1va-mgnsX{jx*Ho8ZA_A
zal&e~alnSfOL@$hsWBnJ)w-xk_6?=Mr3xVBYhue<Rt#m70PgsU?N>WX50(NBK^qUa
ztM4!o6}8-hYDAyjBd!jAV8}3oIEbW7P^`aw{1SVV22PsPzOCXo+Cb`eB*>5jf%FG@
zEak0vsg^o^WIKBEUdp{mry2?iD0cY*4dq>fp)yMPoYR|;n*KG2LH*Vk%UJh-GTrK8
z1=>7G+QsWm5`sc>f)UkRx3M+Dp$ZW~cfjNfJCq8zS@gQ+ul@KjiFMGx%Ztz7jo%4H
zp=XZVBRZ$D70m0>jhww`B*}}*-r$u$ytR2r%OiS`gm<=jMVc5U+tK^Kf5>13T1jzL
zmAjs3e2%&c4uak-olqm}(HmP@NsKc+;Ofv>kQW&(1qy6))4_a%eF&7_4@=yoe#U^~
z(ABzx!hy^7<gO!&+T(3RSp@<rM(7i=(J*fBR$wBL@ZZU;p+t4Lo0SCp@xr^mVB#_I
z54R&{+cm?X;wbo1R=E9N6h3kbcqCt`wM?Vk5c5e(FC`IFK*HD5(jF$`sChnkHg^sF
zRL>HlH>V>zbA!l=l>hFc=mUO%g3{2bzR4D{+=?gGLBoui6&CuQ)BiClw)k7TNEtNV
zpwr8@kx8k_iWZY`-9sm+%PRgS9x(ObU{|@H?%r;{5*_N9=9aO18IAS`wz4e~^~MnJ
z!QPy!9nT}vFp!)<Deuf&of%$NX}SK;^`A+3fV~y<^3ZqwAZ)-w#6oH5)1sM-E~x5D
zb2Y7l{gM+f_uwTeRc$;f_PU?us@kju0M1R*T5M%q=(otqy+-UE;f>Qk^TpTxqu|vd
zyDL@|g)tq^$`D`hG83Fc17;Dmr!aD6lzoFv3tWRbQ_=-}&By7-_%QG{LJ3h*A49t@
z4#~VB<HJRzZ*6{zHhLx>g|qtAT$MI6tIVn?gA!4thdH~Upf%-$RewZS0r6$yVIVuK
zH9-BnIE;N9b2F(g_B6Tnr+ZLur5w|yLm41Bx3K#|r3=fDh{L<gQH0=YYGv<T<z^wd
zXdZf=m{)6<R$8_+{)o^eq$^F;V}uLpOq@c|2&{pu6Um7g6_PnCC!D`K<XoHE%G(!a
zc8d>;J{XE&R!!UM?`O}l`~vov6_eJoG~YY$nz<U2_|hHFNR#n97h!ZldgWu$?7N+`
zp!<4j-vQ5s@Ns5@f(xZ+ID=$5nk<}4mG6LAEb@GF0i=6dM9@C-+$XD6BNO8&6t`-_
znZUCu2lkv9?nY^Q@g?(k)?Q&<U#p5m6?&a-fAI5_tdXTs!0E4JU0a{?3Ci@XIm$>Z
z;WSz0_8T^?r>fGmZa6eW(|HovAtgkeFNkKDtT|$9NZIa5h(6TI_`r_tO4n4;he0pM
z>f`42^}~>Hqw0H*`9X~8$%$~BKAWi<!C;phkn>mN)G05y5Le%<9CS}|4M`GXUWc7W
zGr^GGGaf>9)-yz`5||-6YSbLSbu+(KX?roz`f!@&;`S>mgLka>$f5X*stDO0;-|GZ
z(_C!v>)Svh3`3t8oDN)cIXo6;f<Khl+=+GKyY|L+i?w##-z>b~>DCN?PO@GITPNHn
z=}B-uF7i^X3$fUXg5(aj2tpUMELs{OhJR~SfOAsfL~K)4HGO#Mjn<C1y!Bg^vU*iy
zX%^-w9#!MAu|K!G7{!0=$&r-{Jv6XToSwRy0IP=(1IVj|ft*mn)!X>;mQ^SvyXIHE
zGsM7hWME|2)$GczL(Wf9%vg{+gNu~sV|ERt=Xm2tcy>0!`lH4SfCSD}2P7dwrDHGg
zr$QGkPf&S~=CD7g50lxVHGbA?JKnO#PCq^#Me{_kPS2ua{OrISSUSdM5ybq!5NKG!
z6IKlkwq`O^CK4}KhYN#6(*s>DzhT~)#!3Yj+Vz&U@~+ye93*5^??c)pY=B<firtLS
zK_sm#J)c@wj+i3eyr^3&wV$Q$5@mjm!QN5fVRN6S4ED2g)gJgvtU(>{^<HOa6_cm;
z-0;n|rNPcaIP`(w%&;0os$dC*2Z)!Lg8vebKgiBgDpMUsTNPNZ)r_?Q;eEV<$GZ}4
zO1eQaT+7~^1;2IXdbRqZCTW7zb<!pDb>WKw;d@30_5V)c58Z0vNQ4F&qLh(MNsrE+
zW1kek$_t8D5`TA89iM`xCQspii53}m*lcQBCNCPq*XF_S$}Y<spqyA(3PYRP0`PRP
zEjrB@=zq*MPA<N`BW%L6xP@lwV~sqIn&l>NQQ(Onw;S{%W#OMyR`vrY<b5POYwWSf
zp3lJE_w}WxaQ^v7-^S9eh!<zFHilZlVLzu{6Tzgj%>4M28~SKEkTJgn@UsSgF}Cpj
zZ0me0J^1;H*%RFF*rh-_*R5$de~=Hw^A5go_G{#JO!jsD@J=g<Jur2R^GtAwYL>aD
z|DXhQ_V8=RvjB2nYATXTY2Q9oLEXMDIuZ8y&Sn>tsRxNVOhvQMJ>!HQ#denQgxSPU
z=%}%5Y-P-;VHv{P35f^eGfwmr=$xKDDU{skae0(MYGm)-$7E7t;%lzk?Bwa(S@YiE
zm~1IcT|3-|W1yM>3=F20ym5(6KVQa@Tv(E!=cm853vRM%b3pP!qI56%Wi$?vx{>UL
z`UwA@avR96qSta#od%~?c*K)DMXiTq+ih?y_77eydkWLA+f|S3#bXVi^rDZ8x=xgC
z<eo91g`7f}Ita}Xc_Vc~-7MNY?E>1w2H{#yuyqHazHX3Zd#9O@RJy*wO#&xrPpDRE
zzwg@OZ;!s0iCQh5qzCsnI>vB*wElTvZGTOprvq$)8rkHj7~MI{k5kv~3RQT(S~hfq
z^jb9~4y>20(d%60kAXYhHpEd@Y+IECl$EuJm7v^uXz?j)6+;Ra`cEdzR?*F&_X;QQ
z@2F7nyVIG&I)yC5W|aCteV^Rnn-Zj)zNCk=vjPZ_TBVX^C*>pJo{QEJDL~iyO(`!O
zgM(TK_x!90ZZObD^4q)#M5EplKx{UF!1U-WH4(0((8YPGlW3kP;78i01ky=625S|=
ziO#1@xN@ccO7w4k3RW&JMS1URfz(jydQ3!|IQX&l2&c1J<0jJh-(47DQVqrTQI6}Q
zkAw+s7VliT=YY7o^cm<AzleP@ADh$b1ck=Fg5}XaL2}C_UJ_1Kg5jdiqC6O~ic;C4
zej$B+BIBjr+l<H_5#Ft7)8^CH6L+2^WPT2=jM4SUs!jbqlF4GN4rQ>oB_}-;s9c}7
z@4jri%w~M~F9Wo|#`rHggu6{Ko<A+|`-j+!$gDuCq!|5eXh|PzG9J__%HB&5&40Z)
z$O`)zot7iZgCInYBpVh{pYHz2i*7!nQpugjDutw`vNpRt2SR(b>4x&E=)DuP7c){Y
z@n<ix_^nMnwJTr5APbn-QcHrnlIv3)WPftY(uCmRtH)?4HJ(lm9LtK7IF_W18&5hM
z=7G8$DEDGmt!M@Pqa;`~ovVbd8w+-oLQdXT3Tzm<aLLr%9nqphL+hW@SOo#AOo+q}
zW-$=H9Ep~e6s~3srYq~b=tujtY0<x7<|+Zf@2<zB(FVScCJ0U6QfwC#qU!{^UXrGe
zNOasY@6ZfH=j3IEA-f`wb?OAHV91Ga#Q&hyGRznm#2uWwtIXfFGzR^MPq$o4LWMv4
zfW`S88@Wz<{pSY~RQ@8~w(WWq>+C}n!r12%<Bn|~8?eKK7F1Z^EAgdMM7@_{xd1dV
zI3+ttBA6Ufa*B(j_RA^6rARD{7wxlrzh4+(dX=K}kOXXw19U*AxJGGCjw+>gZTw+-
z@f5B0g;GqlL}XZ0L}s)Q`4weH`5fi?X*`L-r~`S84i4dDQB!#OIhww*Tf@(mqM>C;
zK`GyS|4eu5NU8UN{)|Q6Nf`|2EX%oCLQ~gdVuyysko#7NcsEp8eB3f$5;BR6s$<d@
zA0z{sXXIC{KyR4z^<$Z#_hxLkKa2QSi80feL&(Pdn6vWT4VHnzcW*tLHn0;YZDhH=
zpRqq?2ZYS@DR8YWLDFGryRG8f2Lig(m3wifD-}bC+u(a_nboQk9s|Uo5|srGqb<N;
znLOP<k3bd`+o}mh*R_7qfH;<eU;8ibjSLLJx1Gqwtlp8`3>%Bbp-4ASJR<~kSlklT
zPSmtc!C!n|n}Ic4?X4;5SnW;J3&hYCRUnTL5K=&S+^703uuZDYirh}TR-lcq*RHn~
z&GR7oV~##BI~m878RKu)dNh%MXb^m_Kde5onh+v7ct&^nD~@GqBb8T$sfJW%x=Va?
zKuU>E^RPpXrugSICrE`=S*-=$G3N21qd}_J|5*E&OltPIK=$3pr7O1>6nLY-pyUc}
z`t{Vl?x;J&U@vH_kN%4|9PaK*x97WT>9M${VfQIqW?ODO)qgCb!HkSmcdaFG-gg0H
z%nc+={icyCZDk}1e@I13QdsfH#|xHihE!BUr=G$216Uf5HZ93+SqD7ffoXeX3TcOT
z3mMsnn)UA+X{#c7v-7DerXd9qqnG0lb&hkh!`v+_B*b6TDy+L`;)ky+J5r;l@@Wz&
zfrEq#GzwdH5L-F97>0;w!bO65L#S*eyh9%*O*7C4VjVi27DOZU5AxAGRB~LniJ&K0
z!nPLmFGuM%`vYQkbG}gmH-zvJ98?p6i9jZhqH@FhL%YIo$k3)o;wQg5$%4#9rL>^L
zPwvs;em!@LXZK&3$ZI*rMy;q^4B#5OqyXdoSH)ZloT{~i6Sm%JoIX)s-LYSHBoPuv
zkz(Vc))jw53$M+7lMcKI0$<GPj$Dt7nd*HG+P$FO?yaR!lj@B!$NBhaX?|=m;@wQh
zz7owwQfG<sM?0|@@Z3-r4JLl&a3ucXjaf$ZFns;E+{A~Kt*;Ctvm<$AW1AB@=7reU
zA<}@tHz_m)apXFc^}9=quhT2oj=JI;Ar5O84wJMp)*)gku6YT(Y(`k`-ZLxeaQB~~
z#rYOG7o-kn5lR9Kpk+cNjPEtlY6p=HLA|(P0+qv3LB2WFHsQoIMs*lIJK^HILpqyK
zm(jT+Ql_<ysMp8A-I`x>xc+<<AF}04EHp{W)gow%1@EXGz^dNnS756c_3t<$ow(^r
zjPVES%EkSqor2;UbzN*dnz=XB0>S=a_tF-)&QaySZ$vDIvUfr|s8MZY3N36&VeTSb
zoGyhy{Fvw1i|Dku;f<d-KD5i8JC5x`spSHBoLB47O5~Nz(sw!eYJ%b;+&Cc$5e8o2
z@^#W~0<ysV;8SQ=`DjElN%79Odzr<x8e}m2hBACZh&}#d>~(Fe>RER>q26jaYE$Q|
zMX1`Kw{&+;q;Yzfjj(kZ+>M>X0L8|(8Q#}syAfzjEN|e*6--`x**OHUJ421kI*EX`
zZYK8?#$c|WGt;UBIn!?Lvv}%*NL%xJ+e$rUWmPlGUeQX|UZ~j^i`8Ppb}Z$SxMV)|
z&vW%y;grl-T2$le*0LL2zhIW(in5vnp7dyxf_exdp5jc@mx=r553_ZME6#|y+r1?K
zn%{)mVyoj5NThO326u*>IhxkSM;u1sC0(-ni+T)PA~g!!1kdQHMuO|mR<OsaE7c!S
zUVIHm039_nA-2m7i>Bk7j2TDk&%_q5BT${R%0TrdF$|6rF0v+n0w#SDF4gI@82Pv7
zChI>X;u7WBEoSj_oOtvy)hd{wZ7%KnxcVlDWo0<&k8;`BV}*QB+;;;gIZ6h&Ko+Lf
zyUuYLQJMW=Nn_8waE;(IJn9nj*TUfag8ruwfxWx_n(*|;*N3Zu<%>UOe2zl9u2Bn!
z-VIdL5;pFA8_?Q<g*fjBQDtyt53o3pJD7&d5Te1+ApGILRGt8suRy5<orc&ax9jig
zMEev%0z7R);BMsOaMlP$$Y}+YONfVm-?&F!ce10&okf!v;4oQI&7HJS-u3zYKnOu*
z%p2aP9EpVkO{v%On$cEH9y+Qpr&uvRe%lqBA1|w-O7-rIw<P>JN;NKtS^3I_ZUEV$
zyoS*htw77F9U`ByVGl|zE>E7$Q?eeB(z84JpgqS5;}fEZak8Wx@VL%{(p9E_Vv13-
z8IGYiZJ6wd7b>u%%&-&Pjyaj7-jUsB<)f_V(%x&9kvyne?0i-1X~fi$VH^XUw*p)I
zJZ==e^P7#JX+WzToiB-yFT>iYKMAvTOs<Sp=$i4E%{)lBl>a4>;(4Jmf41BK2wwhH
zy^qp8@(R>Lbd{??BAqs=MIx~tG2=b#J3-1SnJI<Cc7hFf1C0Pde7_hAzn@6;XJm_8
zWAQ#2^JyJ%ZfSi4!i~#nr?BXtSwKCH`(?JJ!EYG!GN`)l_EV`K4yms(`BEb;l_S>~
zikd-Ge_teUQK#sI0(I`@Jmnn7k(z?eDljmRVhu>;z7ekn@?%RWReDoT0mVW;_WHB1
zyG4m0!E)XUy5kX-rt}MHZ#*h6yqETl>}T!|fqa6RzU<3kGh0UuNHBT)bCtmHKQpW9
z2+m>}2%vHg#V**{&(})S85qHQzc*EmY@e^A@Fq7`62zX+Mdy!CiMXp&e|M-<G3TCs
zG$4LDW<_U73rHVM1GAy|mYU3IFz9L_zzKnx`~ti68v*4{sT!bj-T69r8baoN9?P=N
zRyHS2Lrw_<R&Nt$T`3AeOpLG@c0gZLc%e<!k+5Dl>O#lCW6u^Ibk<dq2d2fo;H&El
zM?VHB%w0>M;81<Q#eTcG{^C_>j(nklROt}gmHty6^5hXdfqF4@im6)h^<Z^6Tf{LL
zYgwnCbhXT$dE7^f?83w@pRzeQ*GE~1WScp&ih%U(Lki5`oVn6iY}Rs%q&r(v&pN?w
z6j(l^dm;idU0Czp?Utk@^vq&{>^4TW_dBg~<`kYdA7J#&^n48bb*M~yGm}HttJLgU
z%~F=jmJ1*IRK_7%<#u9_)hn6%ESgo9CM?p_QQ9Bh0ytrJ3ycvbV+F22FfM~5gD@~u
zPjs7vxKggdB=~6pgbstgy~bhgqHk{@9H8u9?U-3#IG|fYC<D6q8kV+OHmSW4j3k%j
z^MS(9h@>-jfD5mTPkU#nj+S|~;7hTpa@m_u?s-Ql!r%d4U0UKOAJ3Vsk5qgO9~wN*
zhw@mcwY`;fydX22%OU`>TDqjEH#MfG*KzdrdXln9TG*z{uh|0IW_ki&Od2_m!(l`p
zD`1}RQ2K~g@f8yRVH*xs_5-6#0f;J3#Hd;WdmHy=b)HJH0_M!?(^2}Qa;7WE<5=@^
zd{RaH_g8~Xv5}IlZ3|-U2>LGG`0<)qxf&X$s+6YI9@6~274nA0=Ulss(k}pHY__3E
zOU;7y%;f@g&qWLzh~P7`c_6H<y+4G=F%*3~2$aa2`|oRL20jfQ)$dd`@xbYCkia_f
z%Jqo+A4O|0+uh-vrMf~i0oyXSsJ~uM%LNoN)9S^iwhW=qltcHSLTHzfUA>+$=rOam
z(;yGLkdM0T<>sPD=e3Rml+H&Qoh08LmHU$Ef>J(VyxZUMe5Vxl=%+pXPMXwx%NKvq
zHpcvp9>jW0G%SjZT%=D9#%eiaY^x~Sh#K(dB@Mh{Kjk3(d?1(;G}MIe`_;eM;73}#
zu$l)?gSWGt@d;;+kEz7^1J1U)LGaZb1Oqjf&J~UXh5zzwH~-H|X%?~K2(xRMHhk8s
zQ`*ZJ3d+)J0%4-$33KtarwY=Ku}fL6v6kay2dI^Ap8#~h1B>iq8i-4eP&cO+Xbss4
zxgmq@gbys^B6S3+W!mfL&JdXFUM-9re;Td?4(VRIYujVxS{$RGu}2Jbgi=_KzdPsZ
zyj+VtmgWY4W%UX&9zR_Un`>HfMzxqV^T<&A>QhXT8zCiWpo)znc+ev<8VNj2lwT99
z7dU5((EfzJa##0aP1p>JNMY5e_8q?LwYWt2_QRQK*988_Fkp5BgYvUaB`|%D?(klj
zI_an7Z&QfcY?w62<r-UG0C*Sjh4<!Ml?Vx7I5&cjhfU4!urg2$iS6FSfy$QSD@%9n
zMg>(PM%uU{5%m?SPO$!%Z`XYOpsGw(z8F|x-!Fk_Hc(L6&AN(L>z)%<6ss%(Hzodh
z-JEz;-9BKUg2V|8%a~eTxtALIFfb`ltHz;~mV(8-N@J<g4{}z~^n2f`(LKi8dtemC
zdS=HoPy6G=BBQv)lo&q1=PD_tDdA9I^FXbLx@9O^pJ$+HU$lrBOiRSkH;?j(OGWDM
zY1<2^S(9FW)F9s)#0*SATM~EwiPS|9-yELN9JxQ;G+cj{U|(Ubz`xc2LJfZ&1jS1X
zL6THRN(<R}K8BjskIM(DxWJ7u@>-&N=q8-*2EsO?8H{mgvS<|B1=zRE)trP(sZ&zo
zil!4)JK@T%l4t@y<J2^0Qp%Ft)QSmU+X=tPB~1qQAGfia+n4YBvKwCKr3n%Pq@wOf
zX!a7Iy^tih!SeW=@S3dwm!bUdFyeJpy?Yrhix|KvltXkx6iuV}+L>M)?kWvfuakrK
z28PPHg|qdvF~Qa;ljxxz#)T}ykN4Vm+lkZC3-Lol?E6+)qiW2SuU?>dlQ^&G0Ql%=
zS^~eIy+BHo0Cqt+4*|drFW-Js6M7aQ>-8R<Y-xL*1Tu;Isx`;CF3>zuN)fS74n-vM
zXN1tg1Vu09L)7z`QX|$ZAVpoN;PWyvrq?PoR_)ICEug+D`MG2k?ICs>yf-Wy<8RYH
zPh^(dJn}i3ugL0NM0EI~JqqYRe>GioJR7}JAb11m+a^6fkk!SH=YPe*IZNCer9qq?
zbCDE%luQ!}eC;dp@4jtU7M$wM=B2^kk@f)b-QZdcKC`D=M})@WjY4ja&WLqfZdp{M
zxSr;$WWA|xpp<6s!pWRKWO&CEsl_vU-p*R-aMpwp2eTJte~{92Dbc8j8CP>xUyktu
z@!T-|NiuWiyH<|?t(^o>k=J>6-kr=y4jnr6`AxIpEw;_>(;@QkoS#v2{A`^UFe9#<
z*<Eb{)@s8JR;vvBnE%G-`dAY-O&?;GT%YG&HPsfrsc-mh9hxXY7}<ava&KTH-`ufA
zt6=knW4VD$S3XUW-0@p>-DoFaG|IdNui?9y$ztJ$-QKh#ENN}+LYQU~f4-=oxOrIw
zo?-#l;_IKYbcScV_N_3a&o?zKDt{9JrxU<DWDsz@5bcM$C0o_0OT-<>Eve93yEap#
zGxh961P6p6BZb#obpZ~ZKrF8D4|*wem@!eDVmK}Z+zbzaCwwBDE)KvIAtn_Od(Qr%
z=yEaXF15DOUiGHm$qm|um1`zc`A#><udV$K*XgXN+J2rtaf%?HLLB?rsmf*6lasuO
zmVTuYUB;*zJsL+!yOGG#<GF}6D@cqDZfNJLkP&9wOr-c{C+i;F<afquuBC2qW7X2L
z0W6-$*}9%ie#(7MhDT4M(qC|BXK%~_+PmVvqS3#$aHxqR4?L3GhfDy_L8?+i%iKI9
zekAGj*HPah&#hUNc9my1R?NrZ3(1(AbOpkH40DaYN;2%rpIL!JwTjfv<UMw(K`NnI
z5CQiNU7s0&7w955=r8L(%xurtl<8~V7Cc$j{j4Y}&4l{FD}dHopjz%_fF;)fO){1v
z<~R0QG^jh7kRRoMYARvNUmJ|_UC{R_g+h^}i7;^t5TX@M^BQ(&FI~_RoNW2km=&PR
zf@tP2Cx*vSWaqe*>Q=3M&|{qEaAWL;4~9Z4Kb5tYvth8O+akWgaAzLLiG7gWygK*P
z*j}hsO<%)E4P~3(RbfNq2ZTo75M4|L%Wt9lam(DfwUi5VUk(fQTV>`GaKHxna*dJl
z9#Rl~bKW0#IEo1|%;#o?7P8k}LREg6ywQ`mQz%c|w{E>Kss$YsHw7AS$Ua`|(?9f;
z(BUCsxY5DK%pQ)x>F}6}rHy%%J>{#KdPF;Sds0zglun8wQ>hz`96#mLA~6<N7boM{
z#p5!;C8dmm-CwMm#8e2BU8>QOW@k!d9@_oY1!2)imcK5`VkDpV0$Jh@(wCZ((aVva
z)_UgqbWj<%L@YgyC+}XZx`s7YR+!@xN{UA&1fLT-1)R1|yXYM4FU;pBcBq%@4&J`B
zPMAysv`p+eUDv>nprL*~d$kh)p*_kirk=P8hHX%z5TTR>EC{N%FwNsL`$XM7%WC{X
zMzWbr;J|{Lpx)3NY7~_nuNa)%fZ5qn9%ZJzxu+f(_9JhqPP&C?D~5y<{cJYu9qVJM
zKV!tVzJRe@^c{Mt8jA^QP#<nIq#AXdsl3Y%#36b1gOnlytBc*#QDE872JC@=M`of(
zJ%jiA3^qO^oFXOgiFWK1lFHDNvA8=@0>?)!x;u2y&UGO)zj|=i-Xs)+x6%_o18vjM
z3A~Vce!%iP5#}qnOu}rq6<F{o6v%dMg%Qyk-EHn;fbEL!gF-Aterc5+HIxYLmAZ8q
zWDej%>B6Rb9WmRC*?6Qx=SD02OTO2nZjcF#Ds`p+Lo*IQm^>VQNa$4t-6Pg%P7l&K
zp~y;&iYJ(SCqP?x>IcT%L6Cb+_Jvk}%TtZXT-?P|1zK2y*HQ1QI}e6G*wJ>3DdH&)
z4whq+SHJU!5fn&mX6zFx-d?H!VTx!SWV)N^C5t0DXWg{g;fCSKj&qkc(5HUw@+%76
zwdvPk;>+c}Q7~f}NK1N;B>`weTh#m<Tm{z>^dXCgIxPqy<;T@S(QgC;0P=0osttZ{
zn0Nh_#BgxMvAjSIT5>goEBc19azaoNwKz8s-BAs!8_Pl3d5;$e5?;EYcLC$KaWK*R
z@{-%e-%sS0Eazbj^LiHo1GZUui{&&(r1RCv5M|uoIEn2&wB6EcLYPfsuAd<|>wF)z
zfl;e%)A{e_B5yzA<R>*P3N#g+F{)9$K|ugtDM$&)E%%D3#dHeZkf{yUKI(@OtoQ9~
z@N6A@U>x{`K?qE88B}>ce(WAUiX|hv?xRPH6GVpg)|4VO{`!E^TM;4!>&QY^b}bfu
zP|cDeseW<KQ&Q{8{i=e_kFhXR@yG?M)9q&1b1~lGCe~=(y!Wf1Fv<o9D4&mAk~yNy
zxlyyFT3UISQ6_*qAlmRnSP0JJr)ZwMADY!if+dD>(}8%q2!9dsgCm++s}VWk*&N9A
zO3nJfjk>Hj(~?zpr277JbHf8U7_(6=8m7<F1@pOH&<qjpaHuL3Lzpzm?)IkcxtYZo
zn({M+SeL5-s$hZ+wIoq~EuI_{25w<Ww2cgfsW43R6MRz}DQy7Ztv5&DIChAKoGgTb
zfhy$Fj=$@jmT*!Kmk?36w7rJslbMk1)QAd{|D(8kNGB;tRTj{Tb~M8rpoF8SVT1+f
zYVTqm)E4K+&gNe7qjRZE9PtY#F!t0nyuccQgmkJrt9DSdhh7_pN08Wx4H`*!nr|fS
zEDS~m<p@=NLG1Nbx`!>#w_T<%ssqzv*<JkUTcI3>z0Y#>qieqp2(1x*y4BPh41~+2
zBsj^>=xLgKcfw|M4CTn&qM6@<cW^xEnDinxE&rf(<|;;$fVL%DVlJ{_B5rY_jp+QB
zPgOvC3$sg7_DFpa^X^E(Z)oYC+HT2+w#J(LWEx)<;ZUk$Gj~U;0a$lkwi!dbIO6Pd
z)&g7%8t!rjcHTx1U3<62<o3Y5SqfWZBg@L@65+FTBtF>pWMU51#5O<J>hcn^ih<RT
z9w%eGVPWOn+^+-9x*T`Ko&8&kQ(}XMc_n9lJYTI6g~XO+1q?u@Di`=2nx@^f%0C;&
z7MOC%%(gJuJLi<|WXRZ}D&?5))4xwnMc(}J8wjoYRl!c78R4>&UamUTMZOx*03F6^
z@!Z~rb~rc#BWug{WOFT$;BZ_vPSDB27ZS@z4wN?r=T(AGRh@W!#|pW9KZSv<3~#VV
zH&0X6+MD@|I%`Qn3BG?nS~zYs#K5zb;eF9WXr9_$alTnIyuDhdy|!JQ^@&Ig|8Tz<
z)r?!`{^8&yDAla=>UjK`i*L42x}}#61u4z72Mp7XPq!7nQ&+_ImH>i7Q8f)GsG{|K
zuo=_lhcbu!vgg?rIF}$=7gr=!m+&O(#HeZ*Q9&19)-XcJ+D4B-;(=3<Qk#U3riRq1
zF+2R2OQ^Sylt=zMIURKZvVx|d;Aup$AMFcOJjh@9PXt-myR#YTE&B%tg7NTBdn}#p
zZngHaBv$u~@mb}jDhh;G#=FQRcCK2KWIJyzQa-w*Q$L0CJeHy&MEu<4?_omdP-6i~
z@P+}BT=R67#-K@{8aSg=q9)Fw1MMCv6VQ_$HAxW}uvtfs?e$|#pQoyw)B(j=ffKwK
zd}>D2NJ3(rSR`e%QSh;cBgeW!!t=sxK<_zKT{C_l7y*&%l6o#i=j)@JGoK66Dw^(#
zC{wwT2enpe11pOzME7--di!<Ev?AU{N<#$0^lu@cuxWmsF)$-Dg7aC{uM~7YmQtW*
zY}@#AqZ70#oY<XpxHm{`eYrc8fV>AHv67^_V)c?|+%wQ|9n2qs=(daLL%a<n$OG#-
z-cojw9xD|4gC5N#z@$<noUe6%3}FLacxLYD9*fLyq0cK$g-KvD$cBTSK~RxTE@lps
zhWMD5Gy<f%td=H%WJ!udQ?wGN)6}ml9E&O>xD2A$VdCYj5By>l>jJ#T7E&{89xXn)
zS`CDKQ(Zib{Z5;}=(S<4bQ<<NaAno$pQ*wb93|grmG6VPLpkgZ=rrU_EgIUcA`AuG
zr3#Lk4sonBT8^^uhxUso5*pfXO78U{!z2*C(2h;9GqNa<MEphUF8&ndXvGjvD^@RB
zO;Cc;ND_P9_uE)awv-}r(=JSiZ_Xp8cjo#xIzw!mJNr&hjgILdz%)8NokV0vP66YG
zA(4#zN^?D_^O)RK)|w|ty~6<ApmhZ<)1E3@64kf|i(^^q{)EzFas*22bZyqE%&j?x
z>2VN_EZ{fcj%G9I%&K~JGk6+`sM1d5%Zy2va=)e>?V88)2}$Y~l^N3zj5E#&SZ(Dt
zZ(qnhy>;~BHO{Sz&0r~hjzRpRU6yDxH$K263+jd`q(uJ1^#LtE|8tgoi7GO35T-pA
zimp+-sASTwo{xBmJQ0}Htlh>f^9%e<7J6M3#P{*25~gNM$L`JI0o3eJWWfmPM=V0q
zj>i?9Jx>IZ>G){8So)#5qI<pfB?!y)gxvY|RR^q3OnTB}f`=_w+xH4$xSkGMK$qB!
z*ss$$@2?YYQ<H>ouHvT+3U+-S>KK<1JiU#PrZ$;A6llqV+;{yN%6M5=3<@$F<V{`2
zon;X+p+;=iu_*a?8xlBUcDKCXxFOQsKp1(jc|0@y)IpJV5*KS|9)5wuHbwi`yQ(~%
z!qMu~H*=FB0>a(_8CnX$AfC5hn9U>E0+h6b<F~t*$_fV^>10hrUno~|hikv?(M@Ku
zgeQmfcBjb}E+GPD>lgfmM@{4jH9YS&)S1b&Ge>RV4MF)Fz<)U5zvn=5rKCeD1KUa8
zL`A(Ld*e5D`m@BPm+9V}&B0s@(B4d~o*tyT(q23CR#4f?xF46=A0X(2g}1OB6>g%<
ztWg?d!`!{#u}V6P-E7p|Z`kv@Cs0TqdxxSA9PxHpi<o5!=F$Y4hEEb6pS$5cRBk3J
z#3Yjf<&WhW_nO52<Hx_5SF3Fx_6vz^_B50)&y(m2^~Q8P*sc3*o{XP`@c!16`qFr2
z5kD&8uQp5D{0@RFN_w+OrM1-2S0W47q&S#chMe%Qpl%$^q;_u?27nhG+4228kvM>=
zV)$QtijAftC^9|Hg(-J7mzd8hz`4_vtA<(c(rkCZ{4GMxR%hYsgmT(<-u;;%35^xe
z;j&~c+h7Ctyv4aRKZGplcwjDB#iyAOwZ6z%2=YOKMS!HIMz4tAIm61!Q!xbEr2e$r
ze;2|NlnZ@VBVwh65QS(3V?WF<@(VqF7G>p(_=|kV5@9yL0>RXZ9zx!-av35DU4$_?
zb<7D)&d7@kjF3PDNxU-sY|31mSRdCS+CL7?H7gWgC)sA0u)6x;;%vh4eLa^*^Aqs0
z^0gFK4$iLi2O)U?i@Hyv#l*)20ehT7%sTQsfFQI<Rv)f5w8b|YZN8!apUrI7KDB9C
z5Y~T1xpoGpA0&k~W*Ch6MNy1MeZ-W;87Lgu_pZDVU#8rb_7T|HpMl)ycVd24j_6vE
zd@~FQjw~<2qFP;{c_6Wt>E^Rx!WguL6s}gYmiO;+*6F;I=!zko02&r+XtQo`8%#v6
zXdLyNvz*YvVk4A+@llJ1clcu5?O4HDX%FO~iCOeA+Egy{m?tcs?RIBh3d624@y9Qk
zyI@prHIunO?E)cbPXU+Fw>OzhJM1J<=kG;xdEMllz~1sanC&Rz;jE?i2BvzLKSS5=
zEQ8KBValu=F<&DbCaAb;mL2W!un)tQHwYcun)tR@1wcH#de-`0JhlVNXZ*7F*C%pB
z)JxFP#*8U!ymx;vLKltTb=B&JTtsGaZ~e(XFjR3tn8t4}+lVQp7^1Vr_rjYnv_x-_
zLQv-Ixw{UME*P7wU59;8o(wAzn@RkXYCZ-(5l)=R`_H;0EX!_03nVVe{a1VrjMu!m
zM_t8_Z0!W-ORI?V!-$}5lBAAjaT9SXvweK`JNBW{#lT_U+!MGhHts9J_DZgLCPtoR
zpDT04IQ^9(I=&vGFe8?7#UvbQCWbZ_<~}Y_>I`y+JS>(a^)J`sR^ZW?2iEZQWvtFG
z8Pl<fF`J=He$bAt6fL{;hAK>~AhYu}T12ShDlv<Hgiy|bdjvR7t1USM;3V|dBj>GJ
zxDVEZw;-gm@J~SuI?GnawX4ZFB_R@kpjj=hCO_K6k#WCiQe=DKI4_b~cWtnUAWlwA
zkfbt0lGbVLfB&dL6dvrh%)X<S8wHLjX_W&z7ir<28kG==P85%C$ReZf2Brpk&QgaI
zlmsGJ@V7O9TyYlQ!yCyqA&uYBg+&)80e0zQ4GC}2?Y)tx%0iIUOtGzq@6m8#j?S%1
z{($#&f&gQ$qjE<2<4@@Jex42GXLa2;@8K`z113Ny)lsP|jW>V4UwYf#|EbPde<MP4
zsQCt~`g9#YdN1|aS^H;Zu`^-4M5EiA`9ZF6_8^2ID`c~*k(CZi5!y4XsUHRF=>#2i
zP%0P-2`zZZx$YQ>rXAa6xc3lEx3L!oN)wXo6t<1UYcmP+^4w1-Uj0E_%ObG%?sz~@
zxSi>gMWk8A1-MTR_V6v)zRiR>Cg(aZl`We~1YUeKL1UV<;O7`L^pD}wEkXY{Y&4aI
ztODBD$C*8@$@NrI1IoR^Mi*k<3)PUY$vypg;wE)vmYFnr3;k&cMhQ5^nYQtx*wJ9P
z3z6Hq#W-Kyn2si&W1}hL(d7eRY~N+WY9#xE$IAnEg_C~J+^$M*SzDuu?KLJBH}w^j
zbX~kTfDEsFXqB1VpjjCok*#+^<3n8)L_U)SNzUy>!6LN!)}iQm*nph}MSaZ~9t*9+
z9yT=Fie{r5aIKypxIa8+@{LBcY!;&3FDreUZo%GBH$_J@aW*irW;+HR#@yavTXRna
zvw@@T@kaYNs6e(cp0^A#<FX$0X_l2E!;N>~Y`Ydl%|19v{h5sW!a}t#Bf<oXNURog
zP@c6ppjg)ZKj&U<8n;blG)Wr8p4z&c(73x?;fSk7Gni%FFH)L{g@YXAdeMls6R8ja
z&n^3nqWpB5SN)W&;>Zf~(6Z@Im(14=lP%NyqlE&#dDiDXNAZbNCydU}JNC-;3VZTU
zJC2Q+U;hsSK>WYDx0sCY9Ud#Q_SuJiE4Bo6;XCOgxDfXD&4_m6mwQ5|O34lr-cGz)
zCs+a-X;&3>_IfV}_LQ^%xVyUlj0_!DyChOnmwb|}H=32Tm`PdbS)*&yDAP+m-jdrH
zTQll+u9xK?C&)o8?)3mg;ZS^NgIYq?90hRnNGyc#6Q)Qr<kJ8N&IIfJvz+~gBPc9U
zb<`6^JD0?kf?Kqnx@?$K#yl1<4PdA`w~T5)Sr?Vdx`61vyUSLaJhc7MC#H0+3aejq
z98UBazz;jGj&YNW4Qt4=7*Kyx)ZnwTz=<A!f%eeXOj#O8Vxfiw>cFHOL>AW3t|I1W
zzlE3<w7Ztsiz!-cGYiy7uPfH~#jkVLX#z1odkunSQ7L#Jf7MPl$?vYaPl$SG^FGbF
zra>!e%B#~A>H`g3Do>j9o9T7YvUN9@YT_4U?}$JAalf&ucEX&^X6Gl4zjft|A9i3$
zpH0qTYc-$(7SMMEpn2!I9@enMd2{S1qCFLej_uUoj*PG%<4ow<MI`_QFxU2QzH#(d
zO1NU0z|gE$b+Rd0ufS0-stoiiG81^^8z2Bwjjq=;8MF6>n$4%7fnE}+EFC1B9CEjv
zWapD9#V&EFLrr`jo+o?TX@IDoX-nWFmd;IVYcA*rSg)W?A+!b^XohGzVX#o;7(%tj
zvw9eJrV;aIExZ$Jwue5xLx-q00kg5+7?<>L?N7Q0Lo*)rEYliZCd8jnsDwe>9HVlw
zAfvanW&M0Rsj&pMYmq|Xhc$ytRU8Os8>Vi`NawDUtm{J1mf#nv=+R-qwkAAWKAC?M
zisNYg5S%qp283<EX&{TBp7J0AR&xQl@=<F7J_2_qIkgbwMFR?~;zpkHDDtS#5Wv>{
zgZGsa;Q?_t6Q@IYd4j6{FYQWca7a*#GrK8!pNR;6_YbJnc7wxjMuIIT;CFl9FZ=I9
zD6cfHzsVg9rO3zr?M`(X3~H3XF@6gB07F2$zskRvtf9J07aO_wT^{NxwT4V(u6w-H
z<l&}|Z>$N<h_u1WojYl6v9YzACH+(&IIJVV&fRz?xG+$uyzKznp}b$(?D=<6D3Xd{
zfdat88W>V!{|YsmH=bySIHJTRh>F;>eC1XTW=7{$a`S&}LVe`SXDG!6fClsPl^r{L
zY%lGID9R?tTg4_&F4H+&h*g~vO0(g49B3^NR&G<zskPhxp){P>R#n-SEw+#E)V*e&
z9np%(w%96!nV?MA!e=>p<jiw3-H;Lj*OcKo2-YDMm&IP^R<Qk*pyBnmI*WmdEk!&G
zkI9Kv|97#0DOE8R)qXc!F7351>zKto86*0RyfseoyA8B@TBHKv1}HZpW}YzyMfAu=
zB_Qjx&ZvXNOqwEd@}tA2BPyw(o=_m4j7RF6<!<{8{Y>i!l`aCF(=Hq`(92R&;RDbc
z92XBcZeE)fT?CT@)aJsx{`|=(dAN~g3eac)U4r{@N0{mVd4jE)JefsL9qvfr%Wp-K
zOle;4I)(BMaPtW`jBKb9h*0ckp$@tSbD7XQdOLva3}`%AjU8NIT;?tr^2qZ$2P2ee
z|HJ-vKuF|w#iMC`P)(o8Es3H{Yve|BU?48O)}Tf#d!{O~M{TMQ2#Q#-9gYVUMMJZg
zc{vF|fJ(k<n|L$G4Stmk71R&rHxf7bc>P?kB0KBnMh4rE%zIM%0uy_@Tgs*+Cz3)}
zBIAqoU1D+N)#ME$4PWtB(U<Wj7=^=-H40vlS*Rtcf+lp-`i|W*t=0A468$uIjVn*#
z({Uh*@%UVnI4JpqZ0`pZqyX6cpR5tu8Osx{Y13gLU{|E9Lu|#qb9KtA9ElCrAueB#
zY{Og>!uC09>CL5pXcOpSD(d0hIx1hR1gNR^|8--&<r!2^RvP0+gdkw&7q?xn0!`)m
zQWoL5WlP1t?D#)t$*EdwdrXq-VIT=k7*agwtK#_`#)mm$IC!<wj3gb&DfZs<POA}g
z4VY2zCGG>2qL8A>y%LaUZDW{F*D6o+v#I0YUj~zMb%C`3a8|gaUUP4{Iuq!XPu;c7
z%V)NnAxjQS31b^o^1mc6-!vU7oKB=464hPeZXLNgCo$H~I=WY0WY5R(6oGD6BWi5B
zcd-$bE#M@TS6V9f%0&lD`1vy1diYgIq3OZg_L>t$I!{>k$S|(??=lOYogxl#E=W_i
zS1@$fRfT*xueK=<RtySiu1f1Z0diZpYkW(jyav&Gt2@wC7d|$#uP>!SlnAZ?&&p9N
zf{UY$?k;kZ?HlYyTcQwi@kEs?Sb1TiQ9W$Ik0LF3e{anHh7ZQx*{&m5fCcNsM(A>k
zjw^h*;fKH+5k2^6k<nXT5kaOeVu=MaC1`Iggj7cVT$cO|-jxonl}!?2=5fX+>=B06
zxMshDW1x_6P%>)|`u@ewtH+L*`Ff%bAOhlhP>1Slwtcln$(nj)kxv?kWr^!lA+Mjg
zTgS+;_hS9edrf$&?K*^GI>a`(PBH4Cwv?_N=ZjP50|%t4w$whEmo`=zldeoF0FK$S
zkYxI&aU?@`DgD)RC)36t-;{umtEVy@#CFi!q2K~^YE90%g*v;(qCBj;0rmn>a8rC_
zyZtq!s~Qxt%mS7RDtUu5kPeDmbwrWqfNp<oJIdg=U7Do#Zg@Z1nmNgyFld-J@#*Y9
z@C2mh5QZS~eZlFgI6%%Y&p_uU9ZS^AK!M|<&5Vp;=mgzjkp;#ozp>{q4|tvUmA$k_
z*t@kT<6E9wVMn?MV8c3c<?I)4Zvj#$c2x97JbUsBp>SemCph!!ov*6w4{nZ9gxHKA
z(ho^&7EbkL>Tq&qwX6FaR{KRABQ}oOkc&U-S6ZhdHYPkgav5x%z@Hp%70_nQtj7`k
z35sjH{M}R}bqui6>G3f}aq{b<T`)a=Vuyb%5=5Ei^3<SH*&VaMleHsRuYrN$a@l9!
z$kxU=@j|?DO5C)w@(t?Owp?)7WUNsamPvzO7?+#k-b+?k=qr)#?%jOci@J-<Sor+X
zoRd&E_EF=#0-$u#`s5(PM(V1T4ITSM@^k2b66Siq&tI*z-I7`Q2C+ANJe)C;sjApx
z`X;Wx2qC-!M-e2#Npt)2q#vV^Q+txpI*GskzUXycabNC-tMU$m_T%b8I1<HSh+4Z<
zumrwTvxm~&9n2npG9^0NK=r&+9kMa^kuR*iA!1HoCXg&lAPnaM8v-EM8D|3ijB-$)
ze)a1IC;7`!4tJ=ApJtmVM%i1G|MMZoioz}nUR`NNLe`{Yhuu+Lfh+uEFI%Uy92D-{
z#+d~!RncT+Kv*jzl7PfVHQ3^BIxsV21a0jpP=sTPEs_EX)c?2bcLOdUx1OtP3KB%9
z+f<15lQNOtgP#ZXJ%RDwsfqH2<6twZ6$?GWxU5H^(G$)7oDy1t)nFr>oH>{=g$X8t
z=?{jrq~*i>5=80hE!5xob@DKWhSuulSUEfcSk*pO_pMT;d%i2|lI!9IH^7hHY2Eq;
zF^UEl7kG2po;^htez_pJ6VLAZ<9wNi&L2G{$F|AwLbO8qz`Dh##&Fm2@Hq#B3T&PJ
zSb7dKuV!Fza{clVohdw-ye{7rN+#8{wkYdl`z7Jlv=pPNYR2Z04R0PBwg3Nr;cOc<
z(iYp_Y+AeV#1BydF=XfjhMlsZapuUCTK%Pp);$HxcbAQkX%qYhfc01}jx0C><!c(H
z>*Xls!s<ddTY3s%rMI+)f4F%#p;2yuEo4s~U`k5w1}5`ED*y~rO_5ESt{{;MnNZ~y
zPP~C(O55{%Lvv$>UgBWWA6e}OP#xoscct1nLgK`RleQnQF}SQ6n)0pvJ2y@_(LfS^
zan!V1r6N?MU1EpE58TsxK}Rj*&mAgRFo;U81|HQ5s>?t@4AaP@K738x92df+q;Fp)
z9X-$$7T}$(w2GVCT*W(G>(C;}WiQ$^a&~nCM_dx|_AJzBQt6erb3_@R;69LgdLJh(
z08pVW{XkBVfr_P$=o+(W6u7y;W^srTbAG_TISvMZ1B6)n_zP=2SCXACqsq5+KniPd
z?Uq~P*Lys&J)$YPVB`=%Rate(`V^|`dnRliO1u_-?UNfMU2pk|N6>2zkg0Ijtl88d
z5#gd=rM8tuVeX!qHTQneiR-WZ0O;Ed!D}AU4AwqcT3wwV$Bz?&`UVEzes@5<rM#!v
z9JX=9v~_5;cG@kZ8R?TMcX6pEWIi3{(TSLxvhQ*`9uzW32^5~GyqxsMhN_j%6yDo}
z$|VKiYg^~c>lQsmQG?RA<lgfaN(mKqkDw)6LX<DFmkj>un%Pd^-z5#f!)mq$b@#{v
zV{-yl<d<qC_95-@Yn2657#5b_R9-cl1M$*Dwcza^;}j<#IA8?OCNO~Q+<mvfM_bCf
z&1C?|(JS6NNARgI%jA|AFV}PKCGM?G{KcQaz4ZE?wnV0_%-Y3rRc1PtxYykGwm#2B
z+ra9_yc|e2w#)8XEViQC1CU}mb53a+(QItRL9esk12$zve+L|cu;h1zARw>DcB!=A
zCkh|W5ei12E=R8y{wtn`FqmLDM*oO>qW`Ia0f_5cwe2hTIo$FsbBsxbbd{!iFm%;R
zffo;gECi~9wD?)oRV$;s=?1|@MqhmK-rd{9Nv+Kt{zd5Y!cxhc2r}WBk2JTnRH+o*
z&0{(!&g3E$Hxtudigu`*ZZe+y6$bL;;KX-eJuXHi^CxgBd!}}M%9$8q=w8q@RTDVI
zc&U!Mb`B$nJ^i$O(7avYBi$HNWM>er|E9r)QKMiWFdenC!uVwVo*gc)k6Tp)S1*5k
zz|<CMk&7F_o!}aq`^<HU26ekYarYQXfY`a>ZJQ;o?kN4YDO&wtW~Wwv9I%jpTuaD%
z?_;(_*aHmBrYBWOC0yw8JE?7}sUL?PGjqM7$oWiLn1*<lFz6%1PLZR^p&z_UJLR-u
zEr%T0o?BV#N2xMV^%vgiPe(VHZb_$!+S<(d6J`e=7=%-9gp~p4RYVWpMs1+OGJ?qC
zmuvOWDIL=9B`;mG{iX&Y)}u2Bh0=aU{^S7s@_3se!1EzfTuC4zmpvxv%cw#8JED>~
zAM-o$M8?;Awq&Q{Sy*VY2lHcK%W(-pqLoRYLY+4Yj~il@b5rGy$e<cx^{A*{^({>0
z;~&Pp?dH^|!27R?h@6q;x8|z)OEzhE%;6ZPy3_O#SKB|}yfO8xLL`W216?=X<)n!H
z;HXMLZJCEgZ)#_efvOT?TPD{#{Wh^J(fMeRf%DvyN-5q*`;N8x#m)5s`nKm`n}usn
zsjw-(mO#ud3FjGL(4D(V3jbf^QWfYA-|4h07h2Q(xDAd9nw(V)v~hd~R+}pY6UM;Z
zIT);rYc#rj{vO;z^3M0RRg<UQw@4~0$Dt}hXIO?<dRfG~XHQ-qLv)p%Xk1+1ekE}J
zA}g_J<l@le(dLzGeH2EGJB84M^69BIXUJUy1QVXu<;C0_v=&pAN8Kd6zA7DzY7tX9
z`!8>xW}X7jE42UJOnAi{ai^M9^)Hqs8(S#Fz#RzOz)^|wsNyQerIv#I^NF2ivw)sZ
zhl}pr<CXUL&c2}k>&src>BIUh+@g#1BFrar=?FC9-Iy!J=)jzCEbS6k56+eYg!_Lm
zBY-AVvm0;MIPEQ#djM(3X0mj~vFxPwpzC5mSDsXhoY&LROIDSk>u#&95>AOg%!uz?
ztcE?uFR7ACoxEGG%HPe{LfNKkCF3ftz8Za))GABi5j4-lfJWA`vuaMJ3sNz}g?Gv&
zLN!X_W;;QVQ{|lUlcLqk>-`Jua*eOjW3QuJigJi1BV9zEg2(Ozt>Z}R66_F<V?~6W
zm-F65`<Jj~rVT+rWnLZl!Sm_LG|CO?zL@!d{uJLN=8&z39v=^oj<ixEiy|fIWEB9_
zPWicKu_+ScHlu{pA6Lk){o{K#vZ&p`PtA%7ToQrkh!EDLRwoA#$^L9hgw;FUmfUeW
zU!&5OMZNvT|2b}U#3n`**y?U)0;j~<dL*TuizB`KUm!TsQ19135C~nAi}zwrFtQH$
zf2Ew>f1cBblrAU?Pbr^g7EFc%h|p(FBl(m71TYIV8bl15Zw}nY8y<A`A66P#S#>im
zhwEEZ5yb)>6)@=f+l-qk=f+LE2d7;^PL7)HC*_s6Bn|o%&|x<}kk#y!!%GKJ%&eMJ
zG_U~Tt)kVMK`pv#%!_ri2e}2aBQ`OtE(gMcr*leE43b*<qW{%lB86Gt=j<)t+%~*<
z0;kfJQ`kB|Tz{`aJ4Ngks@GGy*%m?a%K1Vxu2k5TmfxkM1Z9aksk$CXy%<Wq8=_)~
ztBH}wr<6kLZFuCO*)+1^Pt#{5lldZ0r`-nFrd>Uyh)A}DIX*)2>0Q{@Vou%q+f-Ai
z><0;8dVjU`&0u#zJxm<<W4(5=+z?Ksn3Pcz#2|T`TIPEB-u;+UF+v(&rUJk?oLhP;
z!s)iu&G*&7#U+F)RwV>Ql4F<<ShQSxa<CsZV`A^%zFb4!SoU%=U=rC}{cu}|s$4l}
zFMnY;>u1}ou3pT)DbX-O$rn&DWo7t$a!5jzer=!;PW8QQi1X1>gtZzw9abqc&yMt-
zOsy+kS=Y5xi-0{OATov}e!0C)W`pQXYm~2l;LP?z!5#nJBPh>fb$=DzZqyrhuoIn1
zeuP#R7Bcyx0Ye*I`)bf&)a98D=*i+|I#MjYy|Be4)HKO7Uzk8>PEy$?Dyj=?gy=pC
z!O4y;%keOWH1S}&CDzH2J9brwBPCno_sN<YGY%SNV#fA5l~=`e?S*iI2&d0lYe#uz
zHWE#jrdlknZ-+O6{>ep~>OYIJuQ_T*5iYkH0Al8N(Z+E|us7K7eBiu70hBYPLz5;N
z{eIw@a;yqlE-J4{%WEo~+gz1&i#!-h|8m3v?z!)WHa@K+!<t{Oh)h@O@aRd++^Jmj
z26667&(#&>02VJLzh#0%q+eF0lq|T4aTCxbrXy;2vcZIoVF&hS+^iHo(Wvwu8SD-_
zL74hfOmRhTX!+|4TB}oP2cdRMFLi1U;_|E&jLu7EPa-O+FNZ3+UP+X~#3@6x^-;~<
z5P)XbGEKiOy|*gvKr3LP#T<rJeMcO<V8o~_wP+)p?QtY4u~1yFNkxEi#~m}n_;ELJ
z6z^-t-wf8NsG~s8(U4iNXn3_d?NZ@&gGherBQPxf*FF7bh1&pFB;quk4MSe*dwnaE
zSjGRdp0k3uiN&q0*+Vhj{O>{pK`5(I-CqKE1rqC`z>_-TR~d)FY#@PMx@gI{Zogrb
zBl44C=aPTnbd@D{nVdeAGS_X5E&|f&2=pH8hJ{6+o!-kYZA>9oL-e<5x1P*PSYFn}
z^*Bi><VT5Vh?%aHtR`MpqI>fs?muuF*ltFM*vM`dA2ofVs`x;f+kA}_>sTiUw5@aA
zxY35Q&!no#>B|Mo#JscOtF3EUA?{{INY$VlwUeRsJr^4_j6%cK%dQVNBurg9=?pg&
z4?0|-jy{e7!e7pHa8oQgo=`5?zrHPFNmSR$(+TFL-vQfbz!9_Jyk3w@S6bk8VC`Ko
zlWZU>veI=tbkzpxJJ9Z#kn8~9dLW`q|D}@wsZuxuc>Dbr%YQwO6^*H;P}`Syymy&-
zg@oRqe&I$1OtT(xzl;=^-`%&z)>yqOG(YUemtGo;fO}V)jew&oCcTL0^{u3>D0~IH
z&y<_hFG(eZqJ@aoL#$EC+rH0JM_HU{S2D{r7;crr(_BuVJ>GJ$v(VZDt{;k{<3~6q
z?qcNaT;dM+6VD=|CuXus&BPamCyu;CONyQ8gMB$xz><b8Kbe{(tSOuWKcyk6J<0EU
zB#wWnss&&?|JPiT{=W`)zmqbgTJuypO!n<__xnrfYRcb(Xv`aA>|Q6D)%@HHNX88Z
zqtqD&f|f+z0rbJ-Y6OLIe3GMyW+Nfch)}Gz))NMqnXcL<FrS53Ita9HD@Wk!s59uw
zxMqQ|L#2T)A4$Ve`{&K5g9$|ENg*PEOyLY{sCbSXHL6Nn$I_S;F7g_ov2zY;*K9L~
z_OcUQnwfTObP-dc{2W^N;DqZB8CZ09kkB(8&P-Wpfa8uWNpMg8A@m7K88Ek^rD5Q>
zL<>G>-`~S<JtWGGlfQlJ0UD|Ld4&A=I3?9<31$-y0#~fHuU$*}B3adPqPxaY-u5fV
zW;CR<J!F3}CnelmI7uOeYu&XQqFq+41EI|yLSi&?7D-&Z6?U$%0bmH#y9Ckc$tbqh
zq2PfFZYh7qXr^kD0Nfvuvf{}&PIEW`{IB};mIW(D0T)OzDe|UY)>4ZD`fY3D8=m!#
zFsnbmql$5fq?9=>#-WBqc}#W2jqtH6wMDfY88}Wa@m=*@p-K~q`56x0iXlO|V4sT>
z0Z_1`p)HF+{gg&qU1F4x;yR}tS^gXhHGX1rCua^nV)9{?ObA%W{N+_D7aLz>PHge+
z0B!z|RvcWyNS%BTg|^`$vDmGwGD5Sc4CoqV{8d7JR(+`mcskDbaBGW2R^h%^$W+mW
z$*Pk~=J;Uwa2W(PXjwe6juI2R8$m&vHh4$l$&@z@x$hxE&{ivn(;_yPC8U3YJ~le3
zL6DX5#|t+b1WPlp*?Yq<=Lc*IwN-dE#QS?Fq1~Vvt3ATM9e!+-E5q1yzOh+=bo!3f
z0!r%UGNW!h(D9xVlA`ypWmW4CKM!s^z3+sXk$x$C6ZELA8i&<*N4lab4QXVJHtfxV
zy{gm2=EYx!Ehem`B){Nv<U;kQi7sY9xYN!(6PrYg;9!2~SgWbo7B#2cKDm$(Lf)Hl
zmEF@o3Kf0h**E9Mz+8fC$@%U11e!Rx$xS6div9`bC)<n`%5$+g_=h!QTu9o;L=qg?
zJMn&XbX3XwPARI84>OhAcp+=X5W|4p4dH`ra`=Sg-K;^6v6}NA&T?pG%C@F6gDk|D
z9!t`eXVdpqSJ<HbY<Wd%Sr86Rxwn6qAuF$*Cm<U$^G_cA^)PckD{BQz^setSo%fFX
zJ(y~r=~4t;TeDM4BIUqq#U~QYUFq1?LX&NQn94qnl6}G^M)Sb~b&yOIINU@IRPv^Z
z?L>lcs7BVQb5$eIuYQ6Y8;vs%r*1#QjkQEK4tUL$xCeHu&D`b-kX~Nk8LjYuW6Tt$
ze@S%L<r#{Ad3aRPT7utr@OeG9s47*hZAG@wthX{Y-ce+Eh*ncMt(;NW>VH91n{Lz2
z{KD6?gK_PUGlQ40`GH%%*{j#;x)E>GNnI>K9qVU(mn!lH7Db-AL~v&B==7|NG|!4a
zGr_b`(KRY5*0c~lyCQz4jznOIOUtRQ#LmZ*LipdCqu^#A^a7Ny=~#HFQH2}9k}i>4
zYD!w^ddgZDyA9fDYoU%RFL)ttdnenD<R&l$nowAVxAszl5Id0^W`GpLYeSGI)QOyK
zzcDf3782){3E4Crx;WkCwsaa*AHV)Civ0mqo-Gcg<qg7$Fk~D}a{%eL92Gf}aaKLS
zPD@QvurxjY^g4wAm4_B9*}I!1eobb=$%@-*j9YvL>im8Ax#WKV)YjsWg16{_jT%4G
zvqQXX7tNy4zN4{Rw*KL&*Dkbx0elwS1FdzN9b%ApQQr$?S-Kptal3=MTTUD%g<GZr
zl~aPxvEgAf9rR|zSU~`<oV52aVpq)(4cRw;3C??My(re{xcpx6E?AGS7TpA{tXap-
zYam40g)be;$@Y~Isy=Ji|6~tjlmYl<kMClPUXxn2+FOo;NPQeS@=7*Mtn(>LE+}F)
zEG2;tVxGSo)2JMd*mp9hyF?het626TPrSm*Dwm}L3VM|9TWQguGh(0h&)#YHyx=p9
zSjItAo_Vj)Q?@O1D{%!Vc|pK<8+#KE4XbnaYt;SbV4FhV-Jy<zPc$KM_7QA>{m6Ip
zD1cv%ZA6EJ)N;LPu=OytM>IRt`5WN?nc+F%h$-cRBIB{{Ee|yhw`Pe~5}z%&chjc_
zBdqnyil}kdfh(NGxk;D~XUl{2uPPOeV^qt-La>=TG}d|rA=m184gE=U7u}KTLPbBN
z=3+vJe^u1y3c>%6Ov|JQLd;tyr!57i_Njg7W4gN@LLQ4<e!*LnP)ki%XwuR=nfVYG
z4_v{US+M|xg8$T5lo-nUz`p?QShgY<{gkFC+&WzXJwO?APnK1p(45Txwwm=+zhI~y
z>W~HP|8hT%;r&xI56}k82%yc@*y2H!-yw!+4$#gdggi)2QP<W|unG<IB-s1HX1n$7
zgJ&@|ALoAI^vJ7akTO8{^jF~72sO14D1GQ7INAR|0wAN{`KAn4AG<}E=&v<X3|g-f
zA#ez0U}F8C*`o=Obww&XcabzxBXU`*%c-<_vB@T%oxbBq_0aVh5{t&B#Pu={8H$?#
zYIsY6`=Jiv7ylO{D#qe~)Y(b9M1Gx!LX{Scpw__Tmzci`_%#(Bt=W#CoFy9bdvK+P
zF~9tDl9~1<f~wN9DGbL14YW{-5l}cHusD$?ztWY`f+-9|&kw21sArV{)g7+7<|sm|
zNX-OhxaLaCDtff&1N6X&vXCK38-F%kgx7_zi(;#g23~!c*llxG9VvpIHt`4<MBJE>
zIjEtVDN(mS9!&WD?)ya3Bqr8_Xo0w;`&xoc?)<<SH{uQ`>MipyM7Xu3QohH>_1Fh@
z_qi|ltP+{ZQ)I~NS6gOz+Q+?xGIywgJZ;q3e@Fs>kC>XI&dD=)&{+7g-5m)^Yf@oP
zyi#3Z9R%@#2gJwl=r5e_65Rf1gu5N~Ir%Q!=TDYTFZNpD2fJ^rT`1Ih)S_Fb{K4M6
zU=1|l_5-ZZ^<BI2nq6(Q%Iox5U%D~K(vP^$?`bVpTiDz>g;2gHIQ+!TEs;bpd6<D~
zaX$Tn9!D|``hYj}XE?nh0fnW)O1Zzyqu~Dp>Y0T!x|cH)&&1hFLbTaGTN+V}TU_Jd
z?34ToeiZDwELrQPNgM>GN;4=UcBo<6IWh~08W5(H)ch_o&HqdwTcPg4COlFi>kkEe
zD(Geazi>)v;%3SG1s1r4Mt+1J_ir8NXiL@<_&~a_cP%(n@Q-0z;VL;uVQ>}pfC`QD
z(>Ljmkpmrfu$Qx=15TAr7YjomUQn8ppOf%V$QSTyQwZDEIK;Q(WuNepT0$g~!@J+G
zPx^T5_pdx4O(e*NiMN-k=Y+p)!kp%BRpF48kB?WMB^9_5tFIz{<^3vwKJTPK=OoD(
zykuZoADek1ej=ve0o?L1wy|caDW9O+0e@x-!M#p5#h!k&rM^p%qNVYbtlj_0M)eo<
z&<uGp)8GOEG?4g~OhAv$70kUY6vS|GK`fwtEi~+hR2H_cQ~Gym!j2k2kH>Iph5z#D
z1BWdtOin*_nT?fG@RMnY@)g77zo@svKu3*|=P9SKQ=M*DAIqWP+X!_9eDX&rIp~F2
z*|Ds&7MK&Ack4;<YgR!U+z%jM$Y;K+iUggNx<KRSk)k2MmQ=RJ2dQG6{45nS(2MO1
z+d=~q?KUGJ?b-J*ux%@L;{PSa_`t914Q<T5Io<FxmPz3-Io}_M*J>~O;A~(b1USRP
zb<t+#HSye#LBVrHZ_VLFHnmI6{}l8J-!6$0<gya37{<M~M<#VY7#YSv3N9A&Xwb#q
zpLi~S%Xc?KMji(JSApNWHo!$>zxf?Ko1ug{xk{7{u_@gH4DQ-(;48cittX`KWc3;^
zmg$BFAv5#or(DSK3X#>fXjD5^cVH8HWghppD`lyNxvi2g<I+UJa;)>BYlCVPyokgE
zTd!=bYx=z3D=|t}jp1Z&+^hAG0+02NPed)qn$rkK&GKrO-Kpd8(`kvOkC?Z!fhbj7
zYgM0PG%d&CxX2gm_~l$Y%_L`HR(Hyyk!E%rBRO>d9{l7c_c7Kc81@gdUCKs9o{vt@
z<rXUz${#>*EPU614Mqmr?XC~meXBH)h^e!5<Hj`Ls51H;KCuZV@Vv)W%T>yT$mem9
z9nR*Psm3Eh+WP3l2gI!0gxi9-zZ{=`U9Qs7-xJ~#o)<y(Is8uZ>S}SZQceCet=$HH
z7iT&5X~fPTb$?b=TSk<tV#jJZz|G9Gc^h;~HIWzCJ`}ylrqLA_X*w@V+yHa%GCqRU
z|H_DBYOOGueP#6V{S-!j)LbyjTyf$2cg?C^s!V%N&x`hM@qY&=1}>R)jsRsBj}Or?
z-~e*{`HfOKo3|NMCGZoqaJ^)OYPV?|LMuL|Cx%1Fst$>yUS=uZ^fhNXOf7zp#3bMt
zhR}Q^G_i}*6<plFlnlW+yGI_k3YENQO9vf<?Hv9<zGE-AW}!|sud^;PduG_g3oCO;
z=b_sERpmj@oKyU=y{``>2%G%1Gj&x`tH2gS6~>xrObp2U>DD4@1jl&S>SE(DCGcig
zEoFsnLGNKQ&Cbk6mVyM8DCW;G8x^2KnOh;Cl@l6Yj!DmzSb96&p0q+4&;Nga`||3>
zLq~bYq$2iifWzC6<anFl<&+IO@X>Sr$&`I|5ff1WR%1o&*b({6l(~LgES8pju{M>w
zDA;3=U6hcX4=fxhnr~VLJeu+9Z;~y?2ZVA42<eyNenNjVlFZ7LaJU=f5^NeoMMhcx
z`Muh)K2Ac%u<CbH)7>yai))Wt&gyOkNxGbD{pf&kloX?{$MVUQ{)wNn!+lIy(HEXU
z4s9+$cU?{`qEk%hbV8{MQkA30nq2e|-HoZ;TY<<W<$)GymVIp1gydnNaC6bQa2<SI
zAMFWy10@2uBz*bO+oYCB?>pD#zf!u}f4mlZy7#d&cI*S_?gZF@x_xx-!E0ukuq>e+
z@{taR2O1^g^ynRt6}-&q{2Z+$D&8v(*y1zNfKWF}KV8t*<P;z)$g<j>)yObuYgY57
zSV;bEcCYB_I(BYI3w-5GNT#~S-pq5XtEnVBW&{X|SnSUbU5^R8`uqEX)BSK&Q#cws
zBoq}EInO?`-!(tv;K1~@8)`!s{cMtANQRW_6f3dK77WK`$=>{tBCn65ku_LTUUb4)
z-2NwCqcD(7eFs+Ap0tNg(M5Mr+$gMvwHBH_p(r8~HkFFUu(EBVsKCa1c78%M!MSo$
zXI*<?ch>F;g0g_mc0s)dll!0}P<j1jX5+e6UrqJis+d$poAD0MgZ<vhUe{<sIH%iR
zwx=%`2fkF^F9^xM`e29_t7MI-#2*Iq)-^{x9obu%yvRCRYUtt`k~5x#@(<{<%mm<=
zUh+@+iax7V^;XR<yJ7RD_QnzYDc2oG{HK5fTdAQS@9M}=d6{4Yn`*pL;ZKN<s}ZRO
z!H{kcwa%OrPh(Z=x!I4e6}KxvX*0^$KY4EJt9>5ponB#YN#lx!$<j|#rqoPlNdoB4
zo;W*)g@x_x+*NjS&juJYWJM_zU$6{d>g0(egrr!c0ftGxziZa(_aVfVZT(YZVtA&d
zaT4;qXYoR9&d6#^Fc8;h{N6WCtD=T#mdIV{+;fPvoIDX}XTqN1QZR1Hg*wVSzQ>k^
zZalOfjAEHP^_@9+Jr4o^eT50aUwVwrBZ=ev=!zugA|F*aPa2_pNXM%i;OsY|M@U<W
z-lY2EgdiQe3x0?rZXJ#dwaQs6KMO7qmHXSL@cgz(P38N|EZV}`;<GX8u#x-~a}#*P
znv1UMA?q0wSHApwQ3%r=*v0fT4n{zuxMqfmpkevPkJqqxF_L(@V<(HniQquI<fmmI
zo3>W6T2YAe^914epVjFi%$00cT)(H0&smica3-GehS>qVL&mXNCPs1M+hur5yf(f$
zbAU*#oZ4KniTOySLRg~hj$#nPzLAL}H{VuG3$fN1OiNFi0Vjm#ftB%<1_TWi+H8gP
zrW%}p>a(z>@8Nkfn(DHF%+P$VN_9aFFb1AsUdV5+Zj!xK6q?~gYv+=3km=ZrFytud
zt;y%Sw=~#M9==9?3H~%4UGPNiUjj#C#<b>=T3p0oX>pD*F;-LUQ#*bV15CR`5nIqn
zY6}8Z8^8d37(e8eUZ0IA8N0Lu`yvK$k+8SmEupQwDnr{<by1s2yGQ0o2SFsP@{yP6
z|K2D!Mx3XeGQ|gyH>2@?*}e6$)Dmg`l`aC0B(1QqU}Lbkyo2k8+QPPO{3F5a7~sFB
zWXHWDvL)dsJ>T3`s>xr&RkzgxA;SWjKF`rg$v4e0Cu2yP9HTBo5S~T`8aN6SA8y^8
z3RE(fBBpAJw=<9ia~Okt4guQ|)YP&8OLdAozQ*edW4KW+<X`!kh|hcciAy=NmVV0_
zjWWU|?B`?F^eoVcZk^9ErjdA}&!!qqlc@AGOvh!4aOgyZG+}Mv-c6}vBj!6mVgE;k
z$QY`r(i9tpZ}!hqeqAWY9Dozhe?5OEYJ8FxaWQ97(qJ+(Sy)<5D+l0{sesN^=M{E7
zOc7XH1UPgAvB)?d+=@{n*p5^}UQ0v0V(Zx=G~6X9`@0DWq9ER{fI{V+jcAHqgwS)Y
z=`L1EHXjfY`4#Li%2LN^boWD(MdBx2z01Ar^Do`#mBtMwdXgSQp7#adogKc*8J3bI
zl|@#qYrEw0n@(r-1xP&~H(XlqE%xCp*?lPVv_Y0v1j`FcM|s9<gRQ}YAiazglVHH{
zY<sFMvnW=ph~L6XAnn^&YD7Xb)PJnJly1U<y9{I3-D#i!>Me;SA3AMQqaOUu0LhZM
zPMq`OiO|nUAU8!Dd6<0jS|N`|=5y`jC+F%k(haj7Hs7c}5+yQ&?YNPV(~esX)rY9d
zdeET}XT*#E<)E_*Zv({}jKlESm^u^_KV*L3nR~kduOC_Q94=5>Q)2Bt6*Y@?RWHtc
zYaoO%3*Tl`Px`f-Xt7(NO#7Q|%6U#SCcK|hF2o+Y_b^i}Cdig<Lit8kJ$GQH2Z_jO
zzL!xumpWg~=)pex62aZc<FWugH5nGw$xEbT<cjX_8BI2sQ~?xe?q5W9>3nC32eE#+
zShSaPTaBNS?C;_(^ce|QS<^Py);orsK@b6av+o0#ysoGa{auZd&5xwk7O*3Bnmgi@
zLNGC|d{yV)bIn6!WH;kIwvuyQ|33U#YG;PJpNxT}me#7~E1bAS%kURT5!f+$qX^tq
zT@CfTpenWLl^s9o3V8zfWuV6KC(mjmZby`aW8k2c;i@Lpx;~;v6Gmt&h!V)L`et9T
z_scAG9MiGN=6<nyCGpHca3+H#BX3*+xa!40z~Z^HJ~QB{@6l`Ec{o^t;(d`&SaVHR
zL*rN>*|yK8CDhaD9xgWn#NVaet?CxbG{WA~_If|C4cc{b0r}hNEFgNQOy|4Zk<@5f
z-ajxHQ5=`xB{S%kL<xg<nxv*jQgK><&+l|zXMo(veMZnpTqK3Mx&y#)e`313fgJe7
zFoRQ7ShO`LHqWDMGUZwc|A@fzW}#B9(>V;?2mFC7v;gd`if%C?S>Y51G>W;gr%0r*
z2J_e@TB9w@u#(W*_U8#;C=Ga@?CHvvShgGBvM7Am8dEYXFw0vAt^cFw^~R~RpH@Ia
z@7?lJrsksdX=fG%QRc(Ij}1{Q0WU<~db^hxmY8g$G*{V#Sz4^_U-LjhI|PkDU|k3i
zjZ~F<vw2~=9BR3P!Ei;JPY$n)tsy5NEMlT;2cxM$E@BP@JM$hM>o@K~7Y8bnR<qj=
zOE2*vivr#Y*Fd@PBfb**tGGw0W`uZ6`bDdE^GUo!Sb-8#J(_8={HX{c(?vb<jmB`Y
z&}`eUXsR@=+OnH+jL$W_YgP_Tu_e?qGRFs`ZISZWp|tqDCb=H^OIo9`g51*O7VwKc
z<A{a?>00~LG=$_sn?dg~j#lsbR3s!S?SWawfgD|tA}`QJrL7zwRDi49*-sflQ%q+0
zM5HEoYTzYC#OI0D)D>@ua@?b=9DgtvWH=a}Sjvmx6h`qM!;6~7b>jozsFR-ACDjme
zNC1~4vvI$MywGb_0-S0{M+l*<`}nY{h{XwKH#aQsG0JBUNk4_yOe7JG&&;B256<O#
zvW)3`PEl8IiGGF6Tqe6CLO%^7N-3|TNHkVIVPkjEkXvE=vhPkE*G}c7&VU7p9m2%6
zRz4;R&{K{U2dCGSYq1`_@+We1i#aqSdTPXDdfgFP!(A_jvKN8eUR{4Ng#|cZU_?`%
zWm`;kKEi$x<)0#1@D_)cCVX#g=O0bVo?$u~9sQ~t8vbd>kHGGnxTFFq!ISk$OFf>V
zvo3EHLY{rm7*VfTDTSt&>oVfNgi>k`MBx(8N~yW&W(x*pb=|_qNQG((t^u!_<<o(W
zdQoC7uhkpFSWoBJz~3Sgz25CQ&pA@KV%o!&bw(7E<{m>H*|Yc1yOv=DZ=cs1!!w(4
zMF~>+&q&unw(x%tRC82I2in|(De&M>G%cqJFc3-A7gbmV&wP>(yR?i}sRyH>os3Yb
zi(*PmH;JjuU<{%&EqBoY=m_{YBEpm|?v5&G=J4pxz8hn;@3G?1ubpl&9^wsH&0z>!
zU~$xs>8e83MmV?LNkBcSEjDLZjp$|mtdBlSpPc!0YjUq{O6zT5uhF=bv0uqGBO4DS
zTv!7BJKG~mwL3^Rn$wrde^fj{i$#1YltbmlK$I~x&NZx%8otSnLTH0A5DMx&{Rn9L
z*=q{A`jv>=Q+YiOn(8|N9RefVX>YxQrK9?<Z?8#NzKuN<u|XpmN6MtvbZ*34cZdnv
zX^^=5S{L^lRHHqMEBvb<EqBEg!u@?tjZGHf`7+abx&S-kZQxZTXAnLrcJJW5F3e=b
zD9UN<u^Sw~YD7w?;=Mj@?Sz`h+BKSjAua&UyC+b-|Jn}soB|D#I9DBuZ66|}@_km*
z$`!4`r=KJ7#3&7Cb470_gzzOJn6L5xZv*FLXK)K-v98Cfh3m1}hLO644}^H+EQr2C
z90$oHtitPSzBfpRAo&#F-P^%jgK$_<)dXF~I9AB?wyFs}11k7!q;$FPdLbVyJ|8So
zH7^(vTo_3t&Ij^`pg&Q~!90de8MtgrPGW#`G6`?y`u(`JUjP{&#z00)Os^ceJ~rpf
z@NRG{EoQZrWj2qS-3F_ph)Fa&jve&izAc1%lgsy<-2-6kSq37;eepb)y`@?1prC!l
z>CNu>VzFWM$S*I*)BnQG&CU}D^>I80mi*|n#~s)srj)lqkVYnDoE+oC_@4Zjk&5rU
zt5f5NiZ*L+B+6D|S`E#*&pG;SL3?8?wJ=;#YC?1p#Gi@}<onj}wI9iQg~6i?{*m)7
zW|hyh#GzrDPOX5Zbj3J!S*vZWw`1|~X;f3rCMOn{bfbh{Tw;~M3Y|PtO%%EdNge-d
z<~-4YvPo_MwLOy@g?bIyS8#2l(V%_Y#kdPhKyur$sI(_NPAk8(E3Iccl03w#5X{=a
z=axd736=YwnNsj-(qd<&uX;iDbJ#C#a1TS_WAxR{_oNE|pGSLD-6vSI+?1IZ-^>qP
zkxr?MiYmQ3b2n!fyZ9oo_3;Hvupg2lm~LO6BJs5b2bFNF<$oZ^3{`1+4dtX4q9^p)
zCe5oGc5g^j>R+{VcpM3oUQ}raB*lkDhDh8X;qh%`rxP<m`P0+PlXpMhW$^(7D~*XB
zkU^&`B@2ciR3j9T0OdKO%bt^%)T<2{@mu&NgLA-Y2=qUI4cV}=+t<`T7^Cvvpw3&`
z_+t@3)pEDdT5n3t2JNBgxh-i%NQccRH^V$ouL_;CFz=S+P<t~A($~J=2VB9Li0=j^
z>h5mCLY&`yApab@-Rvh=x*+Uj0d+7X_ZPP1;&{$LZXY`fV!pO@K^<bDoF5Y2Rf5Gt
z!cwiX6SC>byUds;?d5_>h<b@-s|d3SE+i{Yc-YA<k4;+y3;2<g-|eiE&4}OaH=RD4
zfb)J{E!hOrv47aaV?cVJN^J#GfpM6?dbp!U#~N&YTPhL3F2c~BZ-Q0}V2);OWdq_G
zTT^E2%AunrE9rLJ0Rl9+@f9mW%{Kcumv)zgoV%uYFoP!P71G*^nS7G+n_vfr!H2HG
z8DkKHluI71?ZcKGCq1F_+~*|dLPPM(eEb6;TIl_KQsH=Dqg0AUz_;C0R{26TF1jGt
zSQDI|Iv9@u|AIjpUyS^_d#>x{VoV@<a~!1R%l}Mi(5XugkE~cK#l)=<KR9_pZeNsA
z?exlWy#4+-J$B~=B&3R`(;mDrF<Q)}8m(yc7J--*(mGYBa<*m#ngR&T#CIIgm{b4n
zYe|h!7JQP*T0?{L2nv|I-jp1>+VY!AeDS<9igHv2G*A66Q|n6+r|k_0j$RCL+0;xQ
zBl2R%0FD1FPdp(?DsGGT0xsR4i?Vv8*h9fYGE<nl#r+TvP7BJ;51wAsk#1*~p<c{=
zEem?m+gsI~VqVu<axH%NKZ8abw&L**X|vh-C@Ez~+E46RcdIPq>vl`ZM-n89@NUbK
zIv8v^g*%<jY~f<;y5+;XM`*Z>7l<>W(9}}xHbry6uD)uS1j%cD%sa=wAZ^qtCxrF>
z)hwwx#0(s|%<0mju4LX&GscE(L_Y7|^rid8XFcOEw~c99dkph%pT6<$t1OP=@Diqr
zMdyJ|c&GGp{14^e^KmqVa<ORd!2UbJTKd{a%s{A>HP<ZAW|&o6p$3!Bd$e(7`AM#6
zSs5Ng%Ugvj5ge_cLp}4XiWjSKQV;c<3y~P_wDl}Q<zZE<Y4MCb5D)yY*KoB{+K|V^
zC7ah$59ntNWIlaXYUP;L@=DtE$PyQ$>-k!Qux;QXN`$8zsthq(W#!e5${Uj0ayJF*
zr;OlSPD^jwd~6@x^iUV&jih?OtGZ|kQkKz9k9$usjnMN3Om839F6+w6w){!I-ZX)5
z{d>qf04czh^x`gU`8x3Ea+y=Xd)rMhHKOyo=@F2$q3X39SF}8(2fLENV(*sgOsz4Q
zsDuDG!l;=4o}fg!X2FQ-bVL43yf1U#*02v?Ya?C*WTV(P*P0vljp)^d>?`kbKGPZy
zbf(J3%BJ3Vy9|VS%BmP%=)5REj!LCNO=CcZeGawooI_>P8pB$MWF<0Dcwm?5d{8sv
zK-pK$f-9oVSq&An==|6P*iS?tZ9UcbC8(Q}KvuLsdCE(i1B=~2j0JFh#lU0nD3il+
z0K@s=$oS}6A>|rl?0L2#x`Ps{e0AUynSP22Snw(4J-<ma5Mm+HK;Lm8J((<^_|jYl
zQx9hPq)Yr}&mWCfi75eOxcHEA=U=!WruAZtXB|!nIGP8{!TF1P=*GO4rp4Jfv6;mx
zjJho@sW$Zq`ieh^^Dqa@;$*V2QSEK8=DG{5Wt9si8?wLAr~>B6v;pS0h^w@DczyFY
zI$G6!5t($L^4v0l*wwHi?oz)eAWvh5icb#Tl6eP@rcv(1%y+O)`N4}-GX8&&pt(_7
z56wM3zx+GvM+r)J=|Xcp-ceP`X_*)52VHWV_RQF+Y{Pp{c%DS(8VCsj=yhGs7deX5
zWJd`&NjDU$@aNE^96VL~^$E(m{SyonD)+SC=Cz$45%zi<6t`6ki#MJ68!OrDem50o
z@DcMQzM=xhZq$m;;?1UC;}MZ!Xj|1NnA$Kz1&v8^2=~N&^|FRqku0p6RexHn@QaXt
zNpLb>Qq7d6X<X+V8R%5f77amHphhqa?^b41tz3ilIvBesq;tRyIGZcI42i$<o7=(%
zrNnv$E@{Rs4@21l3m>%R;p0Vsj4tBnrV>LSJ=gZ;*P!~<9)%|%L*jd?s?AKOlz3K^
z(@ZB0?wVVmY_h*TxWUH-SJ<4uVoQ8$FIA?5aP1O>{PVpb#pKWOt4_G@L1^3oP?kk*
zVlf_rVBYKJS3u00u2+fWo!T;&<IUw~bUKWhp7U&C*l;i1+RkmFr26fViizuQ>Gl!f
z{z@%2!7%!G_tSF%4P66+t~|_vgB5-2nUb5seObqQfY@j?B4f)6)kC8;1HO@hKQ4#G
ze5#LS($_11wxOHvr3JL3j9qzugxkngcBQ!-E$6qUME5Go=7xNKJ?L#<@PPBUQiogc
zzn+zh$<21Ks!fJR`-0eRP}2&FlC6N-gX*heC}wqgzoy<zw`gCeAB0u;AE0SpJM?_X
z3<ivh^$gxNDY{JsRL0dL;HC+vs^^!)``sU<Ql6`HWDQvh{$i<qjV0^pz&`E+tDn4l
zgPt3Hm`S|#5f^Rt|Gf(t<o~ZP0iT&$*5?mu3P#-!KfbA3nkI~C7xvy_L|iIh^A;-v
z#Nt(R1_^P$=mCr;g8)%1CHLlyyg8ePP*%E<u0;#f8w^SYS7?=!snNO-3x=Hnsknws
zRCLBu(wT>vvGAPk1z+{Jb+qFyZPIuyTaQ^7&>sV`|3Q*gEjcDSUt%4{ei>1t_r>;#
z7yfvIn7a(!*CbMgZ0Lp#z|hlF(Z~{X>t6?qIv%Wn<(Fs`deHlb<RvGpG8hl$%o@W^
zaP9K6f=q*ChUnz$?TRbW{z90vi3SU+_J*Ar%Ir9Bb-j)a11=`Q1b)u5A<3{}^_9_d
zUiLy1$#Z%FT@||lWD&WmZIZ;doYYvGDFA}gqD=|}mcVICTqGFzIuy@}FgT6as7I#I
z*IgCZ=1P04K0bDJPaZ7%HnGv*Ch>5dh=Jjbs4+4GnzRKXdjnmCG_jZx&yMRp^*p;<
zLK?y34;B5l?E70c{q^n~=CVNn&8FX^p9JS{`L0=&01W2K5<iWr@=2(|xnZ%_GXe*)
zbIay^Q7thG4i_Q3DgNFbnA((cdh<~Hz4*AgHGJ73(<m4Z!=93_*I6cxeYbbXmGOWb
z><)N+I7Gl+wk&Y-HI<iPnj_noOhLPnGe7Y%C)t<W4oE>H_13Q6x!Hu#ZXp_+LvL;Q
zJP!d*FPD@&#v|+Nuk}&(7(qM0<ClzQ?gQDvGU2RCDT~T}m;T+58*nphBsW4=9TJ7q
zX6&c|*b!(9*F&gHetjq(+cZ8sA|#X%18%zJsHRnG!V{vp7(hS{mLNe978A5_=lT4B
z-JR?=uBo==Dv0O|$wW>jUuJ2bKZ>`IMF$hpP4ZV~M#sehN^6BB!+<4d4?M<#lKCm;
z$1|9_OQ0)PeA4<DL^$FB-?Ck~=Woy(E>r6agd5MM4y*;8Rj(SNY6rNmH!sTe8hnU7
zYp=(fS>6+(!q`GXc(^s`HA*dqTI%}K&=Pe1rg9M`x=EqhaE&Dbq|J`g{<ypzs<XAd
zUhZ|W8bJ2ERCVpb3rj{fhK&!|t%oztY#fo%&3NynL`Or>&3mgte~%;yI(3o0>>oH<
zi=DtNlaO~)<O9FIT0hQ{Rp%Ei@o4sNKy=1a7w;5{SC!jbyPP}tZLjI5Ix<)z+XQmZ
zz|aJF{TvI>EZQq6S=!3GTRJ25&DNu9k=_;0KIU<NF!-68#F<R9{_}V^86U-Rg|&Iz
z-REsiCzw2o_VN**a^S7ZQD60*(I^gW615j^>>6SonXs?4Bu;Ol?=$`@4l!Hvc(Evg
z5d`1}NrO*{!G?rC_+Z4dGw~UFgzo3*X9Fu%v%LCFr*=$}d`|=3`;+9|oR;1Yuqb#U
zjhUnkhw5ahuhCCkZq$s=P5Ze(B+DEMrbHu_JL&0&E^vHeb~8K+qxSCmq$Hx@5BQAt
zZbhs!(56A+Wn(GrzRbna6T7-ZSZ`M%rrPKvTpom=WFgIKWt-x2p#@!T2`*P@u9J0}
zWiz~-j`2QQk_{vhUAs~L$4VMB3tHJvcwW!jBwa)?V*H;$dr62D67Auxy(JfBgpd@1
z+p~Jk^8!Nu&a3(c(Mc>YufTQeE$CGN$tgD0N@mYoiauHj6Hc-C&sZI}I+FOU-=+>q
zflb;iu0dK_vj&p$B?<bmkB20n6V?}zALJ(5Qef!JT(0OCf99$6coM-{izItd$l*aP
zBB}%miAr)1w3~!}3*+iJiAM0*yo?dIQDiEUVx2jm)>YhD)#v8=7|$#WiMtYN9$gq|
zjbURK@<#L!sPV)IAW4o4OoBC$L1Vpk7z=_YPJqx^h)Dfy^l3MYy!s71CJH2~7C|zx
zjlcxznFPW2Kt~t`@z87j!IKwAzLQnKBJ!S4%Q?>if1xiBY;v;JDmX3ng#)P3LlLb}
ztL~D6r6PKySJ3Oi+#3~ih5jP4Z=VZI_{itw2;9Rg0t;;XR}l~fOi&v7=(HTT8g1%k
zB<;O`aRYit)caQ+qLALAnH4};?8PBaGX|HZ3YKT<BBtQUbEGwV$wcS4gG@Qc>;PoQ
z5ejdABRe;btB=lbR(Wr3^sx6YU<P3iiy(*hkPH(vF<D;+8I0!%m`nrVtiTCI@%9uX
zJjC?i7pINGJbQQ%xQonbKhuC?Ncn#EB35BrzsGbR^Q7h&rztr1)MJ@1v1J$M0w^wp
zJkEy%OVs8pa-xbZ#EhR2lgEc#10|Qa%0iEhOWmO=M9R1J<*#_a=_3e)oLuEM{hi{U
z;vAXH0*f5Lo4-{9<Vfl}7>bemLxYRDS*NGWVIn$lv++^9S}?NZ8H}97Bo}^M8P^GC
z7Lu*W{9l65V%7D3WBIT7B5#Sa;Y&qMH=g~$-s}>pS)0dd?(1$1_^t!!k_ZQ$Em0uU
zb8Dbm`WeOyTquDaOzsjTKV4iyCv~0KWYSjnPtfu3<`O53(bU4?wFPpCp79AuBjn-7
zrd~4w10i?SbqQeCyd?V#%-=7q=sBT{Q)_1VU3XsaaRi4`P5JX}Mv14QqCiidS$dz)
zLPtqhFS^?rpvvnP;1E82&s32|>#ZY0%Sc~7y|1?-BeP6#eYs&6H5XI^#BV#t12Lp=
z6Ten#<0;O-&s6wMao(2b?-DROpih}g58<Fmh(K9a<hC(*i@ZVZirroKTFTjsbzKKI
zm3`Q!>=mWR$U$V3<Jf!8tn57w4vvm<I72pBWs|*%5{1aj7P2?lB72mC%%bm6-nXvz
z`##sD=YQ_!p1*rMx6v`}=kGt#P44_y<6g<Z{(;M#wLO8}XRP$O4u^X9aA$l$KAZe+
zigw*U=A}UIIhT7)lyoUEHP>GcJd+23IYuZMAu~cNvWKnHW{>@HId~R1j#B7vw(={Z
zIGzlEY9G33s&6#Pe+^|!dz8I+;Rbi|5B1cAHTB0Zp0X@fSGCgbhu^iO{rR;I>ZOG~
z{S?(|FY-<(a$^jPO)jIe4s}1jVzG`Dx~5<FDOL{4L`FvfdqGf)(jeDY7&|Z%p4@Jo
z5kB0c8G|09#XGnxy)S353fxj1qin#o^zB_Bi_jPB?yP8M4wh2!>T0~H+*;btK!c&1
z&f29;oc}V{>MXT6X(ZR_IhLYQ#Q3VwUGB#2GwO_T65r&3<g4GAl<sVma}_KF>ol!e
zw3YHHU&|rBe!!a-es%vEm6@puUcX$ahpGB=XXomLtYEI9s!L9UzUI2!9fUR4HB|P{
zzzVBS3C;A#;i^xRj5bis3-vYV*Sss2+_?(}`x&4`x@&S!>f0YC?@qcAJQ6xCDPwx`
z!cz*&Ii`X~{mnr>0TjcB$u>R;-?|_6^uzarOL<;0?{iAnH1>iI@#qCq6f?aW_C|d^
z3x^Ez^}9K8P4emI%yyAJ(CdAJ=TO01A5v6@AFZ%{JSEpN<qAfu&oeH6vieyp<0W}g
z)Ue`a8e(pH-MG;jzl_%T!v2nEY(p1Xt|IF^i}A#GAxpqB4*0ABsmOf3HP1;bon-E?
zR<|<s%`7smZqA&mUqsBTD&~D18^zvKv9L=x;Uy|0qgRq%T_5=Vf}wZB)F2q^t`Pf~
z-|w6Xe@M$QWemyZ7neJIrxn)3r<Z@K*KIuSsKQYef5XJqCatBsCEQ`EbtZcnurX8U
ztr^6tp-<k1?9Ht*h4nIiZLd{8tbARLUMlIKzt9#*YM&9@XNK)x{8mRj+p+u7^x4}^
z-%c4qW%$rsxmlVp|NWUXq6Nm~sWquN$i^E;_;hcWQ$vs6D=WO$U)$v1jK}kXGZMg;
z2ZC|3bmJ$*>X?<~UUK_rJ`Egc4)|P6#N%9+T!Fv{VV+sWC(mdfj+AZ946HxX%DT7N
zrAm1xa$6-oRe~j(vM>_g)AznA$inA&R^z@*ZLv8K=WAqzbN+Gvu4HwLAgH{i(bJw_
z4y{oj?K@c6Y@+07a_{3t;vNH;qgP<~)7REcA$%P3&sqt6x3V>fWlr`moWpTfMH$y-
zJD*$*D@2%3Z><#)Wqkim%NVji=()fwSWAFVhkSn}WS1+Zdhy3yjpob?QNEu#XTx<q
zJ?+tVBHy>V2`hJn_VXImxQLT6O&z$xzKPT}a8;0HJC9kp<dA!kASGxTB#!#;KP1oE
za*m7L{g&rbf>$yYv_cTICBd3WH&`(h={L^H^7Hca(>U?TtLNmZ#K_;7Lsw084$6b3
zF1Ngnb=LGn?Znwk=-b|IBBi{*#ftabP2F-Gp>&b!F1vK7Uz2Jx!>86ceEJE?viips
zMwebYzDw&V=LBstRTNk|e*!^+Sr43ANqN7#e1QDgv)oK#=uXzNl(`iL8hQL0-#7Bn
zL>c3HgkCOjMHXwmqeXay{HB|=t^<Av36B7z>aCwoa&G1Qn`trP!ii6~-RWkXNGQm;
zM0A#gI|!EV>#xlBTMGTO;cYOoP*J+;4{~dpmBWd2EF~v3T)TR!%(Ib(*X2;#ldj*Q
z?IqzuM9=jlf2PztPS{@malWJC^ZCyAcXcz5s&jJN;lj^@&hge!dcZ$MWq*j9$aC)%
zH{cs4_3QLm@5+4b3{}i3lhEp4yCk=pdq`3`eNm=Tq-(F9l*UU}EQ;}lO;MEV&+SK=
z1LulJ6Wg{ZeCYB$Z}r~5GcxiT^r?O5kg-yA4b|d|acibucT|ecj&WZ6-Ut2~(O7En
z^aCtkc0RIuMiMO25w$=uVQ#_lf;#1zQNbwiebqX?`WGHje(SvpvQB~G?}^H<kPQ);
zpV+2<hOyJTH|b|kvU!FkdPz{sJi4Bg>L54y){IEx?NH{?;7r)V=G$Q;;s?Ts)TZj;
zYhgF$v{;k&n!}W&v|3i#YL<-@Y2xm9GTA=MQ?NY8{jxl7tjL&5CMILX2;0IYY3E#}
ziv4zv>_qeB=M5FQ?5)7!H4)$8TViC_=47qt!u&<D3%_5`G-%_{UZ&}C`4Xr3j7ma+
zsxh%ixIg9I=pv4>63<kaDA0<Y#6usP_v+4&YIH%SjxaZFyWsg6rA`oSA`Ne8OpJ2N
zIFh`TpG-zYB?Mig<?@_th)$G->qYQX!=rmU=_|ujvfO%~xYj%<%OXQPD!vNXH4n9m
zUVM43gK?3ch(PO(_;+#(p6IKlNwQ|<5%ZOsx3(_k)pc8%@6%MjYsu+}&xE8u;u{@1
zd`5irUF27_9J(Dw@mm#o`mbkM{IfTyyvT<-@Go|=vf6a;X$KI*`#yVDms_pWr`)j;
zV81>=rS{6gFe8}Td>$;W9YakPGN%}BYvWa#RKMw!%h<<W1TpqcsbG*Ri@s}k#qFw4
z0@vG5>3i<X`B{V)mvz>ZI08-JVM4StnB;TeG|nA5b%9Dh)#|>m%ffb4zc53VW%M_{
z#1Snm^W(JYE$8@FOzJ&%b;O8AT7%$sM%e1XO^-v;qrgSy%D=i(`ZvL!yj-?1FCv=|
z*tj-jGCE_zxW)N(e)iSJZllqk8%ZtGJWj%5;7)&2`@VPb%~41QeM%xk9r!O?{VC+g
z`hao79m!|zHydD50~*<?ah<C~3((}SxBaGZZlIy5bVzRCfh^CGXU8Ok&S-W}g#_Kw
z{SMt}qiYQV%;fy;885z)j^^MUz6oEiD-TN=Dd4)`)aE$Z^xpF3eNr_J$GLZL3+VBf
z3)@QGik2Tn>MX7>m^H9QST2i&USV|{=2<+vU8SPj_x5T^CY}bahcZX(@&#+f^As^|
zOQRt+O2nTqN{PiNCnrK#vYP}9*48*CQ}R%$=E^PyL3rGAW)J%Ai{!_p&)#gtf6RN}
zr>KS}Nly};-kBjml|{T-25Ed<^X;xf49)J>h~2(NB2OTbd;Ip_=uM5A><@bvDu(u<
z2Xz!j_=e|9;w!f$@p+0<#Z678`e?53)?Geat<Gfmtn{Huc8y$eT`OHx?y+0wPKHN`
z8LjI1mId?=S@4}rfA%e^PCunf+26%pNhcNCt&i5UEa++{hp}C2&WM{WdG}*>P^Y_i
zDgUE)4E-?mky66*wIWYk0U~VoRmjtHdjhAm*8w+0r?whyBUP@xy~8+DA9XU4dWXeg
z;@Tw|Raq+R($Zq&Ab;Qi`o7USl0ttRYV?!+y072Uf<(M~4kZ;Gh`6`I7m>>LcW80<
z@%0wRh%-pd0zPF|q1Pg0lyBZ@h_a;aUA@&t;gt{@k{G*MC_>BU0OL*j*b!cZ+mWDX
zczX19^HsAW%1&BfQKfzmqj1;8u123;{=V?n$Hz3nB0&&`6eiS6rDR@0e?y+V7Sfxh
z$DNn`@p%ceWR|5{Cv;Yv+0mS}ov@*qj_M`@*N=i`g0;Q4RCANZyKnu)zqv7<*f7MV
zPq>;)1U&72mFCz0<6==@ZwjbOC}--gtvIKbQ%ma;BN_B{yO3M%c`tIbzJLt94ZJ^i
z?4&J;@sPPkf8TVL^y|kh{*kRRsai1mo;z!eoa*l)E_P2LjX!H_KlNI;K9I;bQ8Ps*
z&Fq3ZQBJ%zXIH1<mgyH<n>yi1D(tKKqF4@E)yEM!|F(bGuTr00|1LGtCDtpP9?4H4
zHv_hjB$UJ(%0#4}>9GNizY1V9OQLfNVv{7C*<vq6g5W|q%`rtc>Q0Eq^Gq&)t1%Qo
zk`d6hNG#W_IAy1l-DRr<9%&aJAK4{#agY;F4Fum)W#8}K$Iq!dj$$qoQLDHB$H;YG
zHEa81@{6_(5QS}DA*~^~Byk1i<b<mA!Tq6Im={?s&-ci78St8J&i6>UBaDw22h{Yo
zlimijc>+HujF#?t9~aiEy;@+n-#1U8D8%nW-5Rc{vpMgUW_y09>VZxR(^n^9XFQFV
z?Fh%Ql*Y%hR|pTr_>@AkLfqfiYOvnf7WCt5=owqQKbKr-VX=B~QmEwy;{yu__wB*B
zAN4T$`U&F>3+eH2CE)<(+Xs}+cITYxID3{>GcsvhZyeg!Rt>BfDGF2{_1q}Dk`HCA
zsM)MBjo`KLq66Rm=ra;UlI4EseoCPHo<mgWm_V{k&`)_hyZK$a3FRjA$<O|5iF%9N
z4~a}s#)Se+E+1_ypT>;!om1Q8zNeq1_uOm=-QcksjElNeVrv+o#VfOqV2F{;?;2w1
z?k8n6Zd;=Fo0EN?DF%6XK{Mc3E~Ftg_p2av5H+b=#r?bnvA#o(Pp)73uFbnQ3@cdZ
zUgCVUqZL)GVOZ(;^e}xpdCv`xlYuse***V<KF;69VB`(?a|L|huQ<_jIw;cLznkaA
zzqcTV9VoWD?$9VrU$MRUo(8R++p(Re^_ekk>c>Xb3s=dD1wU_ACe09^Bw`A0P(N6S
z{;a8|%$<GkG*q@|(6P*0Ee?HoI1r>e=Jj6G=Yw}UnQMukU&G7rbKB5qdIiP~$Ekr7
z$k>h!18jXi$kpPep1x@L_>^J_j~PSLHU5EF`O8_>@_{&q_UAN<yLqljQ7>g$6cU86
zEoPKvuORW#+B_Jea>4M=hS6yrcV3p3lu!GNItPU&l#e*+CS@k7--&JDey97~b&OX@
zPK?p$X;ZE5m9L8UUmsD6>MmXHxJQ<r@Uiyu<yJ8fo?Mafen(!8yQMsrGEW8t8xd(m
z2JOA%?+MUVAAkH@l3tlxcYG>WZ^BkuX7x(euA0i>5_dQ55`HH6$|rJ-rjhAH(Rji-
zn5%YHJ_iTF^mu;cqP<Oz<WgHsS&^N0+payu=aYjR!KV%noQc?l@T9Bbv$MNjm;J0>
z*0$w<QV=QjZFn3C$}6e2^q@P~7u}1BW$)7#OG`bQ3gdLR-#c~lZe@Mthxp*OY02ex
z<kqGw&`td>Hw+#ZM9Q<@R$Wm@xEKoJIV7Ku;d(^#L>u*D(^G_Go^anO20y#T<o2U$
zLYZOSBNXw$or33SrYRztx)JLiaymv-%I`wlNf|oU+^lQoHRx1kUx6&r+4y2c-fp(}
zs)k<g-QrU>k>(ejlM2@z$~!Sz?1w5hA}E;CBKc)a7;nWjh&Fu4bZj^$E$YhgP928y
z46)z+R<EUHF)T20Quq0qa6P>JoQfK^VKxtWUa5zVwlTNDw&B2AU-Q5T2Q_PB%1$!#
zk}k%}mbv;iBtqtcl95`g8VRP2-!1iSv$`D!m6Tl)<zuF6G5J{e%C}v@b&Z+7E1Wm@
zm2<-Ql|I-TRJD|H#R}Blk?4bxXy10sB_5ethE*QakbT`mxQ<s5N&Jj9<C@y1gAL9U
zUdWLs1vIVAdrOZoue43{MVWK<YJy5E*ea}O7T>t&S(lR1Z6!<P?kUe_Eo6oj+0M;&
zRz?=Wp>Zr!OdoNY8Y?(0p39eP?mYv$*cY9&x587}Q#Cr8xv;8UyQ>qyG_{8(ycU*i
z7|s#L922^LM1240g5Wg@Zv{oRCt*?IbWNc8A0dlGFD)ZvO(cgW^_!lKC%8rHF-(%J
ziC;D3_RVIE_?Q|0GeEtHH$6x;I-&a&ji^8!EZm5eIrrs~&!lqUIn#hRBcB1>(vT=w
z@l0Y``+II`HZxMrjfrGV!lsR64T)vhE$LzeN%*JW*!lh>&s70UFK5GS9W@Jqkia;@
z_29S}qyp&K^sE5RvL%;-OG$SA0~<@{L$niVjWWeOTSq1t!_3G7t#OlAy3CqF^*y<U
z%KG}FNeXkWy9>b!U2ZmOc|G;rQjD9FxS)@@azZQGCl)^YF2k$5#{|2Hh~Y@a#G^|%
zbfDK;KX4CXt>IGo17~D_M@Hqqs%)|6u(y(-PB3X#jyxU5%aZ7Cy7aC&^Er+2$i<^v
z!%->o&~%SRfvwr>lH<sBX%Q5D7Rv76MS#SXo-_2x^Hqwg?OQuyq4Pgd_wU4-vTHmu
z<fZcakP2JvJx+fZDPDNZtxxAkCoDPHCK0dldEbjmGM@t;kIG)YUE?hE+NzS6tQV1d
z*`Erg30GvhaorqU1kMrTy&{bY>YRL)mc%zb<YFCAyV705aG|8SwSc7Gb1RXeFU=+L
z;{!e34I7!8pRGQwMA0|}{M=<Ok4U^bWpVBMEqCRI55kGtzY!_!eQ1Sv*x|3+uv?{d
zUUkhn#7$jR?Y6i;L#y|V_WcW^W<D%=mUn6|i@rjW+MAsVMct_<!GVJrEJw4ZlD#^R
zHc;K@We9Dgqkr+ZFh;5DvKgsKjimR%jjdztt=H*xB|ouP)4ea}Yaeoa&sIE1ZR$6$
z2Hpu>)FVVM%%yEQ*{fJ$wI*KaYI?OU;(igXr!~K!{zkzyHcDi8<xVyhPM`aJ*|b&F
z$SeObyKW5rU{!^?SSs}v5k%^HsW0aP@JkGZbEj;(UEOv&v#%p*dk`@kvKYFVqa4zf
zTqt&lZI(*Cmo>6#VnZMN&`NeZlrr5>f=F6krAyDpzp+g(Ub6gpq3`PBqIP=B&Xi09
zJx!E#(N(h;w%{C(XwZp4+v-B?#iE|{=98HC9m1xy8V4!J>cDORM~FxN`3}{~2ij=2
zRSOa_tCw7AmrBLi<s-!R7uY}2dG;#$Jmhl9&v4IZ;AhLH4bSJrrjrFVuF@f`{BLr<
zp;%O?@%dPA@2gkSl$ei|kDr*82o^c3+o9hR4*bM=pl7Wf((l^WL{leWkofbt`S>MT
zk2P24L>XK)#lA>q76!bi`Ed@_`=*}|HAmhUd@b#+4cYb1?9@TimP@rPv1&B+0i(N<
zyxXFc!S>VVlO*1*JLe*as}8H&MBl8Tb4ghG-$z<8x)IY7tV?qXyraqNusW~*d@Qw_
zZV;~pBe?J+<OsK4Xp^WhrTcZ%oP9>pQAJnvhVIkR(EIzs5r&WOKg*<p*WORd+%DT*
zm3yGV@5DzBd5Y)xeS26@dOEj0WcEgfu>ALaGm&N)$D+z(g5&t6dtnl}vBz~Oa%tuH
zCU>mGx(O~C5QaL$HwRq~FD6T?22qP?&S56QUYl60(^3*EDwD0fs?`8>AHIaju5w1%
z!0{iK70C69Bd3_yUQA065yegJPW#d940IXs<SK9tDQ6tF#6b9Fhsr;73FY$BcT$4x
zMixAh88w$cn!{a1W=(vN;W^86?J_^_I5=EVZT?y@qWQHv*KVdB88+HME+;+umhC9Z
z-rH7Y=BwdRPLu-0G>o`FSZ-5|tYL{-Ml6s-Q?lBGugfKKWv0=C5n(3!ymhce-FI*)
zUOY6i>rqv`+E2f<oRRL)I-^L<i_bpPD#-|)XJvG$s<RYf>+Isk1|+6vkG3qGbNbAU
z*{AU4aaT=oPZ}vUN_8XJF3s~CAS9Z2NUxl%T0R!qRvN}e^sQs#-TfV1X8BT+9uwkd
zjHW>m(=4a;8+R9}3g@gQ5aHF0OV-DP?BKej)pA`j*Y%spTybWEAXJ?q%2T~lO_zH&
z7Vbb?^>Y#Uqw)Ho>-Il{<mgAwcW@a!U?{ud>u~+}(PN1(bUI=*<C@I{%wuUp=0}t%
zZ$5t$Vqz!VSKz7Uh~*24xVSm~`09yeV&!4{jL14zB^yLX$)jOlO6~Q+Aj0}bYkhim
ztQ)2#ipH06<4EquM_Ve{uO*vi_Y=Av_~v`o3NCPkGt}siJicdgnT;FM+9iGi<nEBP
zytdo77EXw7$;YHCVMaV58OF@caeE{z&N`^<`Sn;uZM&-*91DJY6Vg@jZE?CPyEZ{;
zh=Pw7@6yk3bFO)GaTdN$G#|~=Q9Cft?{oC~tY@R!$Un`_e{XY}uG$Nc(hx<?z)P4p
zVjaDaR&f#Uc2U&}dm-yQ<i&zFUtAZ?u}4_dE>wOVmPy$Zn8|rAP@kOerBn)~t)B65
z47_xl@*sh@h_y+G3vbxb;E1nVuaW9PO0h|_`h~52;}Rloa}&5CbqjO$&O?WIBGYZ3
z2w(DHeZ6HjtjZi$vm~XTj@(De*KRJ!Dmq?zed{Ty_l8o1!gyZ3jmwYDMPAPiezp08
zT*Fn`OSVCpxbGnKQhFJ_#J$+-c8tez#NTEGsg}c04Z1$TuxxVv4Yi>cDwn+%_K@q1
zdJ3^&#Z6fwt=&3>&4n|rWjhJ>`U0juxJTYS5_rOuc|%5Kv&r)H6HNtswxY7n%VT%k
z4Ud7Izt-4Cbb_&vK*ED3VKaq37cB%DBj*S3o%W+BhlXw^nXA$7UJtdlNJ#A(noGR>
zuG5jsprc=mcbbly&MYf#!3lg~zQx)#6!OFSLQ&i6aod%1{I@PD<Ctkng^8RmC|Yc$
zuun%H9$`tuFK>bhjXlamT#MF)wq|qnFS-?1$=JDGXj5j+Nr6A!@+621?%cVpUO?Xw
zvByG`JfTg{IDLR0aFB2qN{MfTeA_)#!KTGQaFM?{XD80Zsn`~u^gN72JH1RwK<^2Q
zD^pqj4!)m#-r`%{3DRf!73cTd<%>J^9ZL*|q+{sM2Uf-JtSu9B>Q7flAFG7__^c3u
z@-t~O=q>775L?%X{~DwaX|DZJM#$$XEH*2!Ag@JZ%c?@R1m$hr-|D+zEYs69w@X&k
zl-7R79%>Y|>nm^Mu4DHUjf5nLNh(fshYa4irY_gS{=ub<JlnpAz&|IBJw?Hw<bG;8
zdJ8o4a5^>l<$JS7Rr#Zwi-sVDY+sXDOw+Y<vHmCgX;~h-N^JveAm*8{Bb-arbfMLI
zr<V%EQd5mun~URkSJ|VqNQFvpTzagLUMyv<8}2mM&GLo`#4F+%Z3pWcn|Bx93o5?3
zJ`i#r*X+ieT(7=zd>}4wgeX5}k;)j5?cMWjbwq|lepjM_o#Rcwm=*<}6Ec&n-@<C$
zz~-r=Hw|R|E3#1X<(J0~%(N~)I!IKRkDPFSUg0D0LV=!1&gg9;=4t;<)G|y;^<%aC
zSOcY5P>Z2APJ8p*LV2b`b752leYBC%`-jFMYayHyCliE~%w7KNg6WS)5ht<ZkeT+8
zCqa262JISb&mQM(v%8t|S1+ohvhP)^9(#&zme$>pj+Ye1rv>**9@S*&5I9&FO@@p6
zPUyZTj1>;drN8ukcm6`%l#azq2KRB3i?6k6CI<6{U8R|7@2U}PeoLX+uj>>x!A{Ok
z3-X&T-YD#u;xZ1~Dg5&E7LEmRH~!_2Dm@eKL0jcX0UnFXhk~pc+#>_Za@4IBbKYxI
zZ&~UTJ0I}Vd+aqGuO^QZ2C4Mo+!2{W^&S$aFOJK!78E-i2BW6N9x$$(BZEUa)8)IF
z&3hVm^=m799E!rW`o!=)Ui3sdYxbxIdKi;}8fm$cXFpx*U#AbLTdZ!)5!E0k3;#xi
zF7>%_w^SVexgTpxp*H2l^OC{T=a30cLXfYPp=17Q*OthwR%Y_gcx6ReRnl5qpS!ng
z$;N4e_kyeWY25O32II=&Ps~1-;H6R7wA0VftB}4B?8xw8I`0*|az6INts}nU%9F83
z(H|JTsCeGRScqQ#hf0+lrD|`zXAO^4JzNypQBel3&QBMz*M1(BOsK5hdOH(kIKvsk
z6cXQBRkvg>T7<;TrMxrTSiSdvLY($CALY<(FnyNgn4$FKgjmEC4LoRuB*nmG?6Bj_
zD3$p}-D#h1JQDtgO;83{p$ud0P0M!9)?SBo_hn5T^0Igl8-};<_VmTYdAZ4sbG}|6
z!Jj3~P75knh?%7<%v}DihVHao*S1J0*R*zC<lM?LxCuGRfkx)+q^cg>pI>jZIw^5~
zGc34)a+|;9qdBi^D)*)(UxM+4S@AXNR~H<6UlzX{!wz0P(3*c4jR+~wc^sJN?*DnB
ze(e4h$-&Moj&f#(9$fyB@DKcDZ4sTZCHbEOYP4J|Quh2hChE7(KR!|n+|?oY^wR;M
zM<<#tbXD_<!l&Vf=Ga6kj%JiPZ9?xOnWEG?=X&Wl4WCl~00$1m`pio)I?qE5lClO)
zG=GX$IXE9r($LmhWxN`eE!Mmz+28~-(c>3UsKHt9EEVUJ#qaK@P~XGOHtSbk4uM%I
zP)(}iWf@SsSG%x%$hUO6!F(DWi4ak^e1!a@ZRVy%RBAuV&BO3&b#BfngH(_E-p&JR
zlS}(w9-^^xJM@tO%sir~)ptUrl>|eeWnxyp<XepAQXj?HjW?*t=^n49VTv;eM8>-v
zW}j8hTZFYuDml1cD7Z~46?mvIaFE%WsQYx+%gp{<^O%<9&%2SAWg<M=9eNtWewGbF
zP5G{@U&|DVn;ZobZhIXMCoJ^dIE3Xe2`{auA6Lb&*QV0Es6J9lhp@#QyfwbewK8F=
zki?{fh`RnGf8j=%E5yn`tu^dP%_E{ZR)TXNaZZHEe2`w}Fu@yjgP+-RX+rm*uBe6?
zu<>w=iYY5Jha=2ni*X?)^OD$*Zh?IVY4VjL#^m_~E6yI5ULqkXart5x&(FIQ#wIo_
zrch>|?^OifwyE{o`i6XQrbQJc@SEkcd#<7{+=#-N+$#G~qC8XLn|cg5I<EHDeFVbA
zYQWp+z$9MGTVE{HtL_;|YyK8*@ojUk&YySZt?d&LI@Fud3MfmUan`fe_maz{lA9t_
z$rk1-x2yxU2<IkTND^&wQq87mri1OSdf&4TCtcgQ{8~MLMQ$=QJ`3@vsOJ66jksp2
z0ZJ|QNdg)^uGZX{8H5uwVh4B^W-3NLS%Wp}eSWFdcej*H4R<?gQnC8+>lTIPMfB=R
z#uDdsosoj+fzdU3bT?RKx8g^1smha8v213<aPvadgzMtWBn}d)_felHOE+FQwT^9H
z*h-56evt%v*oRikIqH=oPvZ8bot8bag;fRRcwhpqFWAuBYOiQsB${~sg&?lCX@JdD
znO3>Gcd90^cG%DPTa<)n3FC_!@Sw@#FH0u!VS9WZGOpkGDzt6>%|IE<D(XjlFNiZs
z`>BYH*!Rn!=Vyy2QXa@Xe?F&3Y=jIQxe?vwb^8FZ5T<xZ7SbmpZmX3Mw#&KU^EsZ`
z*gYeH!dG=uM&Z~M-DuFykTzNURqEc!aC@a?q3RgVKIPR{MbQWQs#KP#?F}gqAA*+I
zZP2(|#G;nQ&1D1ifDZxeqjiWeQ_@dD>F+qlFUCLw_^CfAJ`NIiuF-cn{e1D2xRMWb
z96>{QxpeJ2QQqQq3`TKTrfi>BSN9ryqCe3K7FAF22HcS;BgMDM8Z`<l?=*O#C3|8?
zg<CyDQ5tSvus&8Iw>c?4DmyOfnZ43{M}Dvpb)KPa(46E48^-vm$)Z1-Th>#ETDVgj
zaTN*6i#x8ht%LOI<c&7xV$w+n`J|%~n{glJ{J+6C+v93j#gz17^3=5zofL9zb`PYX
zc6=H(B#Ap8+1w_^Qx>|^lU%6qW1I&5V=!df)k~|dtRRLXsyXFRR$#TC8`R;(5yPA8
z?c4EFVq?5J+tr1xG(kg2b<Hk`Qk|`@Te)Z+F;w<-)5k-9goek=S{TV_B#s9<oQvy`
zBCnirtEH0>u4a=Eu$b6W=&hoZgf(D0O_Wx6u9JXu2wzLSwZN-j#5$H5kY}Gy$i13D
zGCeG7P~kM>q{aMA*j`kl@CH$&>S(A%-Z|?4REGX*oBV!#a^aI7gl&FCJt5ZalHc{H
zh=*5-w>h|-`=XM*BU?L}Q*#%X=qnG4H1)EM>k`zjD7<F1O=1WTZO40~aowl;D??$v
z!|<mkq&s^EB=_{4H0s+<vSHfJ6bwPq5)FJW*Q+4ovc0pq>;gKvt)vEX=nmV?1gkhq
z7GeXnAl6k&o{r^{>E#*DqUPl(E4|dh1g3uW@?A7Ni@WiInNUH|=Gchg^VhPw-mVOV
zWFvzX%vGfv%E@Sw0@y!Ux^k_uTqmYR?Q?tGzcF2P@5AKVn^w(FsJ?XLJrk{t9_H|X
zH#cI&KZ^IdNevpNV3z%}Jm^v0rr8yuAxr6o+QLQdg>2%pBvA*Z9gEj_3#^cQn;mhY
z;1Y|2r}SQ856YjH-@e^3*))T=$@DXx#Y2AlrapE=(WWA5zuqyakm7D5D36H7tlhHU
z`_?Ci`4_dv1A!QgKt78T5qhGb+wG=SY#finiQdk9AtP#$eni3Kz|2iUZ?hl<vfXw_
zS^xN>fc%!1h>VGmW`4l?_qqb7Z)={Rw4-|00mapl+H#>oo!@`Fyxb|LOU`qf64FF-
zd2wN2k6@5cS0q_?mSW1`@Z8Fe=qt9>oozm2Dy;hs<hSHE5{Pk3yW9<G-|cDH-@OD+
z;(gcuk%h-?{2s;6od6;1(>b1KERKtO^T^i22(Kx4Kj=+Ya%i>V+){wOxx6i-kAVL4
z0CAE9d#V=e#PoNfAz5XLt}PLA2J$F1-_h=-XB!WhuiWrZ3#C;92h33(;LTAl1-H(>
z8F{6$b^IdPIyXS_lOxiDdw!vAD86OP<H4IY{;y=(juG~DRoC?ueA34@GK5i-nV6lH
zc_YeDp7oPBMZBO@@|Y{hwp&-kqY{`J?=!Uf_dUZLrwvoH_`Z1lQW}cW&)Owbc~N?7
zjSd{2dSlbfIB$<_;^*35zMi@CrYm9sD+t#H`$ToK6IF)X)Cn}1ymhj(Iv$vQICeYl
zHYcs3C$rOGi6C{zFd^pqz+i=SsR^ay_^WOQr~KD4$Yin+xjotVX2^SqhozQ7LJwDK
z-|)D~G|uoZ6Bi`DA+NQgedWg!Vlg{)m;CO5BFoo;*r(=_{>?!ji*XmNZta?NGR-I_
zYB9uILRZ{yi*xH!AlFdl%+SmA*06hs{!G6)GjH46F&pvWJR{y_0-pfibz9Q>*Z=`H
zNsVZcV?CMIyVC^X5%6&@k=T{}530uz*s^VJSb+BZdp}Zr$Oi(9Tlxac&#OILvt(z~
z+9Wy5^B<Jq%*r{b)yf<n<Ptlk)7rw|d%+v8H~PiVnZ2_7Sxfs&RNTj=+n{ajel0^x
zzNa@x{CilBju7m%7%3^m_m_%;X79YYJk}oxTLsJR6shBMJ{hR~_z_`J@>!qUKJcO!
z0x^*=r_yYemO<s{rFqiK!Z=^!e>E~i^24jgHcm%bunuKzMVVpODY<m}mmZNr8{Azl
zmhI_iPS!LgDr#JKqA?{DE-O6IF_I?*K}shMY)2m~@s<4`t4F=qZi67YL96ed^?M7@
zx5_Du8=o1U<j3MZ+014cLG2>Cc*r8kQx<qdZjr%|_82T026Z_-p=_PWAVMI1(CI-+
z3IsOrbcKPyDkvlt1lGsdVt>6nB|-oy9F4&$IziC@KUfobb`}s827!&?_E;wj2qFmF
z(jx;aBkfT3aHJ#XkIH{YWMCz@g98i=L)yVGAoEjId`J}59_9c7+oP~hJ3H8^Z3j5;
zi2y#)P&+Ifg#>|}a7afe8s~z5;(+^_7z7mK1Og*|UCF>GM-&ov7X(J1TK%KZ77cR;
z;&Fi^ahSizW_D2YsZOu6>#qyhseJ?tg8@vRotyy&0KXi5d2{Lx;3>=v2L(=WqyroY
z$9e)T03yGLWMD@CS1b_QZx&Z58ViRa?16YeU<}U11^PP@S2W5VX9on~3RrSPLr*<)
z!vUHxfb$3#85rXVwF77{4?6_Z<?Nol?WuX3i!BU|fja`e+oKQ&AQfO&82UG4P#55c
zfg<h6errR+9i2{%Kmn`~Fo%DRXJjCKFf<$kK<0`##l{W=_y?fu>FNXnv;(T(C_vL0
z%w8}w3Iqlq27$5nfD_gU4Fk>&C>-E{1N<}s4BP|w2Y?%phW$2;gr5r7p%5q}pb6#z
z|8@M07{K%!r8CS<!-Qde0|W(p1%5SB0R&bAft5gDWuW#zU{w%U4Fpyv1K$RLHGm2L
zfwe$jZ4g)o1ik|T>w>^~K*<6X3<4X1z(zn#g1{yquqg;^1_IjxWpoz?WcL&XTi|?F
zHfQvIF|mWg?ciuT9Dq6wX%E0@he87({L0R6VE`jrASM8RAOawEAcFvEWMCkbK(zpH
z0%-!I0T2L~Tmey_CIMUk$N-ffusaBR4+I8~0t(0Lw_mV-yh4-x@)JP*SK^=;z#A+a
zVGo%3w?&k3r-^_gu`r-;08@YJ#ukb>^#_3hTHve#fSUPDg&`1dR}38R!XE1A2m>7Y
zef?EbzXNtazyPm+v}2u6xYH^Ez&vXUM;sjSt2BQ*s-p~G?gDcJkU&H2;dZBPKwW{N
z_BhMvzs~;m@W-v+ntnMC_;41mEe?Tz0Wbk`=~r*VTz+5AFu(%VU@lPfU7(iE`u1PX
z&NQF^1e9P1EZ`af<po1J0<TT&;Xp_J3jSYzes`8V>K=g9@8&t}2E?z~VTVJX4ksW#
zo~Na3i$>jrA#G1P4FfwHMSnEfp<F$G*#;&M+TH;u@!vIn>KFp%aym~=D-MBjJjDr^
z9)CIv4ReH_T7>~fyFl%J%_Dmtt1w^yoVMiI%mUgCgZtHufCs<Y?$1l89S#db=mOLg
z(B)8|B3yw={N1{zVf|Ka2eSv90U!a|_m3>#t_x6MI3Q?1#lQak6Abb+muNu4FEJ-5
z0x0xf*3Za5DgD#0r!WHjdN#=9e>p`4mjAn~<j+zl{|7nw-;-Ki8EEo<JybqZqI{-A
z`EMo4e}vV~_|(r3Q2(1x{STiq7QjRM%(%{(kj|Nq&fh{h|1kX#cEKUAa90FS2s&re
z!thLk;h6@*zcm>Cku*KKZ{UPN1JD2s=K?g8Edl`TG)eA2hM{LH(7#mx&xc>#4s_K&
zB@6ux6BNi1@MJto<Z17pDuJCTf&HxnVERWGe&#RyjPGo={cd&mKYV~5Ebt^no*74-
z38Bt}P=6~&{loN+Ffd8r?r?j+8PwUM0(T~lJCn!#Esy&n=XrJ?m^Y^&dH#FHl1xU1
z46Fc*v)^loHr&M)htY>3HE-#`9D!OtB`d)&c4)XO7KH{ugin_a5Fm86(3~zI|Lij=
z5YVsv=I{G|?>q94zjqW5tg1d1c!C2pDSx_|*g-*%Q^jC;f&ZV4Pvmb#!T*;L*r5LC
z6Z(JI`1yo?F`iy!Wc~{4G|aQG{=_Qse-TChFQUL7;@OsW2kHo{U|<cn{plhGfdCA8
zAWI;26b@Lyeg*!|iujjCu;S^G1K`f31;x4|fDs3`1-T3ILHPOjZ-Sh#SXYb$7?^1P
z+~Y%`9l6QCcYZBBz&!a|%+=ljbow*~%>1JQVEttP*!h%BGr^@KA;2#n%nyMGLj(jN
zLZZC<0yp^iZ*c#Gj0YO#K*kS(h>`LC`vDRW78Vo+Ie`A9K>(~yKS0ReG>DL}kQh)a
z|Dg%+13>?mCI|rVADWN=|No{50U-XDPXs{zZ`$cu<Uck5lHh-B@QaIziT=kwfKPy5
z>_7SdKH>k^J(U&zk1Rw;KwRj5$%=`I|0fQ9adC*ifA|2bgoXan2Rw`62pIa;mkj!F
wFBpJ5uz#U|=iBcp0;?k(P$0-@+5GA}3>J8p{7Ms`Nt9oJjDtg2M}_Qv05%AH8~^|S
literal 0
HcmV?d00001
diff --git a/5_dpnn/results/serial/slurm-25141451.out b/5_dpnn/results/serial/slurm-25141451.out
new file mode 100644
index 0000000..6f81a6a
--- /dev/null
+++ b/5_dpnn/results/serial/slurm-25141451.out
@@ -0,0 +1,427 @@
+Using cuda:0 device.
+Files already downloaded and verified
+Image batch dimensions: torch.Size([256, 3, 64, 64])
+Image label dimensions: torch.Size([256])
+Class labels of 10 examples: tensor([2, 7, 8, 2, 2, 2, 9, 7, 6, 3])
+Epoch: 001/100 | Batch 0000/0175 | Loss: 2.3027
+Epoch: 001/100 | Batch 0100/0175 | Loss: 2.1135
+Epoch: 001/100 | Train: 25.35% | Validation: 26.28%
+Time elapsed: 0.73 min
+Epoch: 002/100 | Batch 0000/0175 | Loss: 1.8631
+Epoch: 002/100 | Batch 0100/0175 | Loss: 1.7782
+Epoch: 002/100 | Train: 39.24% | Validation: 39.66%
+Time elapsed: 1.24 min
+Epoch: 003/100 | Batch 0000/0175 | Loss: 1.6075
+Epoch: 003/100 | Batch 0100/0175 | Loss: 1.5874
+Epoch: 003/100 | Train: 53.95% | Validation: 54.74%
+Time elapsed: 1.76 min
+Epoch: 004/100 | Batch 0000/0175 | Loss: 1.1914
+Epoch: 004/100 | Batch 0100/0175 | Loss: 1.1916
+Epoch: 004/100 | Train: 57.55% | Validation: 57.56%
+Time elapsed: 2.27 min
+Epoch: 005/100 | Batch 0000/0175 | Loss: 1.1251
+Epoch: 005/100 | Batch 0100/0175 | Loss: 1.0534
+Epoch: 005/100 | Train: 61.19% | Validation: 61.28%
+Time elapsed: 2.78 min
+Epoch: 006/100 | Batch 0000/0175 | Loss: 1.1530
+Epoch: 006/100 | Batch 0100/0175 | Loss: 1.1383
+Epoch: 006/100 | Train: 63.20% | Validation: 61.98%
+Time elapsed: 3.30 min
+Epoch: 007/100 | Batch 0000/0175 | Loss: 0.9854
+Epoch: 007/100 | Batch 0100/0175 | Loss: 0.9646
+Epoch: 007/100 | Train: 68.37% | Validation: 65.88%
+Time elapsed: 3.81 min
+Epoch: 008/100 | Batch 0000/0175 | Loss: 0.9393
+Epoch: 008/100 | Batch 0100/0175 | Loss: 0.9734
+Epoch: 008/100 | Train: 70.14% | Validation: 67.34%
+Time elapsed: 4.33 min
+Epoch: 009/100 | Batch 0000/0175 | Loss: 0.8918
+Epoch: 009/100 | Batch 0100/0175 | Loss: 0.8947
+Epoch: 009/100 | Train: 71.09% | Validation: 68.08%
+Time elapsed: 4.84 min
+Epoch: 010/100 | Batch 0000/0175 | Loss: 0.8273
+Epoch: 010/100 | Batch 0100/0175 | Loss: 0.8720
+Epoch: 010/100 | Train: 74.83% | Validation: 70.58%
+Time elapsed: 5.36 min
+Epoch: 011/100 | Batch 0000/0175 | Loss: 0.7854
+Epoch: 011/100 | Batch 0100/0175 | Loss: 0.9307
+Epoch: 011/100 | Train: 75.41% | Validation: 71.16%
+Time elapsed: 5.87 min
+Epoch: 012/100 | Batch 0000/0175 | Loss: 0.6924
+Epoch: 012/100 | Batch 0100/0175 | Loss: 0.8397
+Epoch: 012/100 | Train: 75.55% | Validation: 70.94%
+Time elapsed: 6.39 min
+Epoch: 013/100 | Batch 0000/0175 | Loss: 0.6357
+Epoch: 013/100 | Batch 0100/0175 | Loss: 0.7722
+Epoch: 013/100 | Train: 75.43% | Validation: 70.84%
+Time elapsed: 6.90 min
+Epoch: 014/100 | Batch 0000/0175 | Loss: 0.6798
+Epoch: 014/100 | Batch 0100/0175 | Loss: 0.7900
+Epoch: 014/100 | Train: 77.48% | Validation: 71.84%
+Time elapsed: 7.41 min
+Epoch: 015/100 | Batch 0000/0175 | Loss: 0.7129
+Epoch: 015/100 | Batch 0100/0175 | Loss: 0.6552
+Epoch: 015/100 | Train: 78.69% | Validation: 72.54%
+Time elapsed: 7.92 min
+Epoch: 016/100 | Batch 0000/0175 | Loss: 0.7056
+Epoch: 016/100 | Batch 0100/0175 | Loss: 0.8437
+Epoch: 016/100 | Train: 79.88% | Validation: 73.74%
+Time elapsed: 8.42 min
+Epoch: 017/100 | Batch 0000/0175 | Loss: 0.6594
+Epoch: 017/100 | Batch 0100/0175 | Loss: 0.6912
+Epoch: 017/100 | Train: 75.76% | Validation: 69.72%
+Time elapsed: 8.93 min
+Epoch: 018/100 | Batch 0000/0175 | Loss: 0.8299
+Epoch: 018/100 | Batch 0100/0175 | Loss: 0.8016
+Epoch: 018/100 | Train: 79.15% | Validation: 71.62%
+Time elapsed: 9.44 min
+Epoch: 019/100 | Batch 0000/0175 | Loss: 0.7093
+Epoch: 019/100 | Batch 0100/0175 | Loss: 0.7684
+Epoch: 019/100 | Train: 78.37% | Validation: 71.36%
+Time elapsed: 9.95 min
+Epoch: 020/100 | Batch 0000/0175 | Loss: 0.5805
+Epoch: 020/100 | Batch 0100/0175 | Loss: 0.7225
+Epoch: 020/100 | Train: 78.05% | Validation: 71.20%
+Time elapsed: 10.45 min
+Epoch: 021/100 | Batch 0000/0175 | Loss: 0.7297
+Epoch: 021/100 | Batch 0100/0175 | Loss: 0.6416
+Epoch: 021/100 | Train: 78.48% | Validation: 71.50%
+Time elapsed: 10.96 min
+Epoch: 022/100 | Batch 0000/0175 | Loss: 0.7761
+Epoch: 022/100 | Batch 0100/0175 | Loss: 0.6268
+Epoch: 022/100 | Train: 76.74% | Validation: 69.28%
+Time elapsed: 11.47 min
+Epoch: 023/100 | Batch 0000/0175 | Loss: 0.6827
+Epoch: 023/100 | Batch 0100/0175 | Loss: 0.8298
+Epoch: 023/100 | Train: 77.54% | Validation: 70.56%
+Time elapsed: 11.98 min
+Epoch: 024/100 | Batch 0000/0175 | Loss: 0.7327
+Epoch: 024/100 | Batch 0100/0175 | Loss: 0.6323
+Epoch: 024/100 | Train: 75.57% | Validation: 69.38%
+Time elapsed: 12.48 min
+Epoch: 025/100 | Batch 0000/0175 | Loss: 0.7100
+Epoch: 025/100 | Batch 0100/0175 | Loss: 0.7743
+Epoch: 025/100 | Train: 74.69% | Validation: 68.58%
+Time elapsed: 12.99 min
+Epoch: 026/100 | Batch 0000/0175 | Loss: 0.8528
+Epoch: 026/100 | Batch 0100/0175 | Loss: 0.8293
+Epoch: 026/100 | Train: 72.27% | Validation: 66.64%
+Time elapsed: 13.49 min
+Epoch: 027/100 | Batch 0000/0175 | Loss: 0.7368
+Epoch: 027/100 | Batch 0100/0175 | Loss: 0.8291
+Epoch: 027/100 | Train: 73.77% | Validation: 68.98%
+Time elapsed: 14.00 min
+Epoch: 027/100: LR updated to 0.010000000000000002
+Epoch: 028/100 | Batch 0000/0175 | Loss: 0.8961
+Epoch: 028/100 | Batch 0100/0175 | Loss: 0.6875
+Epoch: 028/100 | Train: 83.81% | Validation: 75.06%
+Time elapsed: 14.50 min
+Epoch: 029/100 | Batch 0000/0175 | Loss: 0.4902
+Epoch: 029/100 | Batch 0100/0175 | Loss: 0.4860
+Epoch: 029/100 | Train: 85.34% | Validation: 76.04%
+Time elapsed: 15.01 min
+Epoch: 030/100 | Batch 0000/0175 | Loss: 0.4664
+Epoch: 030/100 | Batch 0100/0175 | Loss: 0.3775
+Epoch: 030/100 | Train: 86.56% | Validation: 76.78%
+Time elapsed: 15.51 min
+Epoch: 031/100 | Batch 0000/0175 | Loss: 0.4098
+Epoch: 031/100 | Batch 0100/0175 | Loss: 0.3389
+Epoch: 031/100 | Train: 87.53% | Validation: 77.34%
+Time elapsed: 16.02 min
+Epoch: 032/100 | Batch 0000/0175 | Loss: 0.3756
+Epoch: 032/100 | Batch 0100/0175 | Loss: 0.3652
+Epoch: 032/100 | Train: 88.45% | Validation: 77.22%
+Time elapsed: 16.52 min
+Epoch: 033/100 | Batch 0000/0175 | Loss: 0.2903
+Epoch: 033/100 | Batch 0100/0175 | Loss: 0.3860
+Epoch: 033/100 | Train: 89.04% | Validation: 77.12%
+Time elapsed: 17.02 min
+Epoch: 034/100 | Batch 0000/0175 | Loss: 0.3783
+Epoch: 034/100 | Batch 0100/0175 | Loss: 0.2564
+Epoch: 034/100 | Train: 89.98% | Validation: 77.28%
+Time elapsed: 17.53 min
+Epoch: 035/100 | Batch 0000/0175 | Loss: 0.2391
+Epoch: 035/100 | Batch 0100/0175 | Loss: 0.2611
+Epoch: 035/100 | Train: 90.38% | Validation: 77.44%
+Time elapsed: 18.03 min
+Epoch: 036/100 | Batch 0000/0175 | Loss: 0.3834
+Epoch: 036/100 | Batch 0100/0175 | Loss: 0.3105
+Epoch: 036/100 | Train: 90.92% | Validation: 77.52%
+Time elapsed: 18.54 min
+Epoch: 037/100 | Batch 0000/0175 | Loss: 0.3067
+Epoch: 037/100 | Batch 0100/0175 | Loss: 0.2918
+Epoch: 037/100 | Train: 91.25% | Validation: 77.52%
+Time elapsed: 19.16 min
+Epoch: 038/100 | Batch 0000/0175 | Loss: 0.3285
+Epoch: 038/100 | Batch 0100/0175 | Loss: 0.2089
+Epoch: 038/100 | Train: 91.87% | Validation: 77.42%
+Time elapsed: 19.67 min
+Epoch: 039/100 | Batch 0000/0175 | Loss: 0.2146
+Epoch: 039/100 | Batch 0100/0175 | Loss: 0.2337
+Epoch: 039/100 | Train: 92.36% | Validation: 77.44%
+Time elapsed: 20.17 min
+Epoch: 040/100 | Batch 0000/0175 | Loss: 0.2115
+Epoch: 040/100 | Batch 0100/0175 | Loss: 0.2268
+Epoch: 040/100 | Train: 92.83% | Validation: 77.54%
+Time elapsed: 20.67 min
+Epoch: 041/100 | Batch 0000/0175 | Loss: 0.2595
+Epoch: 041/100 | Batch 0100/0175 | Loss: 0.1908
+Epoch: 041/100 | Train: 93.67% | Validation: 77.50%
+Time elapsed: 21.17 min
+Epoch: 042/100 | Batch 0000/0175 | Loss: 0.2448
+Epoch: 042/100 | Batch 0100/0175 | Loss: 0.2056
+Epoch: 042/100 | Train: 93.77% | Validation: 77.18%
+Time elapsed: 21.67 min
+Epoch: 043/100 | Batch 0000/0175 | Loss: 0.2638
+Epoch: 043/100 | Batch 0100/0175 | Loss: 0.2774
+Epoch: 043/100 | Train: 94.04% | Validation: 77.20%
+Time elapsed: 22.17 min
+Epoch: 044/100 | Batch 0000/0175 | Loss: 0.2036
+Epoch: 044/100 | Batch 0100/0175 | Loss: 0.2589
+Epoch: 044/100 | Train: 94.51% | Validation: 77.48%
+Time elapsed: 22.67 min
+Epoch: 045/100 | Batch 0000/0175 | Loss: 0.1772
+Epoch: 045/100 | Batch 0100/0175 | Loss: 0.2554
+Epoch: 045/100 | Train: 94.90% | Validation: 77.06%
+Time elapsed: 23.17 min
+Epoch: 046/100 | Batch 0000/0175 | Loss: 0.1964
+Epoch: 046/100 | Batch 0100/0175 | Loss: 0.2367
+Epoch: 046/100 | Train: 95.22% | Validation: 77.60%
+Time elapsed: 23.67 min
+Epoch: 047/100 | Batch 0000/0175 | Loss: 0.2147
+Epoch: 047/100 | Batch 0100/0175 | Loss: 0.1333
+Epoch: 047/100 | Train: 95.68% | Validation: 77.50%
+Time elapsed: 24.17 min
+Epoch: 048/100 | Batch 0000/0175 | Loss: 0.1200
+Epoch: 048/100 | Batch 0100/0175 | Loss: 0.1642
+Epoch: 048/100 | Train: 95.83% | Validation: 77.50%
+Time elapsed: 24.67 min
+Epoch: 049/100 | Batch 0000/0175 | Loss: 0.1104
+Epoch: 049/100 | Batch 0100/0175 | Loss: 0.2606
+Epoch: 049/100 | Train: 96.37% | Validation: 77.46%
+Time elapsed: 25.17 min
+Epoch: 050/100 | Batch 0000/0175 | Loss: 0.1158
+Epoch: 050/100 | Batch 0100/0175 | Loss: 0.0844
+Epoch: 050/100 | Train: 96.30% | Validation: 76.60%
+Time elapsed: 25.67 min
+Epoch: 051/100 | Batch 0000/0175 | Loss: 0.1552
+Epoch: 051/100 | Batch 0100/0175 | Loss: 0.1725
+Epoch: 051/100 | Train: 96.57% | Validation: 77.22%
+Time elapsed: 26.17 min
+Epoch: 052/100 | Batch 0000/0175 | Loss: 0.1318
+Epoch: 052/100 | Batch 0100/0175 | Loss: 0.1254
+Epoch: 052/100 | Train: 96.76% | Validation: 77.30%
+Time elapsed: 26.67 min
+Epoch: 053/100 | Batch 0000/0175 | Loss: 0.1259
+Epoch: 053/100 | Batch 0100/0175 | Loss: 0.0971
+Epoch: 053/100 | Train: 96.90% | Validation: 77.46%
+Time elapsed: 27.17 min
+Epoch: 054/100 | Batch 0000/0175 | Loss: 0.0962
+Epoch: 054/100 | Batch 0100/0175 | Loss: 0.1066
+Epoch: 054/100 | Train: 97.36% | Validation: 77.18%
+Time elapsed: 27.67 min
+Epoch: 055/100 | Batch 0000/0175 | Loss: 0.0893
+Epoch: 055/100 | Batch 0100/0175 | Loss: 0.1006
+Epoch: 055/100 | Train: 97.43% | Validation: 77.24%
+Time elapsed: 28.18 min
+Epoch: 056/100 | Batch 0000/0175 | Loss: 0.0695
+Epoch: 056/100 | Batch 0100/0175 | Loss: 0.1526
+Epoch: 056/100 | Train: 97.57% | Validation: 77.58%
+Time elapsed: 28.68 min
+Epoch: 057/100 | Batch 0000/0175 | Loss: 0.0656
+Epoch: 057/100 | Batch 0100/0175 | Loss: 0.0998
+Epoch: 057/100 | Train: 97.62% | Validation: 77.36%
+Time elapsed: 29.19 min
+Epoch: 057/100: LR updated to 0.0010000000000000002
+Epoch: 058/100 | Batch 0000/0175 | Loss: 0.0609
+Epoch: 058/100 | Batch 0100/0175 | Loss: 0.0728
+Epoch: 058/100 | Train: 98.02% | Validation: 77.60%
+Time elapsed: 29.69 min
+Epoch: 059/100 | Batch 0000/0175 | Loss: 0.0589
+Epoch: 059/100 | Batch 0100/0175 | Loss: 0.0491
+Epoch: 059/100 | Train: 98.15% | Validation: 77.66%
+Time elapsed: 30.19 min
+Epoch: 060/100 | Batch 0000/0175 | Loss: 0.0829
+Epoch: 060/100 | Batch 0100/0175 | Loss: 0.0516
+Epoch: 060/100 | Train: 98.24% | Validation: 77.90%
+Time elapsed: 30.70 min
+Epoch: 061/100 | Batch 0000/0175 | Loss: 0.0554
+Epoch: 061/100 | Batch 0100/0175 | Loss: 0.0694
+Epoch: 061/100 | Train: 98.26% | Validation: 77.88%
+Time elapsed: 31.20 min
+Epoch: 062/100 | Batch 0000/0175 | Loss: 0.0482
+Epoch: 062/100 | Batch 0100/0175 | Loss: 0.0288
+Epoch: 062/100 | Train: 98.34% | Validation: 77.88%
+Time elapsed: 31.70 min
+Epoch: 063/100 | Batch 0000/0175 | Loss: 0.0533
+Epoch: 063/100 | Batch 0100/0175 | Loss: 0.0425
+Epoch: 063/100 | Train: 98.48% | Validation: 77.72%
+Time elapsed: 32.21 min
+Epoch: 064/100 | Batch 0000/0175 | Loss: 0.0853
+Epoch: 064/100 | Batch 0100/0175 | Loss: 0.0958
+Epoch: 064/100 | Train: 98.38% | Validation: 77.68%
+Time elapsed: 32.71 min
+Epoch: 065/100 | Batch 0000/0175 | Loss: 0.0539
+Epoch: 065/100 | Batch 0100/0175 | Loss: 0.0604
+Epoch: 065/100 | Train: 98.42% | Validation: 77.86%
+Time elapsed: 33.22 min
+Epoch: 066/100 | Batch 0000/0175 | Loss: 0.0653
+Epoch: 066/100 | Batch 0100/0175 | Loss: 0.0355
+Epoch: 066/100 | Train: 98.36% | Validation: 77.72%
+Time elapsed: 33.72 min
+Epoch: 067/100 | Batch 0000/0175 | Loss: 0.0781
+Epoch: 067/100 | Batch 0100/0175 | Loss: 0.0654
+Epoch: 067/100 | Train: 98.50% | Validation: 77.74%
+Time elapsed: 34.24 min
+Epoch: 068/100 | Batch 0000/0175 | Loss: 0.0433
+Epoch: 068/100 | Batch 0100/0175 | Loss: 0.0623
+Epoch: 068/100 | Train: 98.55% | Validation: 77.78%
+Time elapsed: 34.74 min
+Epoch: 069/100 | Batch 0000/0175 | Loss: 0.0385
+Epoch: 069/100 | Batch 0100/0175 | Loss: 0.0597
+Epoch: 069/100 | Train: 98.59% | Validation: 77.74%
+Time elapsed: 35.25 min
+Epoch: 070/100 | Batch 0000/0175 | Loss: 0.0430
+Epoch: 070/100 | Batch 0100/0175 | Loss: 0.0233
+Epoch: 070/100 | Train: 98.53% | Validation: 77.86%
+Time elapsed: 35.75 min
+Epoch: 071/100 | Batch 0000/0175 | Loss: 0.0875
+Epoch: 071/100 | Batch 0100/0175 | Loss: 0.0265
+Epoch: 071/100 | Train: 98.52% | Validation: 77.86%
+Time elapsed: 36.26 min
+Epoch: 071/100: LR updated to 0.00010000000000000003
+Epoch: 072/100 | Batch 0000/0175 | Loss: 0.0353
+Epoch: 072/100 | Batch 0100/0175 | Loss: 0.0917
+Epoch: 072/100 | Train: 98.55% | Validation: 77.78%
+Time elapsed: 36.76 min
+Epoch: 073/100 | Batch 0000/0175 | Loss: 0.0822
+Epoch: 073/100 | Batch 0100/0175 | Loss: 0.0586
+Epoch: 073/100 | Train: 98.63% | Validation: 77.72%
+Time elapsed: 37.27 min
+Epoch: 074/100 | Batch 0000/0175 | Loss: 0.0512
+Epoch: 074/100 | Batch 0100/0175 | Loss: 0.0212
+Epoch: 074/100 | Train: 98.52% | Validation: 77.68%
+Time elapsed: 37.78 min
+Epoch: 075/100 | Batch 0000/0175 | Loss: 0.0342
+Epoch: 075/100 | Batch 0100/0175 | Loss: 0.0602
+Epoch: 075/100 | Train: 98.67% | Validation: 77.72%
+Time elapsed: 38.28 min
+Epoch: 076/100 | Batch 0000/0175 | Loss: 0.0761
+Epoch: 076/100 | Batch 0100/0175 | Loss: 0.0260
+Epoch: 076/100 | Train: 98.62% | Validation: 77.74%
+Time elapsed: 38.79 min
+Epoch: 077/100 | Batch 0000/0175 | Loss: 0.0480
+Epoch: 077/100 | Batch 0100/0175 | Loss: 0.0956
+Epoch: 077/100 | Train: 98.73% | Validation: 77.66%
+Time elapsed: 39.29 min
+Epoch: 078/100 | Batch 0000/0175 | Loss: 0.0558
+Epoch: 078/100 | Batch 0100/0175 | Loss: 0.0449
+Epoch: 078/100 | Train: 98.72% | Validation: 77.66%
+Time elapsed: 39.79 min
+Epoch: 079/100 | Batch 0000/0175 | Loss: 0.0107
+Epoch: 079/100 | Batch 0100/0175 | Loss: 0.0464
+Epoch: 079/100 | Train: 98.60% | Validation: 77.76%
+Time elapsed: 40.29 min
+Epoch: 080/100 | Batch 0000/0175 | Loss: 0.1498
+Epoch: 080/100 | Batch 0100/0175 | Loss: 0.0225
+Epoch: 080/100 | Train: 98.65% | Validation: 77.68%
+Time elapsed: 40.79 min
+Epoch: 081/100 | Batch 0000/0175 | Loss: 0.0386
+Epoch: 081/100 | Batch 0100/0175 | Loss: 0.0564
+Epoch: 081/100 | Train: 98.65% | Validation: 77.62%
+Time elapsed: 41.29 min
+Epoch: 082/100 | Batch 0000/0175 | Loss: 0.0665
+Epoch: 082/100 | Batch 0100/0175 | Loss: 0.0376
+Epoch: 082/100 | Train: 98.68% | Validation: 77.66%
+Time elapsed: 41.79 min
+Epoch: 082/100: LR updated to 1.0000000000000004e-05
+Epoch: 083/100 | Batch 0000/0175 | Loss: 0.0269
+Epoch: 083/100 | Batch 0100/0175 | Loss: 0.0186
+Epoch: 083/100 | Train: 98.63% | Validation: 77.66%
+Time elapsed: 42.29 min
+Epoch: 084/100 | Batch 0000/0175 | Loss: 0.0363
+Epoch: 084/100 | Batch 0100/0175 | Loss: 0.0325
+Epoch: 084/100 | Train: 98.68% | Validation: 77.64%
+Time elapsed: 42.80 min
+Epoch: 085/100 | Batch 0000/0175 | Loss: 0.0427
+Epoch: 085/100 | Batch 0100/0175 | Loss: 0.0408
+Epoch: 085/100 | Train: 98.63% | Validation: 77.66%
+Time elapsed: 43.30 min
+Epoch: 086/100 | Batch 0000/0175 | Loss: 0.0723
+Epoch: 086/100 | Batch 0100/0175 | Loss: 0.0267
+Epoch: 086/100 | Train: 98.68% | Validation: 77.64%
+Time elapsed: 43.80 min
+Epoch: 087/100 | Batch 0000/0175 | Loss: 0.0227
+Epoch: 087/100 | Batch 0100/0175 | Loss: 0.0742
+Epoch: 087/100 | Train: 98.76% | Validation: 77.66%
+Time elapsed: 44.30 min
+Epoch: 088/100 | Batch 0000/0175 | Loss: 0.0652
+Epoch: 088/100 | Batch 0100/0175 | Loss: 0.0463
+Epoch: 088/100 | Train: 98.62% | Validation: 77.66%
+Time elapsed: 44.80 min
+Epoch: 089/100 | Batch 0000/0175 | Loss: 0.0579
+Epoch: 089/100 | Batch 0100/0175 | Loss: 0.0620
+Epoch: 089/100 | Train: 98.69% | Validation: 77.64%
+Time elapsed: 45.31 min
+Epoch: 090/100 | Batch 0000/0175 | Loss: 0.0517
+Epoch: 090/100 | Batch 0100/0175 | Loss: 0.0457
+Epoch: 090/100 | Train: 98.64% | Validation: 77.64%
+Time elapsed: 45.81 min
+Epoch: 091/100 | Batch 0000/0175 | Loss: 0.0616
+Epoch: 091/100 | Batch 0100/0175 | Loss: 0.0604
+Epoch: 091/100 | Train: 98.67% | Validation: 77.62%
+Time elapsed: 46.32 min
+Epoch: 092/100 | Batch 0000/0175 | Loss: 0.0424
+Epoch: 092/100 | Batch 0100/0175 | Loss: 0.0316
+Epoch: 092/100 | Train: 98.64% | Validation: 77.62%
+Time elapsed: 46.82 min
+Epoch: 093/100 | Batch 0000/0175 | Loss: 0.0718
+Epoch: 093/100 | Batch 0100/0175 | Loss: 0.0615
+Epoch: 093/100 | Train: 98.70% | Validation: 77.62%
+Time elapsed: 47.33 min
+Epoch: 093/100: LR updated to 1.0000000000000004e-06
+Epoch: 094/100 | Batch 0000/0175 | Loss: 0.0460
+Epoch: 094/100 | Batch 0100/0175 | Loss: 0.0514
+Epoch: 094/100 | Train: 98.61% | Validation: 77.62%
+Time elapsed: 47.83 min
+Epoch: 095/100 | Batch 0000/0175 | Loss: 0.1110
+Epoch: 095/100 | Batch 0100/0175 | Loss: 0.0346
+Epoch: 095/100 | Train: 98.72% | Validation: 77.62%
+Time elapsed: 48.34 min
+Epoch: 096/100 | Batch 0000/0175 | Loss: 0.0603
+Epoch: 096/100 | Batch 0100/0175 | Loss: 0.0322
+Epoch: 096/100 | Train: 98.77% | Validation: 77.62%
+Time elapsed: 48.84 min
+Epoch: 097/100 | Batch 0000/0175 | Loss: 0.0460
+Epoch: 097/100 | Batch 0100/0175 | Loss: 0.0463
+Epoch: 097/100 | Train: 98.63% | Validation: 77.62%
+Time elapsed: 49.34 min
+Epoch: 098/100 | Batch 0000/0175 | Loss: 0.0693
+Epoch: 098/100 | Batch 0100/0175 | Loss: 0.0466
+Epoch: 098/100 | Train: 98.65% | Validation: 77.62%
+Time elapsed: 49.84 min
+Epoch: 099/100 | Batch 0000/0175 | Loss: 0.0447
+Epoch: 099/100 | Batch 0100/0175 | Loss: 0.0587
+Epoch: 099/100 | Train: 98.72% | Validation: 77.62%
+Time elapsed: 50.34 min
+Epoch: 100/100 | Batch 0000/0175 | Loss: 0.0245
+Epoch: 100/100 | Batch 0100/0175 | Loss: 0.0286
+Epoch: 100/100 | Train: 98.69% | Validation: 77.62%
+Time elapsed: 50.84 min
+Total Training Time: 50.84 min
+Test accuracy 76.97%
+
+============================= JOB FEEDBACK =============================
+
+NodeName=uc2n513
+Job ID: 25141451
+Cluster: uc2
+User/Group: xk5289/scc
+State: COMPLETED (exit code 0)
+Nodes: 1
+Cores per node: 10
+CPU Utilized: 00:50:32
+CPU Efficiency: 9.49% of 08:52:40 core-walltime
+Job Wall-clock time: 00:53:16
+Memory Utilized: 1.43 GB
+Memory Efficiency: 1.14% of 125.00 GB
diff --git a/5_dpnn/results/serial/solution_serial.ipynb b/5_dpnn/results/serial/solution_serial.ipynb
new file mode 100644
index 0000000..1c5e88e
--- /dev/null
+++ b/5_dpnn/results/serial/solution_serial.ipynb
@@ -0,0 +1,1522 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": "# Serial Alex Net Solution"
+ },
+ {
+ "metadata": {
+ "tags": [],
+ "ExecuteTime": {
+ "end_time": "2025-01-14T13:22:41.632319Z",
+ "start_time": "2025-01-14T13:22:40.435888Z"
+ }
+ },
+ "cell_type": "code",
+ "source": [
+ "# IMPORTS\n",
+ "import torch\n",
+ "from torch.optim.lr_scheduler import ReduceLROnPlateau\n",
+ "import torchvision\n",
+ "import time\n",
+ "import numpy as np\n",
+ "from typing import Callable, Tuple, Any, Union, List"
+ ],
+ "outputs": [],
+ "execution_count": 1
+ },
+ {
+ "cell_type": "code",
+ "metadata": {
+ "tags": [],
+ "ExecuteTime": {
+ "end_time": "2025-01-14T13:22:49.823660Z",
+ "start_time": "2025-01-14T13:22:49.818879Z"
+ }
+ },
+ "source": [
+ "class AlexNet(torch.nn.Module):\n",
+ " \"\"\"\n",
+ " AlexNet model for image classification.\n",
+ "\n",
+ " Attributes\n",
+ " ----------\n",
+ " features : torch.nn.container.Sequential\n",
+ " The convolutional feature-extractor part of AlexNet.\n",
+ " avgpool : AdaptiveAvgPool2d\n",
+ " An adaptive pooling layer.\n",
+ " classifier : torch.nn.container.Sequential\n",
+ " The fully connected linear part of AlexNet.\n",
+ "\n",
+ " Methods\n",
+ " -------\n",
+ " forward()\n",
+ " The forward pass.\n",
+ " \"\"\"\n",
+ "\n",
+ " def __init__(self, num_classes: int = 1000, dropout: float = 0.5) -> None:\n",
+ " \"\"\"\n",
+ " Initialize AlexNet architecture.\n",
+ "\n",
+ " Parameters\n",
+ " ----------\n",
+ " num_classes : int\n",
+ " The number of classes in the underlying classification problem.\n",
+ " dropout : float\n",
+ " The dropout probability.\n",
+ " \"\"\"\n",
+ " super().__init__()\n",
+ " self.features = torch.nn.Sequential(\n",
+ " # AlexNet consists of 8 layers:\n",
+ " # 5 convolutional layers, some followed by max-pooling (see figure),\n",
+ " # and 3 fully connected layers.\n",
+ " # IMPLEMENT FEATURE-EXTRACTOR PART OF ALEXNET HERE!\n",
+ " # 1st convolutional layer (+ max-pooling)\n",
+ " torch.nn.Conv2d(3, 64, kernel_size=11, stride=4, padding=2),\n",
+ " torch.nn.ReLU(inplace=True),\n",
+ " torch.nn.MaxPool2d(kernel_size=3, stride=2),\n",
+ " # 2nd convolutional layer (+ max-pooling)\n",
+ " torch.nn.Conv2d(64, 192, kernel_size=5, padding=2),\n",
+ " torch.nn.ReLU(inplace=True),\n",
+ " torch.nn.MaxPool2d(kernel_size=3, stride=2),\n",
+ " # 3rd + 4th convolutional layer\n",
+ " torch.nn.Conv2d(192, 384, kernel_size=3, padding=1),\n",
+ " torch.nn.ReLU(inplace=True),\n",
+ " torch.nn.Conv2d(384, 256, kernel_size=3, padding=1),\n",
+ " torch.nn.ReLU(inplace=True),\n",
+ " # 5th convolutional layer\n",
+ " torch.nn.Conv2d(256, 256, kernel_size=3, padding=1),\n",
+ " torch.nn.ReLU(inplace=True),\n",
+ " torch.nn.MaxPool2d(kernel_size=3, stride=2),\n",
+ " )\n",
+ " # Average pooling to downscale possibly larger input images.\n",
+ " self.avgpool = torch.nn.AdaptiveAvgPool2d((6, 6))\n",
+ " self.classifier = torch.nn.Sequential(\n",
+ " # IMPLEMENT FULLY CONNECTED MULTI-LAYER PERCEPTRON PART HERE!\n",
+ " # 6th, 7th + 8th fully connected layer\n",
+ " torch.nn.Dropout(p=dropout),\n",
+ " torch.nn.Linear(256 * 6 * 6, 4096),\n",
+ " torch.nn.ReLU(inplace=True),\n",
+ " torch.nn.Dropout(p=dropout),\n",
+ " torch.nn.Linear(4096, 4096),\n",
+ " torch.nn.ReLU(inplace=True),\n",
+ " torch.nn.Linear(4096, num_classes),\n",
+ " ###################################\n",
+ " )\n",
+ "\n",
+ " # Every nn.Module subclass implements the operations on the input data in the forward method.\n",
+ " # Forward pass: Apply AlexNet model to input x.\n",
+ " def forward(self, x: torch.Tensor) -> torch.Tensor:\n",
+ " \"\"\"\n",
+ " The forward pass.\n",
+ "\n",
+ " Parameters\n",
+ " ----------\n",
+ " x : torch.Tensor\n",
+ " The input data.\n",
+ "\n",
+ " Returns\n",
+ " -------\n",
+ " torch.Tensor\n",
+ " The output.\n",
+ " \"\"\"\n",
+ " # IMPLEMENT OPERATIONS ON INPUT DATA x HERE!\n",
+ " x = self.features(x)\n",
+ " x = self.avgpool(x)\n",
+ " x = torch.flatten(x, 1)\n",
+ " x = self.classifier(x)\n",
+ " return x"
+ ],
+ "outputs": [],
+ "execution_count": 2
+ },
+ {
+ "cell_type": "code",
+ "metadata": {
+ "tags": [],
+ "ExecuteTime": {
+ "end_time": "2025-01-14T13:22:54.277129Z",
+ "start_time": "2025-01-14T13:22:54.262869Z"
+ }
+ },
+ "source": [
+ "# DATASET TRANSFORMS\n",
+ "def get_transforms_cifar10() -> (\n",
+ " Tuple[torchvision.transforms.Compose, torchvision.transforms.Compose]\n",
+ "):\n",
+ " \"\"\"\n",
+ " Get transforms applied to CIFAR-10 data for AlexNet training and inference.\n",
+ "\n",
+ " Returns\n",
+ " -------\n",
+ " torchvision.transforms.Compose\n",
+ " The transforms applied to CIFAR-10 for training AlexNet.\n",
+ " torchvision.transforms.Compose\n",
+ " The transforms applied to CIFAR-10 to run inference with AlexNet.\n",
+ " \"\"\"\n",
+ " # Transforms applied to training data (randomness to make network more robust against overfitting)\n",
+ " train_transforms = (\n",
+ " torchvision.transforms.Compose( # Compose several transforms together.\n",
+ " [\n",
+ " torchvision.transforms.Resize(\n",
+ " (70, 70)\n",
+ " ), # Upsample CIFAR-10 images to make them work with AlexNet.\n",
+ " torchvision.transforms.RandomCrop(\n",
+ " (64, 64)\n",
+ " ), # Randomly crop image to make NN more robust against overfitting.\n",
+ " torchvision.transforms.ToTensor(), # Convert image into torch tensor.\n",
+ " torchvision.transforms.Normalize(\n",
+ " (0.5, 0.5, 0.5), (0.5, 0.5, 0.5)\n",
+ " ), # Normalize to [-1,1] via (image-mean)/std.\n",
+ " ]\n",
+ " )\n",
+ " )\n",
+ "\n",
+ " test_transforms = torchvision.transforms.Compose(\n",
+ " [\n",
+ " torchvision.transforms.Resize((70, 70)),\n",
+ " torchvision.transforms.CenterCrop((64, 64)),\n",
+ " torchvision.transforms.ToTensor(),\n",
+ " torchvision.transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5)),\n",
+ " ]\n",
+ " )\n",
+ " return train_transforms, test_transforms\n",
+ "\n",
+ "# DETERMINE TRAIN AND VALIDATION SPLIT\n",
+ "def make_train_validation_split(\n",
+ " train_dataset: torchvision.datasets.CIFAR10,\n",
+ " seed: int = 123,\n",
+ " validation_fraction: float = 0.1,\n",
+ ") -> Tuple[np.ndarray, np.ndarray]:\n",
+ " \"\"\"\n",
+ " Split original CIFAR-10 training data into train and validation sets.\n",
+ "\n",
+ " Parameters\n",
+ " ----------\n",
+ " train_dataset : torchvision.datasets.CIFAR10\n",
+ " The original CIFAR-10 training dataset.\n",
+ " seed : int\n",
+ " The seed used to split the data.\n",
+ " validation_fraction : float\n",
+ " The fraction of samples used for validation.\n",
+ "\n",
+ " Returns\n",
+ " -------\n",
+ " numpy.ndarray\n",
+ " The sample indices for the training dataset.\n",
+ " numpy.ndarray\n",
+ " The sample indices for the validation dataset.\n",
+ " \"\"\"\n",
+ " num_samples = len(\n",
+ " train_dataset\n",
+ " ) # Get overall number of samples in original training data.\n",
+ " rng = np.random.default_rng(\n",
+ " seed=seed\n",
+ " ) # Set same seed over all ranks for consistent train-test split.\n",
+ " idx = np.arange(0, num_samples) # Construct array of all indices.\n",
+ " rng.shuffle(idx) # Shuffle them.\n",
+ " num_validate = int(\n",
+ " validation_fraction * num_samples\n",
+ " ) # Determine number of validation samples from validation split.\n",
+ " return (\n",
+ " idx[num_validate:],\n",
+ " idx[0:num_validate],\n",
+ " ) # Extract and return train and validation indices.\n",
+ "\n",
+ "# GET DATALOADERS\n",
+ "def get_dataloaders_cifar10(\n",
+ " batch_size: int,\n",
+ " data_root: str = \"data\",\n",
+ " validation_fraction: float = 0.1,\n",
+ " train_transforms: Callable[[Any], Any] = None,\n",
+ " test_transforms: Callable[[Any], Any] = None,\n",
+ " seed: int = 123,\n",
+ ") -> Tuple[\n",
+ " torch.utils.data.DataLoader,\n",
+ " torch.utils.data.DataLoader,\n",
+ " torch.utils.data.DataLoader,\n",
+ "]:\n",
+ " \"\"\"\n",
+ " Get dataloaders for training, validation, and testing on the CIFAR-10 dataset.\n",
+ "\n",
+ " Parameters\n",
+ " ----------\n",
+ " batch_size : int\n",
+ " The mini-batch size.\n",
+ " data_root : str\n",
+ " The data folder.\n",
+ " validation_fraction : float\n",
+ " The fraction of the original training data used for validating.\n",
+ " train_transforms : Callable[[Any], Any]\n",
+ " The transform applied to the training data.\n",
+ " test_transforms : Callable[[Any], Any]\n",
+ " The transform applied to the testing data (inference).\n",
+ " seed : int\n",
+ " The seed for the validation-train split.\n",
+ "\n",
+ " Returns\n",
+ " -------\n",
+ " torch.utils.data.DataLoader\n",
+ " The train dataloader.\n",
+ " torch.utils.data.DataLoader\n",
+ " The validation dataloader.\n",
+ " torch.utils.data.DataLoader\n",
+ " The test dataloader.\n",
+ " \"\"\"\n",
+ " if train_transforms is None:\n",
+ " train_transforms = torchvision.transforms.ToTensor()\n",
+ "\n",
+ " if test_transforms is None:\n",
+ " test_transforms = torchvision.transforms.ToTensor()\n",
+ "\n",
+ " train_dataset = torchvision.datasets.CIFAR10(\n",
+ " root=data_root, train=True, transform=train_transforms, download=True\n",
+ " )\n",
+ "\n",
+ " valid_dataset = torchvision.datasets.CIFAR10(\n",
+ " root=data_root, train=True, transform=test_transforms\n",
+ " )\n",
+ "\n",
+ " test_dataset = torchvision.datasets.CIFAR10(\n",
+ " root=data_root, train=False, transform=test_transforms\n",
+ " )\n",
+ "\n",
+ " # Perform index-based train-validation split of original training data.\n",
+ " train_indices, valid_indices = make_train_validation_split(\n",
+ " train_dataset, seed, validation_fraction\n",
+ " ) # Get train and validation indices.\n",
+ "\n",
+ " train_sampler = torch.utils.data.SubsetRandomSampler(train_indices)\n",
+ " valid_sampler = torch.utils.data.SubsetRandomSampler(valid_indices)\n",
+ "\n",
+ " valid_loader = torch.utils.data.DataLoader(\n",
+ " dataset=valid_dataset,\n",
+ " batch_size=batch_size,\n",
+ " sampler=valid_sampler,\n",
+ " )\n",
+ "\n",
+ " train_loader = torch.utils.data.DataLoader(\n",
+ " dataset=train_dataset,\n",
+ " batch_size=batch_size,\n",
+ " drop_last=True,\n",
+ " sampler=train_sampler,\n",
+ " )\n",
+ "\n",
+ " test_loader = torch.utils.data.DataLoader(\n",
+ " dataset=test_dataset,\n",
+ " batch_size=batch_size,\n",
+ " shuffle=False,\n",
+ " )\n",
+ "\n",
+ " return train_loader, valid_loader, test_loader\n",
+ "\n",
+ "# COMPUTE ACCURACY\n",
+ "def compute_accuracy(\n",
+ " model: torch.nn.Module,\n",
+ " data_loader: torch.utils.data.DataLoader,\n",
+ " device: torch.device,\n",
+ ") -> float:\n",
+ " \"\"\"\n",
+ " Compute the accuracy of the model's predictions on given labeled data.\n",
+ "\n",
+ " Parameters\n",
+ " ----------\n",
+ " model : torch.nn.Module\n",
+ " The model.\n",
+ " data_loader : torch.utils.data.DataLoader\n",
+ " The dataloader.\n",
+ " device : torch.device\n",
+ " The device to use.\n",
+ "\n",
+ " Returns\n",
+ " -------\n",
+ " float\n",
+ " The model's accuracy on the given dataset in percent.\n",
+ " \"\"\"\n",
+ " with torch.no_grad(): # Disable gradient calculation to reduce memory consumption.\n",
+ " correct_pred, num_examples = (\n",
+ " 0,\n",
+ " 0,\n",
+ " ) # Initialize number of correctly predicted and overall samples, respectively.\n",
+ "\n",
+ " for i, (features, targets) in enumerate(data_loader):\n",
+ " features = features.to(device)\n",
+ " targets = targets.float().to(device)\n",
+ "\n",
+ " logits = model(features)\n",
+ " _, predicted_labels = torch.max(logits, 1) # Get class with highest score.\n",
+ "\n",
+ " num_examples += targets.size(0)\n",
+ " correct_pred += (predicted_labels == targets).sum()\n",
+ " return correct_pred.float() / num_examples * 100\n",
+ "\n",
+ "\n",
+ "# TRAIN MODEL (NON-PARALLEL)\n",
+ "def train_model(\n",
+ " model: torch.nn.Module,\n",
+ " num_epochs: int,\n",
+ " train_loader: torch.utils.data.DataLoader,\n",
+ " valid_loader: torch.utils.data.DataLoader,\n",
+ " test_loader: torch.utils.data.DataLoader,\n",
+ " optimizer: torch.optim.Optimizer,\n",
+ " device: torch.device,\n",
+ " logging_interval: int = 50,\n",
+ " scheduler: Union[torch.optim.lr_scheduler._LRScheduler, ReduceLROnPlateau] = None,\n",
+ ") -> Tuple[List[float], List[float], List[float]]:\n",
+ " \"\"\"\n",
+ " Train your model.\n",
+ "\n",
+ " Parameters\n",
+ " ----------\n",
+ " model : torch.nn.Module\n",
+ " The model to train.\n",
+ " num_epochs : int\n",
+ " The number of epochs to train\n",
+ " train_loader : torch.utils.data.DataLoader\n",
+ " The training dataloader.\n",
+ " valid_loader : torch.utils.data.DataLoader\n",
+ " The validation dataloader.,\n",
+ " test_loader : torch.utils.data.DataLoader\n",
+ " The testing dataloader.\n",
+ " optimizer : torch.optim.Optimizer\n",
+ " The optimizer to use.\n",
+ " device : torch.device\n",
+ " The device to train on.\n",
+ " logging_interval : int\n",
+ " The logging interval.\n",
+ " scheduler : torch.optim.lr_scheduler._LRScheduler\n",
+ " An optional learning rate scheduler.\n",
+ "\n",
+ " Returns\n",
+ " -------\n",
+ " List[float]\n",
+ " The loss history.\n",
+ " List[float]\n",
+ " The training accuracy history.\n",
+ " List[float]\n",
+ " The validation accuracy history.\n",
+ " \"\"\"\n",
+ " start = time.perf_counter() # Measure training time.\n",
+ "\n",
+ " # Initialize history lists for loss, training accuracy, and validation accuracy.\n",
+ " loss_history, train_acc_history, valid_acc_history = [], [], []\n",
+ "\n",
+ " for epoch in range(num_epochs): # Loop over epochs.\n",
+ " model.train() # Set model to training mode.\n",
+ " # Thus, layers like dropout which behave differently on train and test procedures\n",
+ " # know what is going on and can behave accordingly. model.train() sets the mode to\n",
+ " # train. One might expect this to train model but it does not do that.\n",
+ " # Call either model.eval() or model.train(mode=False) to tell that you are testing.\n",
+ "\n",
+ " for batch_idx, (features, targets) in enumerate(\n",
+ " train_loader\n",
+ " ): # Loop over mini batches.\n",
+ " features = features.to(device) # Move features to used device.\n",
+ " targets = targets.to(device) # Move targets to used device.\n",
+ "\n",
+ " # Forward and backward pass.\n",
+ " logits = model(features) # Calculate logits as the model's output.\n",
+ " loss = torch.nn.functional.cross_entropy(\n",
+ " logits, targets\n",
+ " ) # Calculate cross-entropy loss.\n",
+ " optimizer.zero_grad() # Zero out gradients.\n",
+ " # Gradients are accumulated and not overwritten whenever .backward() is called.\n",
+ " loss.backward() # Calculate gradients of loss w.r.t. model parameters in backward pass.\n",
+ " optimizer.step() # Perform single optimization step to update model parameters.\n",
+ "\n",
+ " # Logging.\n",
+ " loss_history.append(loss.item())\n",
+ " if not batch_idx % logging_interval:\n",
+ " print(\n",
+ " f\"Epoch: {epoch+1:03d}/{num_epochs:03d} \"\n",
+ " f\"| Batch {batch_idx:04d}/{len(train_loader):04d} \"\n",
+ " f\"| Loss: {loss:.4f}\"\n",
+ " )\n",
+ "\n",
+ " model.eval() # Set model to evaluation mode.\n",
+ "\n",
+ " with torch.no_grad(): # Disable gradient calculation to reduce memory consumption.\n",
+ " train_acc = compute_accuracy(\n",
+ " model, train_loader, device=device\n",
+ " ) # Compute accuracy on training data.\n",
+ " valid_acc = compute_accuracy(\n",
+ " model, valid_loader, device=device\n",
+ " ) # Compute accuracy on validation data.\n",
+ " print(\n",
+ " f\"Epoch: {epoch+1:03d}/{num_epochs:03d} \"\n",
+ " f\"| Train: {train_acc :.2f}% \"\n",
+ " f\"| Validation: {valid_acc :.2f}%\"\n",
+ " )\n",
+ " train_acc_history.append(train_acc)\n",
+ " valid_acc_history.append(valid_acc)\n",
+ "\n",
+ " elapsed = (time.perf_counter() - start) / 60 # Measure training time per epoch.\n",
+ " print(f\"Time elapsed: {elapsed:.2f} min\")\n",
+ "\n",
+ " if scheduler is not None:\n",
+ " original_lr = scheduler.get_last_lr()[0]\n",
+ " scheduler.step(valid_acc_history[-1])\n",
+ " new_lr = scheduler.get_last_lr()[0]\n",
+ " if original_lr != new_lr:\n",
+ " print(f\"Epoch: {epoch+1:03d}/{num_epochs:03d}: LR updated to {new_lr}\")\n",
+ "\n",
+ " elapsed = (time.perf_counter() - start) / 60 # Measure total training time.\n",
+ " print(f\"Total Training Time: {elapsed:.2f} min\")\n",
+ "\n",
+ " test_acc = compute_accuracy(\n",
+ " model, test_loader, device=device\n",
+ " ) # Compute accuracy on test data.\n",
+ " print(f\"Test accuracy {test_acc :.2f}%\")\n",
+ "\n",
+ " return loss_history, train_acc_history, valid_acc_history"
+ ],
+ "outputs": [],
+ "execution_count": 3
+ },
+ {
+ "cell_type": "code",
+ "metadata": {
+ "tags": [],
+ "ExecuteTime": {
+ "end_time": "2025-01-14T13:23:02.964441Z",
+ "start_time": "2025-01-14T13:23:02.945857Z"
+ }
+ },
+ "source": [
+ "# SETTINGS\n",
+ "# Set random seed.\n",
+ "seed = 123 \n",
+ "torch.manual_seed(seed)\n",
+ "torch.cuda.manual_seed_all(seed)\n",
+ "\n",
+ "b = 256 # Set batch size.\n",
+ "e = 200 # Set number of epochs to be trained.\n",
+ "data_root = \"/pfs/work7/workspace/scratch/ku4408-VL-ScalableAI/data/cifar\"\n",
+ "\n",
+ "# Check for available device\n",
+ "if torch.cuda.is_available():\n",
+ " device = torch.device(\"cuda:0\") # CUDA-enabled GPU\n",
+ "elif torch.backends.mps.is_available():\n",
+ " device = torch.device(\"mps\") # Metal Performance Shaders for Apple Silicon\n",
+ "else:\n",
+ " device = torch.device(\"cpu\")\n",
+ " # Default to CPU\n",
+ "print(f\"Using {device} device.\")"
+ ],
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Using mps device.\n"
+ ]
+ }
+ ],
+ "execution_count": 4
+ },
+ {
+ "metadata": {
+ "ExecuteTime": {
+ "end_time": "2025-01-14T13:23:08.688800Z",
+ "start_time": "2025-01-14T13:23:07.494443Z"
+ }
+ },
+ "cell_type": "code",
+ "source": [
+ "# DATASET\n",
+ "# Using transforms on your data allows you to take it from its source state\n",
+ "# and transform it into data that’s ready for training.\n",
+ "\n",
+ "# Get transforms applied to CIFAR-10 data for training and inference.\n",
+ "train_transforms, test_transforms = get_transforms_cifar10()\n",
+ "\n",
+ "# Get PyTorch dataloaders for training, testing, and validation dataset.\n",
+ "train_loader, valid_loader, test_loader = get_dataloaders_cifar10(\n",
+ " batch_size=b, # The batch size.\n",
+ " data_root=data_root, # The path to the data dir.\n",
+ " validation_fraction=0.1, # The validation fraction.\n",
+ " train_transforms=train_transforms, # The transforms applied to the data at training time.\n",
+ " test_transforms=test_transforms, # The transforms applied to the data at inference time.\n",
+ ")\n",
+ "\n",
+ "# Check loaded dataset.\n",
+ "for images, labels in train_loader:\n",
+ " print(\"Image batch dimensions:\", images.shape)\n",
+ " print(\"Image label dimensions:\", labels.shape)\n",
+ " print(\"Class labels of 10 examples:\", labels[:10])\n",
+ " break"
+ ],
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Files already downloaded and verified\n",
+ "Image batch dimensions: torch.Size([256, 3, 64, 64])\n",
+ "Image label dimensions: torch.Size([256])\n",
+ "Class labels of 10 examples: tensor([2, 7, 8, 2, 2, 2, 9, 7, 6, 3])\n"
+ ]
+ }
+ ],
+ "execution_count": 5
+ },
+ {
+ "metadata": {
+ "ExecuteTime": {
+ "end_time": "2025-01-14T13:23:14.671139Z",
+ "start_time": "2025-01-14T13:23:14.464259Z"
+ }
+ },
+ "cell_type": "code",
+ "source": [
+ "# MODEL\n",
+ "# Define neural network by subclassing PyTorch's nn.Module.\n",
+ "\n",
+ "model = AlexNet(num_classes=10).to(\n",
+ " device\n",
+ ") # Build instance of AlexNet with 10 classes and move it to device.\n",
+ "\n",
+ "# torch.optim package implements various optimization algorithms.\n",
+ "# To use torch.optim you have to construct an optimizer object,\n",
+ "# that will hold the current state and will update the parameters\n",
+ "# based on computed gradients.\n",
+ "# Stochastic gradient descent (optionally with momentum):\n",
+ "optimizer = torch.optim.SGD(model.parameters(), momentum=0.9, lr=0.1)\n",
+ "\n",
+ "# torch.optim.lr_scheduler provides several learning-rate adjustment methods based on number of epochs.\n",
+ "# torch.optim.lr_scheduler.ReduceLROnPlateau: dynamic learning rate reducing based on some validation measurements.\n",
+ "# Reduce learning rate when a metric has stopped improving:\n",
+ "scheduler = torch.optim.lr_scheduler.ReduceLROnPlateau(\n",
+ " optimizer, factor=0.1, mode=\"max\",\n",
+ ")"
+ ],
+ "outputs": [],
+ "execution_count": 6
+ },
+ {
+ "metadata": {
+ "ExecuteTime": {
+ "end_time": "2025-01-14T14:20:37.257677Z",
+ "start_time": "2025-01-14T13:23:19.098407Z"
+ }
+ },
+ "cell_type": "code",
+ "source": [
+ "# Train model.\n",
+ "loss_history, train_acc_history, valid_acc_history = train_model(\n",
+ " model=model, # The model to train.\n",
+ " num_epochs=e, # The number of epochs to train.\n",
+ " train_loader=train_loader, # The training dataloader.\n",
+ " valid_loader=valid_loader, # The validation dataloader.\n",
+ " test_loader=test_loader, # The test dataloader.\n",
+ " optimizer=optimizer, # The optimizer.\n",
+ " device=device, # The device to train on.\n",
+ " scheduler=scheduler, # The learning rate scheduler.\n",
+ " logging_interval=100, # The logging interval.\n",
+ ")"
+ ],
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Epoch: 001/200 | Batch 0000/0175 | Loss: 2.3029\n",
+ "Epoch: 001/200 | Batch 0100/0175 | Loss: 2.1824\n",
+ "Epoch: 001/200 | Train: 25.69% | Validation: 26.06%\n",
+ "Time elapsed: 0.28 min\n",
+ "Epoch: 002/200 | Batch 0000/0175 | Loss: 1.9147\n",
+ "Epoch: 002/200 | Batch 0100/0175 | Loss: 1.8889\n",
+ "Epoch: 002/200 | Train: 22.56% | Validation: 22.38%\n",
+ "Time elapsed: 0.56 min\n",
+ "Epoch: 003/200 | Batch 0000/0175 | Loss: 1.8720\n",
+ "Epoch: 003/200 | Batch 0100/0175 | Loss: 1.5964\n",
+ "Epoch: 003/200 | Train: 40.20% | Validation: 40.26%\n",
+ "Time elapsed: 0.85 min\n",
+ "Epoch: 004/200 | Batch 0000/0175 | Loss: 1.5165\n",
+ "Epoch: 004/200 | Batch 0100/0175 | Loss: 1.4990\n",
+ "Epoch: 004/200 | Train: 44.60% | Validation: 45.90%\n",
+ "Time elapsed: 1.15 min\n",
+ "Epoch: 005/200 | Batch 0000/0175 | Loss: 1.4800\n",
+ "Epoch: 005/200 | Batch 0100/0175 | Loss: 1.2271\n",
+ "Epoch: 005/200 | Train: 55.16% | Validation: 55.96%\n",
+ "Time elapsed: 1.44 min\n",
+ "Epoch: 006/200 | Batch 0000/0175 | Loss: 1.3376\n",
+ "Epoch: 006/200 | Batch 0100/0175 | Loss: 1.2459\n",
+ "Epoch: 006/200 | Train: 56.44% | Validation: 56.28%\n",
+ "Time elapsed: 1.73 min\n",
+ "Epoch: 007/200 | Batch 0000/0175 | Loss: 1.2705\n",
+ "Epoch: 007/200 | Batch 0100/0175 | Loss: 1.0853\n",
+ "Epoch: 007/200 | Train: 58.54% | Validation: 57.30%\n",
+ "Time elapsed: 2.02 min\n",
+ "Epoch: 008/200 | Batch 0000/0175 | Loss: 1.2854\n",
+ "Epoch: 008/200 | Batch 0100/0175 | Loss: 1.0862\n",
+ "Epoch: 008/200 | Train: 62.36% | Validation: 62.02%\n",
+ "Time elapsed: 2.32 min\n",
+ "Epoch: 009/200 | Batch 0000/0175 | Loss: 1.1864\n",
+ "Epoch: 009/200 | Batch 0100/0175 | Loss: 1.0656\n",
+ "Epoch: 009/200 | Train: 61.21% | Validation: 59.90%\n",
+ "Time elapsed: 2.61 min\n",
+ "Epoch: 010/200 | Batch 0000/0175 | Loss: 1.1219\n",
+ "Epoch: 010/200 | Batch 0100/0175 | Loss: 1.1063\n",
+ "Epoch: 010/200 | Train: 67.37% | Validation: 66.28%\n",
+ "Time elapsed: 2.90 min\n",
+ "Epoch: 011/200 | Batch 0000/0175 | Loss: 1.0636\n",
+ "Epoch: 011/200 | Batch 0100/0175 | Loss: 1.0564\n",
+ "Epoch: 011/200 | Train: 66.91% | Validation: 65.04%\n",
+ "Time elapsed: 3.19 min\n",
+ "Epoch: 012/200 | Batch 0000/0175 | Loss: 0.9975\n",
+ "Epoch: 012/200 | Batch 0100/0175 | Loss: 1.0009\n",
+ "Epoch: 012/200 | Train: 69.77% | Validation: 67.50%\n",
+ "Time elapsed: 3.48 min\n",
+ "Epoch: 013/200 | Batch 0000/0175 | Loss: 0.8987\n",
+ "Epoch: 013/200 | Batch 0100/0175 | Loss: 0.9828\n",
+ "Epoch: 013/200 | Train: 70.00% | Validation: 67.32%\n",
+ "Time elapsed: 3.76 min\n",
+ "Epoch: 014/200 | Batch 0000/0175 | Loss: 0.8048\n",
+ "Epoch: 014/200 | Batch 0100/0175 | Loss: 1.0409\n",
+ "Epoch: 014/200 | Train: 70.07% | Validation: 67.52%\n",
+ "Time elapsed: 4.05 min\n",
+ "Epoch: 015/200 | Batch 0000/0175 | Loss: 0.8784\n",
+ "Epoch: 015/200 | Batch 0100/0175 | Loss: 0.9021\n",
+ "Epoch: 015/200 | Train: 68.94% | Validation: 65.92%\n",
+ "Time elapsed: 4.36 min\n",
+ "Epoch: 016/200 | Batch 0000/0175 | Loss: 0.9163\n",
+ "Epoch: 016/200 | Batch 0100/0175 | Loss: 0.9878\n",
+ "Epoch: 016/200 | Train: 72.91% | Validation: 69.38%\n",
+ "Time elapsed: 4.64 min\n",
+ "Epoch: 017/200 | Batch 0000/0175 | Loss: 0.7744\n",
+ "Epoch: 017/200 | Batch 0100/0175 | Loss: 0.9099\n",
+ "Epoch: 017/200 | Train: 72.86% | Validation: 69.96%\n",
+ "Time elapsed: 4.93 min\n",
+ "Epoch: 018/200 | Batch 0000/0175 | Loss: 0.9303\n",
+ "Epoch: 018/200 | Batch 0100/0175 | Loss: 0.9369\n",
+ "Epoch: 018/200 | Train: 70.88% | Validation: 67.64%\n",
+ "Time elapsed: 5.21 min\n",
+ "Epoch: 019/200 | Batch 0000/0175 | Loss: 0.9380\n",
+ "Epoch: 019/200 | Batch 0100/0175 | Loss: 0.8983\n",
+ "Epoch: 019/200 | Train: 73.48% | Validation: 69.56%\n",
+ "Time elapsed: 5.50 min\n",
+ "Epoch: 020/200 | Batch 0000/0175 | Loss: 0.7723\n",
+ "Epoch: 020/200 | Batch 0100/0175 | Loss: 0.8005\n",
+ "Epoch: 020/200 | Train: 75.03% | Validation: 70.52%\n",
+ "Time elapsed: 5.79 min\n",
+ "Epoch: 021/200 | Batch 0000/0175 | Loss: 0.6704\n",
+ "Epoch: 021/200 | Batch 0100/0175 | Loss: 0.9358\n",
+ "Epoch: 021/200 | Train: 72.56% | Validation: 68.30%\n",
+ "Time elapsed: 6.08 min\n",
+ "Epoch: 022/200 | Batch 0000/0175 | Loss: 0.7927\n",
+ "Epoch: 022/200 | Batch 0100/0175 | Loss: 0.7194\n",
+ "Epoch: 022/200 | Train: 72.83% | Validation: 68.94%\n",
+ "Time elapsed: 6.38 min\n",
+ "Epoch: 023/200 | Batch 0000/0175 | Loss: 0.7630\n",
+ "Epoch: 023/200 | Batch 0100/0175 | Loss: 0.8087\n",
+ "Epoch: 023/200 | Train: 71.80% | Validation: 68.86%\n",
+ "Time elapsed: 6.67 min\n",
+ "Epoch: 024/200 | Batch 0000/0175 | Loss: 0.8250\n",
+ "Epoch: 024/200 | Batch 0100/0175 | Loss: 0.8487\n",
+ "Epoch: 024/200 | Train: 72.42% | Validation: 67.86%\n",
+ "Time elapsed: 6.95 min\n",
+ "Epoch: 025/200 | Batch 0000/0175 | Loss: 0.9284\n",
+ "Epoch: 025/200 | Batch 0100/0175 | Loss: 0.7290\n",
+ "Epoch: 025/200 | Train: 70.52% | Validation: 67.18%\n",
+ "Time elapsed: 7.27 min\n",
+ "Epoch: 026/200 | Batch 0000/0175 | Loss: 1.1236\n",
+ "Epoch: 026/200 | Batch 0100/0175 | Loss: 0.9713\n",
+ "Epoch: 026/200 | Train: 70.92% | Validation: 67.32%\n",
+ "Time elapsed: 7.55 min\n",
+ "Epoch: 027/200 | Batch 0000/0175 | Loss: 0.8326\n",
+ "Epoch: 027/200 | Batch 0100/0175 | Loss: 0.9268\n",
+ "Epoch: 027/200 | Train: 71.66% | Validation: 67.40%\n",
+ "Time elapsed: 7.84 min\n",
+ "Epoch: 028/200 | Batch 0000/0175 | Loss: 0.9239\n",
+ "Epoch: 028/200 | Batch 0100/0175 | Loss: 1.0577\n",
+ "Epoch: 028/200 | Train: 70.71% | Validation: 66.82%\n",
+ "Time elapsed: 8.12 min\n",
+ "Epoch: 029/200 | Batch 0000/0175 | Loss: 0.9516\n",
+ "Epoch: 029/200 | Batch 0100/0175 | Loss: 0.9217\n",
+ "Epoch: 029/200 | Train: 62.73% | Validation: 59.28%\n",
+ "Time elapsed: 8.40 min\n",
+ "Epoch: 030/200 | Batch 0000/0175 | Loss: 0.9482\n",
+ "Epoch: 030/200 | Batch 0100/0175 | Loss: 1.0360\n",
+ "Epoch: 030/200 | Train: 63.89% | Validation: 61.22%\n",
+ "Time elapsed: 8.69 min\n",
+ "Epoch: 031/200 | Batch 0000/0175 | Loss: 1.0182\n",
+ "Epoch: 031/200 | Batch 0100/0175 | Loss: 1.0315\n",
+ "Epoch: 031/200 | Train: 64.46% | Validation: 61.14%\n",
+ "Time elapsed: 8.97 min\n",
+ "Epoch: 031/200: LR updated to 0.010000000000000002\n",
+ "Epoch: 032/200 | Batch 0000/0175 | Loss: 1.2962\n",
+ "Epoch: 032/200 | Batch 0100/0175 | Loss: 0.6454\n",
+ "Epoch: 032/200 | Train: 77.36% | Validation: 72.40%\n",
+ "Time elapsed: 9.26 min\n",
+ "Epoch: 033/200 | Batch 0000/0175 | Loss: 0.7028\n",
+ "Epoch: 033/200 | Batch 0100/0175 | Loss: 0.7370\n",
+ "Epoch: 033/200 | Train: 78.80% | Validation: 73.40%\n",
+ "Time elapsed: 9.55 min\n",
+ "Epoch: 034/200 | Batch 0000/0175 | Loss: 0.6822\n",
+ "Epoch: 034/200 | Batch 0100/0175 | Loss: 0.5398\n",
+ "Epoch: 034/200 | Train: 79.61% | Validation: 74.10%\n",
+ "Time elapsed: 9.83 min\n",
+ "Epoch: 035/200 | Batch 0000/0175 | Loss: 0.5499\n",
+ "Epoch: 035/200 | Batch 0100/0175 | Loss: 0.5204\n",
+ "Epoch: 035/200 | Train: 80.86% | Validation: 75.22%\n",
+ "Time elapsed: 10.11 min\n",
+ "Epoch: 036/200 | Batch 0000/0175 | Loss: 0.6343\n",
+ "Epoch: 036/200 | Batch 0100/0175 | Loss: 0.5943\n",
+ "Epoch: 036/200 | Train: 81.44% | Validation: 75.48%\n",
+ "Time elapsed: 10.40 min\n",
+ "Epoch: 037/200 | Batch 0000/0175 | Loss: 0.5913\n",
+ "Epoch: 037/200 | Batch 0100/0175 | Loss: 0.5732\n",
+ "Epoch: 037/200 | Train: 82.02% | Validation: 75.54%\n",
+ "Time elapsed: 10.72 min\n",
+ "Epoch: 038/200 | Batch 0000/0175 | Loss: 0.5913\n",
+ "Epoch: 038/200 | Batch 0100/0175 | Loss: 0.5917\n",
+ "Epoch: 038/200 | Train: 82.42% | Validation: 75.56%\n",
+ "Time elapsed: 11.02 min\n",
+ "Epoch: 039/200 | Batch 0000/0175 | Loss: 0.4646\n",
+ "Epoch: 039/200 | Batch 0100/0175 | Loss: 0.6245\n",
+ "Epoch: 039/200 | Train: 83.04% | Validation: 75.60%\n",
+ "Time elapsed: 11.33 min\n",
+ "Epoch: 040/200 | Batch 0000/0175 | Loss: 0.4767\n",
+ "Epoch: 040/200 | Batch 0100/0175 | Loss: 0.4135\n",
+ "Epoch: 040/200 | Train: 83.26% | Validation: 75.40%\n",
+ "Time elapsed: 11.65 min\n",
+ "Epoch: 041/200 | Batch 0000/0175 | Loss: 0.4731\n",
+ "Epoch: 041/200 | Batch 0100/0175 | Loss: 0.5066\n",
+ "Epoch: 041/200 | Train: 84.04% | Validation: 76.36%\n",
+ "Time elapsed: 11.96 min\n",
+ "Epoch: 042/200 | Batch 0000/0175 | Loss: 0.4673\n",
+ "Epoch: 042/200 | Batch 0100/0175 | Loss: 0.4167\n",
+ "Epoch: 042/200 | Train: 84.30% | Validation: 76.42%\n",
+ "Time elapsed: 12.30 min\n",
+ "Epoch: 043/200 | Batch 0000/0175 | Loss: 0.4727\n",
+ "Epoch: 043/200 | Batch 0100/0175 | Loss: 0.5059\n",
+ "Epoch: 043/200 | Train: 84.98% | Validation: 76.28%\n",
+ "Time elapsed: 12.61 min\n",
+ "Epoch: 044/200 | Batch 0000/0175 | Loss: 0.4300\n",
+ "Epoch: 044/200 | Batch 0100/0175 | Loss: 0.5211\n",
+ "Epoch: 044/200 | Train: 85.42% | Validation: 76.20%\n",
+ "Time elapsed: 12.93 min\n",
+ "Epoch: 045/200 | Batch 0000/0175 | Loss: 0.3317\n",
+ "Epoch: 045/200 | Batch 0100/0175 | Loss: 0.4534\n",
+ "Epoch: 045/200 | Train: 85.95% | Validation: 76.32%\n",
+ "Time elapsed: 13.23 min\n",
+ "Epoch: 046/200 | Batch 0000/0175 | Loss: 0.4260\n",
+ "Epoch: 046/200 | Batch 0100/0175 | Loss: 0.5116\n",
+ "Epoch: 046/200 | Train: 86.42% | Validation: 76.44%\n",
+ "Time elapsed: 13.51 min\n",
+ "Epoch: 047/200 | Batch 0000/0175 | Loss: 0.3952\n",
+ "Epoch: 047/200 | Batch 0100/0175 | Loss: 0.3795\n",
+ "Epoch: 047/200 | Train: 86.61% | Validation: 76.54%\n",
+ "Time elapsed: 13.80 min\n",
+ "Epoch: 048/200 | Batch 0000/0175 | Loss: 0.4316\n",
+ "Epoch: 048/200 | Batch 0100/0175 | Loss: 0.3896\n",
+ "Epoch: 048/200 | Train: 87.42% | Validation: 76.46%\n",
+ "Time elapsed: 14.08 min\n",
+ "Epoch: 049/200 | Batch 0000/0175 | Loss: 0.2942\n",
+ "Epoch: 049/200 | Batch 0100/0175 | Loss: 0.4303\n",
+ "Epoch: 049/200 | Train: 87.36% | Validation: 76.10%\n",
+ "Time elapsed: 14.35 min\n",
+ "Epoch: 050/200 | Batch 0000/0175 | Loss: 0.4284\n",
+ "Epoch: 050/200 | Batch 0100/0175 | Loss: 0.3098\n",
+ "Epoch: 050/200 | Train: 88.25% | Validation: 76.12%\n",
+ "Time elapsed: 14.63 min\n",
+ "Epoch: 051/200 | Batch 0000/0175 | Loss: 0.3231\n",
+ "Epoch: 051/200 | Batch 0100/0175 | Loss: 0.3541\n",
+ "Epoch: 051/200 | Train: 88.56% | Validation: 75.90%\n",
+ "Time elapsed: 14.91 min\n",
+ "Epoch: 052/200 | Batch 0000/0175 | Loss: 0.2490\n",
+ "Epoch: 052/200 | Batch 0100/0175 | Loss: 0.3158\n",
+ "Epoch: 052/200 | Train: 88.78% | Validation: 76.28%\n",
+ "Time elapsed: 15.20 min\n",
+ "Epoch: 053/200 | Batch 0000/0175 | Loss: 0.3133\n",
+ "Epoch: 053/200 | Batch 0100/0175 | Loss: 0.3114\n",
+ "Epoch: 053/200 | Train: 88.99% | Validation: 76.20%\n",
+ "Time elapsed: 15.48 min\n",
+ "Epoch: 054/200 | Batch 0000/0175 | Loss: 0.2950\n",
+ "Epoch: 054/200 | Batch 0100/0175 | Loss: 0.3328\n",
+ "Epoch: 054/200 | Train: 89.62% | Validation: 76.14%\n",
+ "Time elapsed: 15.76 min\n",
+ "Epoch: 055/200 | Batch 0000/0175 | Loss: 0.3265\n",
+ "Epoch: 055/200 | Batch 0100/0175 | Loss: 0.3143\n",
+ "Epoch: 055/200 | Train: 90.23% | Validation: 76.42%\n",
+ "Time elapsed: 16.05 min\n",
+ "Epoch: 056/200 | Batch 0000/0175 | Loss: 0.2905\n",
+ "Epoch: 056/200 | Batch 0100/0175 | Loss: 0.3923\n",
+ "Epoch: 056/200 | Train: 90.56% | Validation: 76.40%\n",
+ "Time elapsed: 16.33 min\n",
+ "Epoch: 057/200 | Batch 0000/0175 | Loss: 0.3742\n",
+ "Epoch: 057/200 | Batch 0100/0175 | Loss: 0.2765\n",
+ "Epoch: 057/200 | Train: 90.82% | Validation: 76.10%\n",
+ "Time elapsed: 16.61 min\n",
+ "Epoch: 058/200 | Batch 0000/0175 | Loss: 0.2525\n",
+ "Epoch: 058/200 | Batch 0100/0175 | Loss: 0.3044\n",
+ "Epoch: 058/200 | Train: 91.26% | Validation: 76.14%\n",
+ "Time elapsed: 16.89 min\n",
+ "Epoch: 058/200: LR updated to 0.0010000000000000002\n",
+ "Epoch: 059/200 | Batch 0000/0175 | Loss: 0.2535\n",
+ "Epoch: 059/200 | Batch 0100/0175 | Loss: 0.2677\n",
+ "Epoch: 059/200 | Train: 91.91% | Validation: 76.78%\n",
+ "Time elapsed: 17.17 min\n",
+ "Epoch: 060/200 | Batch 0000/0175 | Loss: 0.2500\n",
+ "Epoch: 060/200 | Batch 0100/0175 | Loss: 0.2527\n",
+ "Epoch: 060/200 | Train: 92.31% | Validation: 76.52%\n",
+ "Time elapsed: 17.45 min\n",
+ "Epoch: 061/200 | Batch 0000/0175 | Loss: 0.2716\n",
+ "Epoch: 061/200 | Batch 0100/0175 | Loss: 0.1871\n",
+ "Epoch: 061/200 | Train: 92.23% | Validation: 76.62%\n",
+ "Time elapsed: 17.73 min\n",
+ "Epoch: 062/200 | Batch 0000/0175 | Loss: 0.2492\n",
+ "Epoch: 062/200 | Batch 0100/0175 | Loss: 0.2027\n",
+ "Epoch: 062/200 | Train: 92.71% | Validation: 76.50%\n",
+ "Time elapsed: 18.01 min\n",
+ "Epoch: 063/200 | Batch 0000/0175 | Loss: 0.2168\n",
+ "Epoch: 063/200 | Batch 0100/0175 | Loss: 0.2215\n",
+ "Epoch: 063/200 | Train: 92.61% | Validation: 76.38%\n",
+ "Time elapsed: 18.29 min\n",
+ "Epoch: 064/200 | Batch 0000/0175 | Loss: 0.2323\n",
+ "Epoch: 064/200 | Batch 0100/0175 | Loss: 0.2461\n",
+ "Epoch: 064/200 | Train: 92.79% | Validation: 76.50%\n",
+ "Time elapsed: 18.57 min\n",
+ "Epoch: 065/200 | Batch 0000/0175 | Loss: 0.1353\n",
+ "Epoch: 065/200 | Batch 0100/0175 | Loss: 0.2739\n",
+ "Epoch: 065/200 | Train: 92.75% | Validation: 76.42%\n",
+ "Time elapsed: 18.85 min\n",
+ "Epoch: 066/200 | Batch 0000/0175 | Loss: 0.2344\n",
+ "Epoch: 066/200 | Batch 0100/0175 | Loss: 0.2782\n",
+ "Epoch: 066/200 | Train: 92.74% | Validation: 76.28%\n",
+ "Time elapsed: 19.13 min\n",
+ "Epoch: 067/200 | Batch 0000/0175 | Loss: 0.1724\n",
+ "Epoch: 067/200 | Batch 0100/0175 | Loss: 0.3138\n",
+ "Epoch: 067/200 | Train: 93.05% | Validation: 76.36%\n",
+ "Time elapsed: 19.41 min\n",
+ "Epoch: 068/200 | Batch 0000/0175 | Loss: 0.1485\n",
+ "Epoch: 068/200 | Batch 0100/0175 | Loss: 0.2083\n",
+ "Epoch: 068/200 | Train: 93.06% | Validation: 76.12%\n",
+ "Time elapsed: 19.69 min\n",
+ "Epoch: 069/200 | Batch 0000/0175 | Loss: 0.2044\n",
+ "Epoch: 069/200 | Batch 0100/0175 | Loss: 0.1851\n",
+ "Epoch: 069/200 | Train: 93.07% | Validation: 76.36%\n",
+ "Time elapsed: 19.97 min\n",
+ "Epoch: 070/200 | Batch 0000/0175 | Loss: 0.2625\n",
+ "Epoch: 070/200 | Batch 0100/0175 | Loss: 0.2367\n",
+ "Epoch: 070/200 | Train: 93.26% | Validation: 76.00%\n",
+ "Time elapsed: 20.25 min\n",
+ "Epoch: 070/200: LR updated to 0.00010000000000000003\n",
+ "Epoch: 071/200 | Batch 0000/0175 | Loss: 0.1700\n",
+ "Epoch: 071/200 | Batch 0100/0175 | Loss: 0.2359\n",
+ "Epoch: 071/200 | Train: 93.07% | Validation: 76.00%\n",
+ "Time elapsed: 20.53 min\n",
+ "Epoch: 072/200 | Batch 0000/0175 | Loss: 0.1903\n",
+ "Epoch: 072/200 | Batch 0100/0175 | Loss: 0.2366\n",
+ "Epoch: 072/200 | Train: 93.04% | Validation: 76.16%\n",
+ "Time elapsed: 20.81 min\n",
+ "Epoch: 073/200 | Batch 0000/0175 | Loss: 0.1941\n",
+ "Epoch: 073/200 | Batch 0100/0175 | Loss: 0.1818\n",
+ "Epoch: 073/200 | Train: 93.19% | Validation: 76.22%\n",
+ "Time elapsed: 21.09 min\n",
+ "Epoch: 074/200 | Batch 0000/0175 | Loss: 0.2620\n",
+ "Epoch: 074/200 | Batch 0100/0175 | Loss: 0.1616\n",
+ "Epoch: 074/200 | Train: 93.15% | Validation: 76.14%\n",
+ "Time elapsed: 21.40 min\n",
+ "Epoch: 075/200 | Batch 0000/0175 | Loss: 0.1505\n",
+ "Epoch: 075/200 | Batch 0100/0175 | Loss: 0.2382\n",
+ "Epoch: 075/200 | Train: 93.22% | Validation: 76.10%\n",
+ "Time elapsed: 21.68 min\n",
+ "Epoch: 076/200 | Batch 0000/0175 | Loss: 0.2063\n",
+ "Epoch: 076/200 | Batch 0100/0175 | Loss: 0.1891\n",
+ "Epoch: 076/200 | Train: 93.24% | Validation: 76.04%\n",
+ "Time elapsed: 21.97 min\n",
+ "Epoch: 077/200 | Batch 0000/0175 | Loss: 0.1941\n",
+ "Epoch: 077/200 | Batch 0100/0175 | Loss: 0.1875\n",
+ "Epoch: 077/200 | Train: 93.28% | Validation: 76.10%\n",
+ "Time elapsed: 22.25 min\n",
+ "Epoch: 078/200 | Batch 0000/0175 | Loss: 0.2101\n",
+ "Epoch: 078/200 | Batch 0100/0175 | Loss: 0.1619\n",
+ "Epoch: 078/200 | Train: 93.34% | Validation: 76.08%\n",
+ "Time elapsed: 22.54 min\n",
+ "Epoch: 079/200 | Batch 0000/0175 | Loss: 0.1868\n",
+ "Epoch: 079/200 | Batch 0100/0175 | Loss: 0.2370\n",
+ "Epoch: 079/200 | Train: 93.27% | Validation: 76.14%\n",
+ "Time elapsed: 22.82 min\n",
+ "Epoch: 080/200 | Batch 0000/0175 | Loss: 0.2559\n",
+ "Epoch: 080/200 | Batch 0100/0175 | Loss: 0.1775\n",
+ "Epoch: 080/200 | Train: 93.22% | Validation: 76.12%\n",
+ "Time elapsed: 23.12 min\n",
+ "Epoch: 081/200 | Batch 0000/0175 | Loss: 0.2177\n",
+ "Epoch: 081/200 | Batch 0100/0175 | Loss: 0.2389\n",
+ "Epoch: 081/200 | Train: 93.34% | Validation: 76.14%\n",
+ "Time elapsed: 23.40 min\n",
+ "Epoch: 081/200: LR updated to 1.0000000000000004e-05\n",
+ "Epoch: 082/200 | Batch 0000/0175 | Loss: 0.1632\n",
+ "Epoch: 082/200 | Batch 0100/0175 | Loss: 0.1998\n",
+ "Epoch: 082/200 | Train: 93.34% | Validation: 76.14%\n",
+ "Time elapsed: 23.69 min\n",
+ "Epoch: 083/200 | Batch 0000/0175 | Loss: 0.2569\n",
+ "Epoch: 083/200 | Batch 0100/0175 | Loss: 0.1700\n",
+ "Epoch: 083/200 | Train: 93.24% | Validation: 76.12%\n",
+ "Time elapsed: 23.97 min\n",
+ "Epoch: 084/200 | Batch 0000/0175 | Loss: 0.2742\n",
+ "Epoch: 084/200 | Batch 0100/0175 | Loss: 0.2500\n",
+ "Epoch: 084/200 | Train: 93.31% | Validation: 76.14%\n",
+ "Time elapsed: 24.25 min\n",
+ "Epoch: 085/200 | Batch 0000/0175 | Loss: 0.1839\n",
+ "Epoch: 085/200 | Batch 0100/0175 | Loss: 0.1553\n",
+ "Epoch: 085/200 | Train: 93.35% | Validation: 76.16%\n",
+ "Time elapsed: 24.54 min\n",
+ "Epoch: 086/200 | Batch 0000/0175 | Loss: 0.1573\n",
+ "Epoch: 086/200 | Batch 0100/0175 | Loss: 0.1159\n",
+ "Epoch: 086/200 | Train: 93.40% | Validation: 76.14%\n",
+ "Time elapsed: 24.82 min\n",
+ "Epoch: 087/200 | Batch 0000/0175 | Loss: 0.1442\n",
+ "Epoch: 087/200 | Batch 0100/0175 | Loss: 0.2050\n",
+ "Epoch: 087/200 | Train: 93.23% | Validation: 76.12%\n",
+ "Time elapsed: 25.11 min\n",
+ "Epoch: 088/200 | Batch 0000/0175 | Loss: 0.2447\n",
+ "Epoch: 088/200 | Batch 0100/0175 | Loss: 0.2788\n",
+ "Epoch: 088/200 | Train: 93.41% | Validation: 76.12%\n",
+ "Time elapsed: 25.39 min\n",
+ "Epoch: 089/200 | Batch 0000/0175 | Loss: 0.2651\n",
+ "Epoch: 089/200 | Batch 0100/0175 | Loss: 0.1958\n",
+ "Epoch: 089/200 | Train: 93.34% | Validation: 76.12%\n",
+ "Time elapsed: 25.67 min\n",
+ "Epoch: 090/200 | Batch 0000/0175 | Loss: 0.2521\n",
+ "Epoch: 090/200 | Batch 0100/0175 | Loss: 0.3240\n",
+ "Epoch: 090/200 | Train: 93.40% | Validation: 76.10%\n",
+ "Time elapsed: 25.95 min\n",
+ "Epoch: 091/200 | Batch 0000/0175 | Loss: 0.2921\n",
+ "Epoch: 091/200 | Batch 0100/0175 | Loss: 0.1958\n",
+ "Epoch: 091/200 | Train: 93.42% | Validation: 76.10%\n",
+ "Time elapsed: 26.23 min\n",
+ "Epoch: 092/200 | Batch 0000/0175 | Loss: 0.2033\n",
+ "Epoch: 092/200 | Batch 0100/0175 | Loss: 0.1764\n",
+ "Epoch: 092/200 | Train: 93.41% | Validation: 76.08%\n",
+ "Time elapsed: 26.52 min\n",
+ "Epoch: 092/200: LR updated to 1.0000000000000004e-06\n",
+ "Epoch: 093/200 | Batch 0000/0175 | Loss: 0.2396\n",
+ "Epoch: 093/200 | Batch 0100/0175 | Loss: 0.1957\n",
+ "Epoch: 093/200 | Train: 93.42% | Validation: 76.08%\n",
+ "Time elapsed: 26.81 min\n",
+ "Epoch: 094/200 | Batch 0000/0175 | Loss: 0.2137\n",
+ "Epoch: 094/200 | Batch 0100/0175 | Loss: 0.2045\n",
+ "Epoch: 094/200 | Train: 93.39% | Validation: 76.08%\n",
+ "Time elapsed: 27.10 min\n",
+ "Epoch: 095/200 | Batch 0000/0175 | Loss: 0.3079\n",
+ "Epoch: 095/200 | Batch 0100/0175 | Loss: 0.2056\n",
+ "Epoch: 095/200 | Train: 93.44% | Validation: 76.08%\n",
+ "Time elapsed: 27.38 min\n",
+ "Epoch: 096/200 | Batch 0000/0175 | Loss: 0.2246\n",
+ "Epoch: 096/200 | Batch 0100/0175 | Loss: 0.1919\n",
+ "Epoch: 096/200 | Train: 93.28% | Validation: 76.08%\n",
+ "Time elapsed: 27.67 min\n",
+ "Epoch: 097/200 | Batch 0000/0175 | Loss: 0.2829\n",
+ "Epoch: 097/200 | Batch 0100/0175 | Loss: 0.2486\n",
+ "Epoch: 097/200 | Train: 93.34% | Validation: 76.08%\n",
+ "Time elapsed: 27.96 min\n",
+ "Epoch: 098/200 | Batch 0000/0175 | Loss: 0.2047\n",
+ "Epoch: 098/200 | Batch 0100/0175 | Loss: 0.1282\n",
+ "Epoch: 098/200 | Train: 93.37% | Validation: 76.08%\n",
+ "Time elapsed: 28.25 min\n",
+ "Epoch: 099/200 | Batch 0000/0175 | Loss: 0.1638\n",
+ "Epoch: 099/200 | Batch 0100/0175 | Loss: 0.3237\n",
+ "Epoch: 099/200 | Train: 93.20% | Validation: 76.08%\n",
+ "Time elapsed: 28.54 min\n",
+ "Epoch: 100/200 | Batch 0000/0175 | Loss: 0.2104\n",
+ "Epoch: 100/200 | Batch 0100/0175 | Loss: 0.2579\n",
+ "Epoch: 100/200 | Train: 93.45% | Validation: 76.08%\n",
+ "Time elapsed: 28.82 min\n",
+ "Epoch: 101/200 | Batch 0000/0175 | Loss: 0.2770\n",
+ "Epoch: 101/200 | Batch 0100/0175 | Loss: 0.2861\n",
+ "Epoch: 101/200 | Train: 93.29% | Validation: 76.08%\n",
+ "Time elapsed: 29.11 min\n",
+ "Epoch: 102/200 | Batch 0000/0175 | Loss: 0.1737\n",
+ "Epoch: 102/200 | Batch 0100/0175 | Loss: 0.2289\n",
+ "Epoch: 102/200 | Train: 93.39% | Validation: 76.08%\n",
+ "Time elapsed: 29.40 min\n",
+ "Epoch: 103/200 | Batch 0000/0175 | Loss: 0.2508\n",
+ "Epoch: 103/200 | Batch 0100/0175 | Loss: 0.2159\n",
+ "Epoch: 103/200 | Train: 93.37% | Validation: 76.08%\n",
+ "Time elapsed: 29.67 min\n",
+ "Epoch: 103/200: LR updated to 1.0000000000000005e-07\n",
+ "Epoch: 104/200 | Batch 0000/0175 | Loss: 0.1972\n",
+ "Epoch: 104/200 | Batch 0100/0175 | Loss: 0.1779\n",
+ "Epoch: 104/200 | Train: 93.31% | Validation: 76.08%\n",
+ "Time elapsed: 29.95 min\n",
+ "Epoch: 105/200 | Batch 0000/0175 | Loss: 0.2059\n",
+ "Epoch: 105/200 | Batch 0100/0175 | Loss: 0.1426\n",
+ "Epoch: 105/200 | Train: 93.30% | Validation: 76.08%\n",
+ "Time elapsed: 30.23 min\n",
+ "Epoch: 106/200 | Batch 0000/0175 | Loss: 0.1956\n",
+ "Epoch: 106/200 | Batch 0100/0175 | Loss: 0.1668\n",
+ "Epoch: 106/200 | Train: 93.30% | Validation: 76.08%\n",
+ "Time elapsed: 30.51 min\n",
+ "Epoch: 107/200 | Batch 0000/0175 | Loss: 0.2412\n",
+ "Epoch: 107/200 | Batch 0100/0175 | Loss: 0.1974\n",
+ "Epoch: 107/200 | Train: 93.19% | Validation: 76.08%\n",
+ "Time elapsed: 30.80 min\n",
+ "Epoch: 108/200 | Batch 0000/0175 | Loss: 0.1725\n",
+ "Epoch: 108/200 | Batch 0100/0175 | Loss: 0.2587\n",
+ "Epoch: 108/200 | Train: 93.26% | Validation: 76.08%\n",
+ "Time elapsed: 31.08 min\n",
+ "Epoch: 109/200 | Batch 0000/0175 | Loss: 0.2682\n",
+ "Epoch: 109/200 | Batch 0100/0175 | Loss: 0.1625\n",
+ "Epoch: 109/200 | Train: 93.41% | Validation: 76.08%\n",
+ "Time elapsed: 31.40 min\n",
+ "Epoch: 110/200 | Batch 0000/0175 | Loss: 0.2526\n",
+ "Epoch: 110/200 | Batch 0100/0175 | Loss: 0.2128\n",
+ "Epoch: 110/200 | Train: 93.21% | Validation: 76.08%\n",
+ "Time elapsed: 31.68 min\n",
+ "Epoch: 111/200 | Batch 0000/0175 | Loss: 0.1973\n",
+ "Epoch: 111/200 | Batch 0100/0175 | Loss: 0.2599\n",
+ "Epoch: 111/200 | Train: 93.26% | Validation: 76.08%\n",
+ "Time elapsed: 31.98 min\n",
+ "Epoch: 112/200 | Batch 0000/0175 | Loss: 0.2444\n",
+ "Epoch: 112/200 | Batch 0100/0175 | Loss: 0.1985\n",
+ "Epoch: 112/200 | Train: 93.38% | Validation: 76.08%\n",
+ "Time elapsed: 32.29 min\n",
+ "Epoch: 113/200 | Batch 0000/0175 | Loss: 0.1945\n",
+ "Epoch: 113/200 | Batch 0100/0175 | Loss: 0.2026\n",
+ "Epoch: 113/200 | Train: 93.39% | Validation: 76.08%\n",
+ "Time elapsed: 32.59 min\n",
+ "Epoch: 114/200 | Batch 0000/0175 | Loss: 0.2036\n",
+ "Epoch: 114/200 | Batch 0100/0175 | Loss: 0.2501\n",
+ "Epoch: 114/200 | Train: 93.21% | Validation: 76.08%\n",
+ "Time elapsed: 32.87 min\n",
+ "Epoch: 114/200: LR updated to 1.0000000000000005e-08\n",
+ "Epoch: 115/200 | Batch 0000/0175 | Loss: 0.1947\n",
+ "Epoch: 115/200 | Batch 0100/0175 | Loss: 0.2291\n",
+ "Epoch: 115/200 | Train: 93.17% | Validation: 76.08%\n",
+ "Time elapsed: 33.16 min\n",
+ "Epoch: 116/200 | Batch 0000/0175 | Loss: 0.1759\n",
+ "Epoch: 116/200 | Batch 0100/0175 | Loss: 0.2064\n",
+ "Epoch: 116/200 | Train: 93.44% | Validation: 76.08%\n",
+ "Time elapsed: 33.44 min\n",
+ "Epoch: 117/200 | Batch 0000/0175 | Loss: 0.1927\n",
+ "Epoch: 117/200 | Batch 0100/0175 | Loss: 0.2139\n",
+ "Epoch: 117/200 | Train: 93.37% | Validation: 76.08%\n",
+ "Time elapsed: 33.73 min\n",
+ "Epoch: 118/200 | Batch 0000/0175 | Loss: 0.1942\n",
+ "Epoch: 118/200 | Batch 0100/0175 | Loss: 0.2636\n",
+ "Epoch: 118/200 | Train: 93.26% | Validation: 76.08%\n",
+ "Time elapsed: 34.01 min\n",
+ "Epoch: 119/200 | Batch 0000/0175 | Loss: 0.2051\n",
+ "Epoch: 119/200 | Batch 0100/0175 | Loss: 0.1565\n",
+ "Epoch: 119/200 | Train: 93.39% | Validation: 76.08%\n",
+ "Time elapsed: 34.29 min\n",
+ "Epoch: 120/200 | Batch 0000/0175 | Loss: 0.2065\n",
+ "Epoch: 120/200 | Batch 0100/0175 | Loss: 0.2613\n",
+ "Epoch: 120/200 | Train: 93.34% | Validation: 76.08%\n",
+ "Time elapsed: 34.57 min\n",
+ "Epoch: 121/200 | Batch 0000/0175 | Loss: 0.1853\n",
+ "Epoch: 121/200 | Batch 0100/0175 | Loss: 0.2363\n",
+ "Epoch: 121/200 | Train: 93.31% | Validation: 76.08%\n",
+ "Time elapsed: 34.85 min\n",
+ "Epoch: 122/200 | Batch 0000/0175 | Loss: 0.1943\n",
+ "Epoch: 122/200 | Batch 0100/0175 | Loss: 0.1610\n",
+ "Epoch: 122/200 | Train: 93.43% | Validation: 76.08%\n",
+ "Time elapsed: 35.13 min\n",
+ "Epoch: 123/200 | Batch 0000/0175 | Loss: 0.1659\n",
+ "Epoch: 123/200 | Batch 0100/0175 | Loss: 0.1966\n",
+ "Epoch: 123/200 | Train: 93.23% | Validation: 76.08%\n",
+ "Time elapsed: 35.41 min\n",
+ "Epoch: 124/200 | Batch 0000/0175 | Loss: 0.1870\n",
+ "Epoch: 124/200 | Batch 0100/0175 | Loss: 0.2138\n",
+ "Epoch: 124/200 | Train: 93.31% | Validation: 76.08%\n",
+ "Time elapsed: 35.69 min\n",
+ "Epoch: 125/200 | Batch 0000/0175 | Loss: 0.2993\n",
+ "Epoch: 125/200 | Batch 0100/0175 | Loss: 0.2569\n",
+ "Epoch: 125/200 | Train: 93.20% | Validation: 76.08%\n",
+ "Time elapsed: 35.97 min\n",
+ "Epoch: 126/200 | Batch 0000/0175 | Loss: 0.2542\n",
+ "Epoch: 126/200 | Batch 0100/0175 | Loss: 0.2266\n",
+ "Epoch: 126/200 | Train: 93.39% | Validation: 76.08%\n",
+ "Time elapsed: 36.25 min\n",
+ "Epoch: 127/200 | Batch 0000/0175 | Loss: 0.1569\n",
+ "Epoch: 127/200 | Batch 0100/0175 | Loss: 0.2849\n",
+ "Epoch: 127/200 | Train: 93.23% | Validation: 76.08%\n",
+ "Time elapsed: 36.54 min\n",
+ "Epoch: 128/200 | Batch 0000/0175 | Loss: 0.2357\n",
+ "Epoch: 128/200 | Batch 0100/0175 | Loss: 0.1459\n",
+ "Epoch: 128/200 | Train: 93.20% | Validation: 76.08%\n",
+ "Time elapsed: 36.82 min\n",
+ "Epoch: 129/200 | Batch 0000/0175 | Loss: 0.2577\n",
+ "Epoch: 129/200 | Batch 0100/0175 | Loss: 0.2151\n",
+ "Epoch: 129/200 | Train: 93.26% | Validation: 76.08%\n",
+ "Time elapsed: 37.11 min\n",
+ "Epoch: 130/200 | Batch 0000/0175 | Loss: 0.1919\n",
+ "Epoch: 130/200 | Batch 0100/0175 | Loss: 0.2438\n",
+ "Epoch: 130/200 | Train: 93.27% | Validation: 76.08%\n",
+ "Time elapsed: 37.40 min\n",
+ "Epoch: 131/200 | Batch 0000/0175 | Loss: 0.2404\n",
+ "Epoch: 131/200 | Batch 0100/0175 | Loss: 0.1923\n",
+ "Epoch: 131/200 | Train: 93.37% | Validation: 76.08%\n",
+ "Time elapsed: 37.69 min\n",
+ "Epoch: 132/200 | Batch 0000/0175 | Loss: 0.1882\n",
+ "Epoch: 132/200 | Batch 0100/0175 | Loss: 0.2157\n",
+ "Epoch: 132/200 | Train: 93.31% | Validation: 76.08%\n",
+ "Time elapsed: 37.97 min\n",
+ "Epoch: 133/200 | Batch 0000/0175 | Loss: 0.2647\n",
+ "Epoch: 133/200 | Batch 0100/0175 | Loss: 0.2634\n",
+ "Epoch: 133/200 | Train: 93.28% | Validation: 76.08%\n",
+ "Time elapsed: 38.26 min\n",
+ "Epoch: 134/200 | Batch 0000/0175 | Loss: 0.1767\n",
+ "Epoch: 134/200 | Batch 0100/0175 | Loss: 0.1547\n",
+ "Epoch: 134/200 | Train: 93.31% | Validation: 76.08%\n",
+ "Time elapsed: 38.55 min\n",
+ "Epoch: 135/200 | Batch 0000/0175 | Loss: 0.2636\n",
+ "Epoch: 135/200 | Batch 0100/0175 | Loss: 0.2203\n",
+ "Epoch: 135/200 | Train: 93.21% | Validation: 76.08%\n",
+ "Time elapsed: 38.83 min\n",
+ "Epoch: 136/200 | Batch 0000/0175 | Loss: 0.2580\n",
+ "Epoch: 136/200 | Batch 0100/0175 | Loss: 0.2864\n",
+ "Epoch: 136/200 | Train: 93.24% | Validation: 76.08%\n",
+ "Time elapsed: 39.11 min\n",
+ "Epoch: 137/200 | Batch 0000/0175 | Loss: 0.2357\n",
+ "Epoch: 137/200 | Batch 0100/0175 | Loss: 0.2243\n",
+ "Epoch: 137/200 | Train: 93.16% | Validation: 76.08%\n",
+ "Time elapsed: 39.38 min\n",
+ "Epoch: 138/200 | Batch 0000/0175 | Loss: 0.1865\n",
+ "Epoch: 138/200 | Batch 0100/0175 | Loss: 0.2034\n",
+ "Epoch: 138/200 | Train: 93.29% | Validation: 76.08%\n",
+ "Time elapsed: 39.66 min\n",
+ "Epoch: 139/200 | Batch 0000/0175 | Loss: 0.2210\n",
+ "Epoch: 139/200 | Batch 0100/0175 | Loss: 0.2541\n",
+ "Epoch: 139/200 | Train: 93.16% | Validation: 76.08%\n",
+ "Time elapsed: 39.94 min\n",
+ "Epoch: 140/200 | Batch 0000/0175 | Loss: 0.1954\n",
+ "Epoch: 140/200 | Batch 0100/0175 | Loss: 0.2377\n",
+ "Epoch: 140/200 | Train: 93.38% | Validation: 76.08%\n",
+ "Time elapsed: 40.22 min\n",
+ "Epoch: 141/200 | Batch 0000/0175 | Loss: 0.2013\n",
+ "Epoch: 141/200 | Batch 0100/0175 | Loss: 0.2696\n",
+ "Epoch: 141/200 | Train: 93.23% | Validation: 76.08%\n",
+ "Time elapsed: 40.50 min\n",
+ "Epoch: 142/200 | Batch 0000/0175 | Loss: 0.1989\n",
+ "Epoch: 142/200 | Batch 0100/0175 | Loss: 0.1540\n",
+ "Epoch: 142/200 | Train: 93.23% | Validation: 76.08%\n",
+ "Time elapsed: 40.79 min\n",
+ "Epoch: 143/200 | Batch 0000/0175 | Loss: 0.1721\n",
+ "Epoch: 143/200 | Batch 0100/0175 | Loss: 0.2168\n",
+ "Epoch: 143/200 | Train: 93.37% | Validation: 76.08%\n",
+ "Time elapsed: 41.07 min\n",
+ "Epoch: 144/200 | Batch 0000/0175 | Loss: 0.2290\n",
+ "Epoch: 144/200 | Batch 0100/0175 | Loss: 0.2365\n",
+ "Epoch: 144/200 | Train: 93.23% | Validation: 76.08%\n",
+ "Time elapsed: 41.35 min\n",
+ "Epoch: 145/200 | Batch 0000/0175 | Loss: 0.1810\n",
+ "Epoch: 145/200 | Batch 0100/0175 | Loss: 0.1901\n",
+ "Epoch: 145/200 | Train: 93.45% | Validation: 76.08%\n",
+ "Time elapsed: 41.63 min\n",
+ "Epoch: 146/200 | Batch 0000/0175 | Loss: 0.2031\n",
+ "Epoch: 146/200 | Batch 0100/0175 | Loss: 0.2626\n",
+ "Epoch: 146/200 | Train: 93.30% | Validation: 76.08%\n",
+ "Time elapsed: 41.91 min\n",
+ "Epoch: 147/200 | Batch 0000/0175 | Loss: 0.2411\n",
+ "Epoch: 147/200 | Batch 0100/0175 | Loss: 0.2293\n",
+ "Epoch: 147/200 | Train: 93.42% | Validation: 76.08%\n",
+ "Time elapsed: 42.18 min\n",
+ "Epoch: 148/200 | Batch 0000/0175 | Loss: 0.2135\n",
+ "Epoch: 148/200 | Batch 0100/0175 | Loss: 0.2439\n",
+ "Epoch: 148/200 | Train: 93.32% | Validation: 76.08%\n",
+ "Time elapsed: 42.46 min\n",
+ "Epoch: 149/200 | Batch 0000/0175 | Loss: 0.1918\n",
+ "Epoch: 149/200 | Batch 0100/0175 | Loss: 0.2237\n",
+ "Epoch: 149/200 | Train: 93.33% | Validation: 76.08%\n",
+ "Time elapsed: 42.74 min\n",
+ "Epoch: 150/200 | Batch 0000/0175 | Loss: 0.1645\n",
+ "Epoch: 150/200 | Batch 0100/0175 | Loss: 0.2257\n",
+ "Epoch: 150/200 | Train: 93.37% | Validation: 76.08%\n",
+ "Time elapsed: 43.02 min\n",
+ "Epoch: 151/200 | Batch 0000/0175 | Loss: 0.2217\n",
+ "Epoch: 151/200 | Batch 0100/0175 | Loss: 0.1910\n",
+ "Epoch: 151/200 | Train: 93.24% | Validation: 76.08%\n",
+ "Time elapsed: 43.29 min\n",
+ "Epoch: 152/200 | Batch 0000/0175 | Loss: 0.2408\n",
+ "Epoch: 152/200 | Batch 0100/0175 | Loss: 0.2346\n",
+ "Epoch: 152/200 | Train: 93.32% | Validation: 76.08%\n",
+ "Time elapsed: 43.57 min\n",
+ "Epoch: 153/200 | Batch 0000/0175 | Loss: 0.2266\n",
+ "Epoch: 153/200 | Batch 0100/0175 | Loss: 0.2770\n",
+ "Epoch: 153/200 | Train: 93.29% | Validation: 76.08%\n",
+ "Time elapsed: 43.84 min\n",
+ "Epoch: 154/200 | Batch 0000/0175 | Loss: 0.2149\n",
+ "Epoch: 154/200 | Batch 0100/0175 | Loss: 0.2034\n",
+ "Epoch: 154/200 | Train: 93.41% | Validation: 76.08%\n",
+ "Time elapsed: 44.13 min\n",
+ "Epoch: 155/200 | Batch 0000/0175 | Loss: 0.1959\n",
+ "Epoch: 155/200 | Batch 0100/0175 | Loss: 0.1590\n",
+ "Epoch: 155/200 | Train: 93.38% | Validation: 76.08%\n",
+ "Time elapsed: 44.41 min\n",
+ "Epoch: 156/200 | Batch 0000/0175 | Loss: 0.1885\n",
+ "Epoch: 156/200 | Batch 0100/0175 | Loss: 0.1852\n",
+ "Epoch: 156/200 | Train: 93.21% | Validation: 76.08%\n",
+ "Time elapsed: 44.68 min\n",
+ "Epoch: 157/200 | Batch 0000/0175 | Loss: 0.1926\n",
+ "Epoch: 157/200 | Batch 0100/0175 | Loss: 0.1297\n",
+ "Epoch: 157/200 | Train: 93.47% | Validation: 76.08%\n",
+ "Time elapsed: 44.95 min\n",
+ "Epoch: 158/200 | Batch 0000/0175 | Loss: 0.2349\n",
+ "Epoch: 158/200 | Batch 0100/0175 | Loss: 0.1806\n",
+ "Epoch: 158/200 | Train: 93.45% | Validation: 76.08%\n",
+ "Time elapsed: 45.23 min\n",
+ "Epoch: 159/200 | Batch 0000/0175 | Loss: 0.1464\n",
+ "Epoch: 159/200 | Batch 0100/0175 | Loss: 0.2517\n",
+ "Epoch: 159/200 | Train: 93.31% | Validation: 76.08%\n",
+ "Time elapsed: 45.50 min\n",
+ "Epoch: 160/200 | Batch 0000/0175 | Loss: 0.1642\n",
+ "Epoch: 160/200 | Batch 0100/0175 | Loss: 0.2214\n",
+ "Epoch: 160/200 | Train: 93.38% | Validation: 76.08%\n",
+ "Time elapsed: 45.77 min\n",
+ "Epoch: 161/200 | Batch 0000/0175 | Loss: 0.1531\n",
+ "Epoch: 161/200 | Batch 0100/0175 | Loss: 0.2125\n",
+ "Epoch: 161/200 | Train: 93.30% | Validation: 76.08%\n",
+ "Time elapsed: 46.05 min\n",
+ "Epoch: 162/200 | Batch 0000/0175 | Loss: 0.1755\n",
+ "Epoch: 162/200 | Batch 0100/0175 | Loss: 0.2653\n",
+ "Epoch: 162/200 | Train: 93.26% | Validation: 76.08%\n",
+ "Time elapsed: 46.32 min\n",
+ "Epoch: 163/200 | Batch 0000/0175 | Loss: 0.2494\n",
+ "Epoch: 163/200 | Batch 0100/0175 | Loss: 0.2969\n",
+ "Epoch: 163/200 | Train: 93.49% | Validation: 76.08%\n",
+ "Time elapsed: 46.59 min\n",
+ "Epoch: 164/200 | Batch 0000/0175 | Loss: 0.1901\n",
+ "Epoch: 164/200 | Batch 0100/0175 | Loss: 0.2521\n",
+ "Epoch: 164/200 | Train: 93.20% | Validation: 76.08%\n",
+ "Time elapsed: 46.87 min\n",
+ "Epoch: 165/200 | Batch 0000/0175 | Loss: 0.2593\n",
+ "Epoch: 165/200 | Batch 0100/0175 | Loss: 0.1652\n",
+ "Epoch: 165/200 | Train: 93.29% | Validation: 76.08%\n",
+ "Time elapsed: 47.14 min\n",
+ "Epoch: 166/200 | Batch 0000/0175 | Loss: 0.2033\n",
+ "Epoch: 166/200 | Batch 0100/0175 | Loss: 0.2589\n",
+ "Epoch: 166/200 | Train: 93.32% | Validation: 76.08%\n",
+ "Time elapsed: 47.41 min\n",
+ "Epoch: 167/200 | Batch 0000/0175 | Loss: 0.2056\n",
+ "Epoch: 167/200 | Batch 0100/0175 | Loss: 0.1752\n",
+ "Epoch: 167/200 | Train: 93.30% | Validation: 76.08%\n",
+ "Time elapsed: 47.68 min\n",
+ "Epoch: 168/200 | Batch 0000/0175 | Loss: 0.1360\n",
+ "Epoch: 168/200 | Batch 0100/0175 | Loss: 0.1685\n",
+ "Epoch: 168/200 | Train: 93.52% | Validation: 76.08%\n",
+ "Time elapsed: 47.95 min\n",
+ "Epoch: 169/200 | Batch 0000/0175 | Loss: 0.2110\n",
+ "Epoch: 169/200 | Batch 0100/0175 | Loss: 0.1651\n",
+ "Epoch: 169/200 | Train: 93.31% | Validation: 76.08%\n",
+ "Time elapsed: 48.23 min\n",
+ "Epoch: 170/200 | Batch 0000/0175 | Loss: 0.2299\n",
+ "Epoch: 170/200 | Batch 0100/0175 | Loss: 0.2341\n",
+ "Epoch: 170/200 | Train: 93.23% | Validation: 76.08%\n",
+ "Time elapsed: 48.50 min\n",
+ "Epoch: 171/200 | Batch 0000/0175 | Loss: 0.1046\n",
+ "Epoch: 171/200 | Batch 0100/0175 | Loss: 0.2091\n",
+ "Epoch: 171/200 | Train: 93.11% | Validation: 76.08%\n",
+ "Time elapsed: 48.77 min\n",
+ "Epoch: 172/200 | Batch 0000/0175 | Loss: 0.1939\n",
+ "Epoch: 172/200 | Batch 0100/0175 | Loss: 0.2016\n",
+ "Epoch: 172/200 | Train: 93.21% | Validation: 76.08%\n",
+ "Time elapsed: 49.07 min\n",
+ "Epoch: 173/200 | Batch 0000/0175 | Loss: 0.1865\n",
+ "Epoch: 173/200 | Batch 0100/0175 | Loss: 0.1867\n",
+ "Epoch: 173/200 | Train: 93.48% | Validation: 76.08%\n",
+ "Time elapsed: 49.34 min\n",
+ "Epoch: 174/200 | Batch 0000/0175 | Loss: 0.1711\n",
+ "Epoch: 174/200 | Batch 0100/0175 | Loss: 0.2631\n",
+ "Epoch: 174/200 | Train: 93.32% | Validation: 76.08%\n",
+ "Time elapsed: 49.61 min\n",
+ "Epoch: 175/200 | Batch 0000/0175 | Loss: 0.1993\n",
+ "Epoch: 175/200 | Batch 0100/0175 | Loss: 0.2012\n",
+ "Epoch: 175/200 | Train: 93.35% | Validation: 76.08%\n",
+ "Time elapsed: 49.90 min\n",
+ "Epoch: 176/200 | Batch 0000/0175 | Loss: 0.1483\n",
+ "Epoch: 176/200 | Batch 0100/0175 | Loss: 0.2022\n",
+ "Epoch: 176/200 | Train: 93.30% | Validation: 76.08%\n",
+ "Time elapsed: 50.17 min\n",
+ "Epoch: 177/200 | Batch 0000/0175 | Loss: 0.1589\n",
+ "Epoch: 177/200 | Batch 0100/0175 | Loss: 0.2454\n",
+ "Epoch: 177/200 | Train: 93.18% | Validation: 76.08%\n",
+ "Time elapsed: 50.44 min\n",
+ "Epoch: 178/200 | Batch 0000/0175 | Loss: 0.1682\n",
+ "Epoch: 178/200 | Batch 0100/0175 | Loss: 0.2324\n",
+ "Epoch: 178/200 | Train: 93.41% | Validation: 76.08%\n",
+ "Time elapsed: 50.73 min\n",
+ "Epoch: 179/200 | Batch 0000/0175 | Loss: 0.2154\n",
+ "Epoch: 179/200 | Batch 0100/0175 | Loss: 0.1974\n",
+ "Epoch: 179/200 | Train: 93.46% | Validation: 76.08%\n",
+ "Time elapsed: 51.01 min\n",
+ "Epoch: 180/200 | Batch 0000/0175 | Loss: 0.2283\n",
+ "Epoch: 180/200 | Batch 0100/0175 | Loss: 0.2324\n",
+ "Epoch: 180/200 | Train: 93.23% | Validation: 76.08%\n",
+ "Time elapsed: 51.28 min\n",
+ "Epoch: 181/200 | Batch 0000/0175 | Loss: 0.2894\n",
+ "Epoch: 181/200 | Batch 0100/0175 | Loss: 0.2023\n",
+ "Epoch: 181/200 | Train: 93.47% | Validation: 76.08%\n",
+ "Time elapsed: 51.57 min\n",
+ "Epoch: 182/200 | Batch 0000/0175 | Loss: 0.1447\n",
+ "Epoch: 182/200 | Batch 0100/0175 | Loss: 0.1684\n",
+ "Epoch: 182/200 | Train: 93.25% | Validation: 76.08%\n",
+ "Time elapsed: 51.84 min\n",
+ "Epoch: 183/200 | Batch 0000/0175 | Loss: 0.1816\n",
+ "Epoch: 183/200 | Batch 0100/0175 | Loss: 0.1985\n",
+ "Epoch: 183/200 | Train: 93.31% | Validation: 76.08%\n",
+ "Time elapsed: 52.11 min\n",
+ "Epoch: 184/200 | Batch 0000/0175 | Loss: 0.1902\n",
+ "Epoch: 184/200 | Batch 0100/0175 | Loss: 0.1736\n",
+ "Epoch: 184/200 | Train: 93.35% | Validation: 76.08%\n",
+ "Time elapsed: 52.41 min\n",
+ "Epoch: 185/200 | Batch 0000/0175 | Loss: 0.1788\n",
+ "Epoch: 185/200 | Batch 0100/0175 | Loss: 0.1937\n",
+ "Epoch: 185/200 | Train: 93.35% | Validation: 76.08%\n",
+ "Time elapsed: 52.69 min\n",
+ "Epoch: 186/200 | Batch 0000/0175 | Loss: 0.1830\n",
+ "Epoch: 186/200 | Batch 0100/0175 | Loss: 0.1975\n",
+ "Epoch: 186/200 | Train: 93.39% | Validation: 76.08%\n",
+ "Time elapsed: 52.96 min\n",
+ "Epoch: 187/200 | Batch 0000/0175 | Loss: 0.2079\n",
+ "Epoch: 187/200 | Batch 0100/0175 | Loss: 0.2169\n",
+ "Epoch: 187/200 | Train: 93.38% | Validation: 76.08%\n",
+ "Time elapsed: 53.25 min\n",
+ "Epoch: 188/200 | Batch 0000/0175 | Loss: 0.2875\n",
+ "Epoch: 188/200 | Batch 0100/0175 | Loss: 0.1724\n",
+ "Epoch: 188/200 | Train: 93.36% | Validation: 76.08%\n",
+ "Time elapsed: 53.53 min\n",
+ "Epoch: 189/200 | Batch 0000/0175 | Loss: 0.2111\n",
+ "Epoch: 189/200 | Batch 0100/0175 | Loss: 0.2205\n",
+ "Epoch: 189/200 | Train: 93.35% | Validation: 76.08%\n",
+ "Time elapsed: 53.82 min\n",
+ "Epoch: 190/200 | Batch 0000/0175 | Loss: 0.2656\n",
+ "Epoch: 190/200 | Batch 0100/0175 | Loss: 0.2060\n",
+ "Epoch: 190/200 | Train: 93.33% | Validation: 76.08%\n",
+ "Time elapsed: 54.10 min\n",
+ "Epoch: 191/200 | Batch 0000/0175 | Loss: 0.1901\n",
+ "Epoch: 191/200 | Batch 0100/0175 | Loss: 0.2243\n",
+ "Epoch: 191/200 | Train: 93.41% | Validation: 76.08%\n",
+ "Time elapsed: 54.37 min\n",
+ "Epoch: 192/200 | Batch 0000/0175 | Loss: 0.1988\n",
+ "Epoch: 192/200 | Batch 0100/0175 | Loss: 0.1692\n",
+ "Epoch: 192/200 | Train: 93.32% | Validation: 76.08%\n",
+ "Time elapsed: 54.66 min\n",
+ "Epoch: 193/200 | Batch 0000/0175 | Loss: 0.1406\n",
+ "Epoch: 193/200 | Batch 0100/0175 | Loss: 0.1615\n",
+ "Epoch: 193/200 | Train: 93.38% | Validation: 76.08%\n",
+ "Time elapsed: 54.98 min\n",
+ "Epoch: 194/200 | Batch 0000/0175 | Loss: 0.2369\n",
+ "Epoch: 194/200 | Batch 0100/0175 | Loss: 0.1783\n",
+ "Epoch: 194/200 | Train: 93.40% | Validation: 76.08%\n",
+ "Time elapsed: 55.27 min\n",
+ "Epoch: 195/200 | Batch 0000/0175 | Loss: 0.1645\n",
+ "Epoch: 195/200 | Batch 0100/0175 | Loss: 0.1641\n",
+ "Epoch: 195/200 | Train: 93.24% | Validation: 76.08%\n",
+ "Time elapsed: 55.55 min\n",
+ "Epoch: 196/200 | Batch 0000/0175 | Loss: 0.1828\n",
+ "Epoch: 196/200 | Batch 0100/0175 | Loss: 0.2722\n",
+ "Epoch: 196/200 | Train: 93.36% | Validation: 76.08%\n",
+ "Time elapsed: 55.85 min\n",
+ "Epoch: 197/200 | Batch 0000/0175 | Loss: 0.2249\n",
+ "Epoch: 197/200 | Batch 0100/0175 | Loss: 0.2003\n",
+ "Epoch: 197/200 | Train: 93.47% | Validation: 76.08%\n",
+ "Time elapsed: 56.16 min\n",
+ "Epoch: 198/200 | Batch 0000/0175 | Loss: 0.1724\n",
+ "Epoch: 198/200 | Batch 0100/0175 | Loss: 0.2729\n",
+ "Epoch: 198/200 | Train: 93.40% | Validation: 76.08%\n",
+ "Time elapsed: 56.50 min\n",
+ "Epoch: 199/200 | Batch 0000/0175 | Loss: 0.2268\n",
+ "Epoch: 199/200 | Batch 0100/0175 | Loss: 0.2668\n",
+ "Epoch: 199/200 | Train: 93.39% | Validation: 76.08%\n",
+ "Time elapsed: 56.87 min\n",
+ "Epoch: 200/200 | Batch 0000/0175 | Loss: 0.2366\n",
+ "Epoch: 200/200 | Batch 0100/0175 | Loss: 0.2278\n",
+ "Epoch: 200/200 | Train: 93.43% | Validation: 76.08%\n",
+ "Time elapsed: 57.27 min\n",
+ "Total Training Time: 57.27 min\n",
+ "Test accuracy 75.05%\n"
+ ]
+ }
+ ],
+ "execution_count": 7
+ },
+ {
+ "metadata": {
+ "ExecuteTime": {
+ "end_time": "2025-01-14T14:44:39.950051Z",
+ "start_time": "2025-01-14T14:44:39.680975Z"
+ }
+ },
+ "cell_type": "code",
+ "source": [
+ "from matplotlib import pyplot as plt\n",
+ "\n",
+ "fig, ax = plt.subplots(1, 2, figsize=(16, 6))\n",
+ " \n",
+ "# First subplot: Losses\n",
+ "x_loss = np.arange(0, len(loss_history))\n",
+ "loss_x_tick_positions = np.arange(0, len(loss_history)+1, 350*10)\n",
+ "loss_x_tick_labels = np.arange(0, 201, 20)\n",
+ "ax[0].plot(x_loss, loss_history, color='black', linewidth=1)\n",
+ "ax[0].set_ylabel('Training Loss', fontsize=22, fontweight='bold')\n",
+ "ax[0].set_xlabel('Epoch', fontsize=22, fontweight='bold')\n",
+ "ax[0].set_xticks(loss_x_tick_positions)\n",
+ "ax[0].set_xticklabels(loss_x_tick_labels)\n",
+ "ax[0].tick_params(axis='both', labelsize=16)\n",
+ "ax[0].grid(True, linestyle='--', alpha=0.6) \n",
+ "\n",
+ "# Second subplot: Accuracies\n",
+ "float_train_accuracy = [t.item() for t in train_acc_history]\n",
+ "float_valid_accuracy = [t.item() for t in valid_acc_history]\n",
+ "ax[1].plot(float_train_accuracy, color='black', linewidth=2, label='Training') \n",
+ "ax[1].plot(float_valid_accuracy, color='teal', linewidth=2, label='Validation') \n",
+ "ax[1].set_ylabel('Accuracy [\\%]', fontsize=22, fontweight='bold')\n",
+ "ax[1].set_xlabel('Epoch', fontsize=22, fontweight='bold')\n",
+ "ax[1].legend(loc='lower right', fontsize=18, frameon=True, fancybox=True, shadow=True, borderpad=1.5, handlelength=3)\n",
+ "ax[1].tick_params(axis='both', labelsize=16)\n",
+ "ax[1].grid(True, linestyle='--', alpha=0.6)"
+ ],
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "<Figure size 1600x600 with 2 Axes>"
+ ],
+ "image/png": "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"
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "execution_count": 28
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3 (ipykernel)",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.6.8"
+ },
+ "toc": {
+ "nav_menu": {},
+ "number_sections": true,
+ "sideBar": true,
+ "skip_h1_title": false,
+ "title_cell": "Table of Contents",
+ "title_sidebar": "Contents",
+ "toc_cell": false,
+ "toc_position": {},
+ "toc_section_display": true,
+ "toc_window_display": false
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/5_dpnn/results/serial/train_acc.pt b/5_dpnn/results/serial/train_acc.pt
new file mode 100644
index 0000000000000000000000000000000000000000..eda87a7f6b5037234bfc658595622974ca04f95b
GIT binary patch
literal 26362
zcmbuH2YgjU_QxNffUyvYQbcK?DZ<NpFRvg)0t7=qVZ16$A%<wskTAobE4p@FQ9ukT
zc3CUBx;6w<1i_ALzv608*1GDhU6Hlz|2OaMuIFT~#{Zq?CucZgzQ14YdnYGl&dsEX
zR=dPvZQI7IM~7HpOsmQ{bu%lgs)o$2)Rlwhom5*^x<_mpd7Rbq7#SN-rR!By&ES~}
z^qkuH1v6FkaSP_u&Yr2O>*m+1nI|U-WQzd<rnG^_e7tAn3DvS?fE1L~)>rByg3SWi
zYDx>dPh9RYr8U;71+y!M#^tUREwpSsxgs`j;G(Lk`r6v+Dm|yZZhpbxYIe1%o;_wx
zl`fEN1`63$%iV-5sDW8c%bGx}Snf{6_Ap|524Xda*-OiILKZS6A=^`?gJC)nrjwR?
z3%L(tlCm>px)`P_VY+F#uaMmtlal*UW`D!<AWTm!dkNW_F(tAOWr_^bmoWXb>@Va1
z#-!yy${b*r0|_%o%Y%d*%$STELYcT>iV2g@GAU$=F<DtcnY3Xtgvn}|6LKhHa`Irx
z3^UAd!i>;zq>zXFk7<d*D2~ff6e=~)XabGVa;%VL94eOMD0HZS$_X@H%L*YUa3~=s
zQs^)P9ZsN0T22;Ha40E{ppa*vBMEermPZRYg+nPhl|s`DG@U>*v^+-0V>whJXHsaE
zfhq}foR(EW&gM{BR#WJB1D!yi8ZGAtc_N20@+1n?8fY$o>a?sEavq1WQc_46Xg-0o
zmJ5VDnL{~w3Wa`Wpi>F7P|MSVJe@&_xcohZ&M?rK1UgI0vxPi|L&fr33Y}-5^9gi;
zmVXfPLJlS5A_`q(po<B#Sj#0sF6B^CUP7T|2D+3$4O%W2@-hyk<mD8)!ayqsbfuP8
z33)Y#O5`;Zy4FC~5$Jj?ZxC`Nhtl$o6uQwsHxcM&EpHL>Rt{z4Z4_E%pxX&_hn9a5
z@=gwA<y{o|vw`j=&^=n-E989~%E|jFwAw%q5NM5-YlU3Lpk!P=NTG)e^e}<eYx#(f
z8#q)fH&W<P18pMEW-T8R@^KC&<P#Kn(m+oU=xHsV5poNMlJZ#!{l!4f5$Jg>Ul8&|
z4yEKv6nfb}uMo)B@>L<Xa;QYUMxnnN=yd|Up=G0x+c=b#Z&K(j18pbJ-?V&N$agrD
zk?&IIJp;W@pbxbCyO1ApC@Vjr(8mV)2Z8>n<tIXZ%AuV6j6$Cq=nDdUspVHfe$Aj%
zTz*5L9R~WAKs&Ykmyq9as91hap&tzNBY}R>^4~(m@O=yC5~>CE2&!5l2d!FR{aHZm
zf>f(r5rt}v$YqkM4fd!^TjWM&H>^aa0I62HBMP+#B9}?2J+Vh+_CjuC+F>O!g-Eq(
zk0?|JL@raJI%1E?bV6=q_QpzN_CcyuXGEd8Aaa?s>WV!o(+#<i*%vF3>5f#Z{SbxP
zACb#sR1fS?nV!gvOfRfNrZ-Zp`XCBbgve#GsxS7aOh4pCrax99GXSYp0}+Kf0Fle&
z)PdNeGJ}vCnS-zrnZZc48iFWP9FfVC#8ol&s7wO6kx61DGAX25l^_b0M&vTZDuX>L
zlSOW1a#)GXP^4NNj40GFL@tw1!?8zYMj$sbBe4>hLy&4U3Q?$1L@tw5qp?S2#vnH`
zW3dvMGNf9KLlo*zL@tw3<=CS#<B=Pg3amtC0#dCeA_{dFB9|#qhhvY*OhRsCCSxTs
z0;yI<APVIna+$O`5_?qUDC9=wXskrW{7|FSR79bsA#$0FnvOjxGXuGiIR-0{IToo_
zGZBTFg~(;HsuFut<~Za=rV1;OnT=GdYDA%qN8~a&bprONObv1)GY2b?IT5K=Cm{+|
zi^ycsaWxlvRHhEOk*UW@Wac5&N+JrS5V=gTnvXpyqmdh#1z3s9$w;+21yQKqA#$07
zIu(0VW+8GTa~f76b2?J3evc^B8Hijasm{b6l{pK!kvSVHkvRvcR_7uLbsi#@NvZR(
zM`bQRZe;#|mB?I(RI5dZLS2N&WlGe=*rPIwksFyMSc%M1q*`5qDAY1UE|XT5Vvou+
zAU870u@affkZN@~qEJ^La+!=;fjug7C2}Kk6;>j1HBzmvK@{p*L@tw6*I|#!T#wwy
z+<=wHtVF8S9}$JR5s}N})J@o<GB+bPGPhtQGPfet>NZ57Rv|K(OkCZLJt}htawGF6
ztVHHcq*~pDDAb=3xlFOT8+%md9^^*mUaUmsKBQXRk0{h?L@tw14`7eVtU+#M)?y_x
z>yT>oAfiwYA#$0ddKi0DW<7Ev^9WWVvjM4A8xe(i6p_oM)F$jvna#+J%wt%I%;QM4
zdIC|XClR?!iFyiqROV^qM&=o;L}m+8t)4{`>Mw{~Cas>s9+i0>xsiDRE0K8-sa7u`
z3iUD~m&vGCut#Nl<VNOItVCujQmtM?6zZ>tTqdht#~zh=1G$lD#7bngA=T<lM4{e7
z<T5$69eY&fZ^(_z+gORrJ4m&97g4D95SdIiuHMHUmH7a<k@-7TBJ&|qtv*5&>SIJM
zQ>^}hJu350<VNNbtVHHhq*{H3DAebOTqdEuz#f(P61kE23M-NM8mU&_APThuk;^32
zx7edHJCPfie_<su-yzlNdqkmrK;$wh^&|GE%umRT%)hY`nHYZWq*V(<p;{txnG)3s
zdsJo@<VI#!tVE_YQmxt`3e^^o%cRwA*rPH9$c@bISc%LYNVVD%QK-ETxlBg2!yc6>
zL~dl-V<j>jkZRQtQK(LcTqdjb#vYZ~2f2~yjFrfAL8?_(M4`GNa+#dk7kgBuJ8~nl
zA66o>KT@rFAPUtJk;&xZsu%XCOmE~yrVmykQ-oBjzKBBgL*z2Wsz3Iq%mCy@W*}A~
za{yAU4n!1c5F(dJsDrRaWd<WRGDENunK)9diV=lMAaa?cN@9=7q>vk#60AffjZ~`)
zqEJ~xE|XF@>`|Ga$c@awSc%Lqq*@I}6lw$_mnl&ru}5VNL2hJ5VI?x9NVOV`DAX84
zE|XScu}5XfkQ<qCSc%M`NVO_Q6ly#om&vFK>`|Es$c@ZItVHH8q*@)0DAXiGE|XQ0
zu}5VDawBsDRwCme)#^w@p^if2GC6fL_NdGh<VI#HRw6SEsaDewg_?m_J$O<@o3<@`
zckkY(Yb<8oCjH`(OUGDuu>U=BIqp<lU2l#!%1XPiGZ%9FGOc!w*0t5f;{4(7b~UFa
z=Ixg)3S+~Fg*nw35-%&w{-Qqh3j24%<6A}<_qi8TFT+7n%)I@Y`-rCQ#bu>KBfC$V
z*0`g4sqt^K{Kp>6`?rq`xBPMWx#%B2`O8=2{a-D*c1JXSJOK{=eY>v6`!6k8Vh-|}
zdH%z$KjGwG{zh>9a~E8)Gn#+W$v?L9>iqRj+4$L_X#OcD|G8Dy<^3m3zWJwU{v}TS
zeK*~Z_up&F^M^+BPdoWHUJ%sZYY&~aJeq&T&Hu3L^Vfg;le@kd%|Gkp-?0ALy#FVi
z>NiI7&pG+Wx&;1bJbm=Z&HER|tp1Bnf0zEd_PL|;*WYT~h9jHz59vP%xBeU38C?Hc
zT^`sT&EKv6mM^?2fBl<N^|wazck90movz9If4OvQ|7iYh{daDk;Pby*>Lne`->v_~
zW?i1Y{_W#W`7xTmTmNlXye#i8H{6?v=I_>j>n>iN_a7jtUXAAO)_-FwZ_E3Cdf}?G
zqxn1b-#hQf`+wN)$OEGJyY=6O?sw(=M;<nQW%K?a{TH8pF8#MLeSF^k;LFaL-n@TE
z|4lmf-@?1|{yTmcvN)Q*TmOxHcu(Fx_Tr7-Me}#-zialsFYiBY^raJ{`MdSs*v|X&
z{`dF1{kdrVj{W!1)p`FJEtjo|=I_>jw{3kO@Bh=sH$D-~->v^Pp0Osl{w2wqqxrk_
z--h*T^Zw_3)Z>|G{%-v@wr*YCf1Q5n*l7N4{WrGi!My*X55D^i{V$~dqWqou?~^+3
zpIvd^6_Kt#<<@^2yodAt%lc2>9L?XY|CV1A_$OL)T^-Het^YQ3TA#oE6~oSo_WfVV
zt^dw#^+?|T<qhBViFW;N{kMEe;I9vSy&;;vTmOyC+mOHhCp-085zXJN|CUeNnD=k}
z#Sh0s^LOjNu~mV8yHP9mjOOpwf7d?pX#V<_CJ!GO&EK*AUK#k`x~cXzeE%QPe^Gue
z{kQ0u)AQH=!;Hkmk*>eQt^cm+yD9Jg&f*W}M)P;;zjfyZ{$;~{JR+LETmN1AT;PAh
ziMOa|{%-xZan0uZ^MBNiZS$h}yY=6?#=w6`;hEi{`MdSshN8#v*FR$Y1R2fWt^YQT
z2>gdEdHaHB{%-xZXt&4n*T2W7XQ!k2yY=7l@qvHEm+hBC^LOjN<qHG<#?A|GZ{9zo
z|KjC$>A#H|F3dmwm9eSOe*c$t>%VKy53c{l(&^FaKke3kWA_C9-*r6pmFA!Skl(+g
z-TLp`uLA!^+C5(r&EKv6mTwOH=da5g7tP<T|2A$7{6D{Xa$PikxBeSj7x<T!PK<W`
zly>XCMJ=AlfBt>5bWyKp*YDPU%ew{s4}G#G+WBwVt^aNt9Qe<=cl_jN*YDPUV}*f#
z<C()dHt!$OfARFY^xth|i}SDl>g^@bzW!z0`fuOeg6ki!u3xmD|1xg<ckaG{|9cyY
zwl@F#hxFf!TmP*a9r(Ab9}(^RFXPsK=k6Z(U)H_lmT1@S)_;qp2L3;O(QkA#f4BbI
zFf8z&_femcX#Q^fckP0}KN&kOTK~_u_21aM!2kQzAOD8me}>h+TmP*~2L7VW3tvWi
z{@wcT@(F?e*bnMnZ{9zo|Kjpp`fualrTPBbxkq8N`Gc%m|E;SCuK%GC2Sq!7$-4F5
zy6J)cmg{zp_WR$gTmNks6Zl{G;->oMU;mK$&${*B@@awp!dZhZjOOpwf8DR&t^c~e
z{*L|k{NVY2cm5&Ke*Ve2_1|qH0{_AewG*Sg{%-xZs50<>b?mWoqWQb^-=gxse~F0A
zY~DYl|H9s-|MtB-nE&a%?)|Tu_Ye8}&$;#A*p%S<Z@Ou9wBP^c-1={0S>Qja<J60y
zUB6rZjnxGH$+GR`(fr-|Z!8=5pIcMaH=4g&|1DR6fA3S&hG_n7{kLIp;6JPL?>~>`
z@790IO9TJ=8oFK`&EKv6mX8no3mz-DESkSt|8@TUJLlMc9s2*t)oMQeke1W@ak9YN
zeeu%uCzu;&%uRB?a`#8iSe)E_F*two$^2?Hr?Pg=!b<Zey=KmtT~<2i|8+A)aWbAM
z&ZJ9{sdO%$$QEZ(*?2sch?iv2i1B2$xFnOv<TB=?+;sZ07Y)EiXx=7On0xtt{>x|d
z2AvM(f3U~TH|6~I@1Qm1>D@OS89N`h{QS)S@1~xn0`vx&LME`42?V$7G^Oe7FP!$e
zQTsey+?1v_ue4|0eh3|Jiqm^fIPUFhn_KSm))9{TA$X#xJH7LR<6d{$J=qkeH*Rp;
z55ZGSaeChc$G!b*_mZYKy={Wyeh8j!iqpFuIPUFlyW^997d^dcf#ZG%o^6WLdk#46
z^|0M@O>uh50Iyl!#;uo~4*WMU?`fGQteU0s`W*K|aI0qNxIM?cUbegS!P9AZj{70F
zRkL*1o#S3_+uf>JI;+lcKLodGmX4Zp-0NezTQy53#X0VW;8xAjL2r(GMYg+Dvvf|I
z<9-Nk)hr#e<~8f}wcM@OjSfjSF%O}wnx!+&9QXR!=2p$p5oM11A-Gkubi$b9UVq!&
zx^_As%yB;iw`!Ko_j26Bm-C;$zUIHcC9RsBLB?}A?uX!3&C=;wj(Y=bcdKUUa4g6D
z5ZtO+I@`)|?*QA~s#!Xk%4^oQaqEevgQZQ(546lvR?X76P>%Z{xK*=stdry3Alu!l
zSvr-;aX$pNYL*UFa@;$}cDJ57IuprpKLodGmX0)X+#77WTQ%E`OeAvL55cXPr2~Z=
z_lDT+R?X6RK#uz%xK*=soR8P67q{H4XO0fzH8BsNt(v8?bR74JZF8$;=_nk>{Se%$
zSvtwaaW7%JTQy4u(Kzmh;8xAjIWmrWN!#6e=IIz1$Ndo8s#!Y4#c?lXyIVC&hp;&A
zhu~Jt(itj_dnLBJRkL&iir1`f<JP6q0j4JAY0Erq)hwMq;<z7zTQy6^hdAzKY<H_>
z>GTiB{Se%$SvtJKaW8ATTQy5(Z#eFU;8xAj(HV|=IosWO;ppTG$Ndo8s#!X?!f|h?
z?QYjBnLFXQAA(yoOUFid&3Xr0?$*anhd!E^htO8d(wPm8d&6vVt7hrQ1;_mm+^ShR
zF~M<fxb1G$EFE~@xF3RBHB09eIPQ(G-L0CX;|3h}LvX8R>9hdHy^*%N^}5r;evbPg
zxK*?Cte)fEA-20!v-GH)*Q{^jmN`8rZ(=^mGS6ByOV7DE?uX!3&C+9Pj(erHyH&IF
zl$qmx2yWHvuegu4-L0Dam1_^dt(v7rwETnj#@Oyw&C(NAj{70FRkQSfl;hr5+ubS~
zdcMhVKLodGmL5;?n)S*ocZ;Tni%rZ!Xsc%F*&xTgakjZtv-D_><9-Nk)hs=k<G6RI
z?QYd9Jy_$oAA(yoOV7nP?v>l_R?X65EspykxK*?CREp!?c-!5o*<Wtceh6;W?5~u3
zh3#%t9X-_HAH2`-{~RMV{q-(AZNaZp$wN$_?&0S^lPX%b#*gUv$NsJG<MMwVHBHu)
mI2<vLozOH7aBa+b<?$xQVlD7CsbbgG=HE5=&%A%|zW)c;Z<-DO
literal 0
HcmV?d00001
diff --git a/5_dpnn/results/serial/valid_acc.pt b/5_dpnn/results/serial/valid_acc.pt
new file mode 100644
index 0000000000000000000000000000000000000000..afbce64d00dc3da0e07760354d0f801efbe48a21
GIT binary patch
literal 26362
zcmc(nd3aUT)y6L%gAtX<0Sy5`Bq+n32OtPh0}_X@Jt`s^LkNRJVJ|kQP+KRcbF{XK
zIMjh^wH9kzZQ6<h4z<q85v>+2MO$06jyTn}-@EVWQ`f%x+~nizKe#?wi?z+~eeXGU
z-Id|&lZN_kyToF>dc~|qzgVBxvZmGra~hkPo5#*+QcYtPFKn%=?H!xR9_MvEMaHU|
z)go%1Kc;c1TF|<rqLEr=FI~_&r%|<>xnvPFE=yL(F4fgDdctD~KC@|Fi|iU86?LtP
zn$&T@VukEBqYFMKE_a#H1G#zWoThPc*}cAtl0Bx^$3~94q`7%fYimohTCnKMB^Afg
zoEB=CbNGU0RUvzh6tb6+y9!w`A7=Br&JV;AayKs4TZ`=;h|Sl`9!l;hWFLb`%D$ZG
zr<wkY8KC4|LhfxaDY*}4_SMWl#_Xr${z6t7Oj-`&OqFH^GiHdA2M9UTU@~$TXAacN
zaK;>@<Om_F4JIo`a^_&o9Kx7UN{$wCjKSpOSkA;XlVD6z$&`?3gUQPbXR?~fF($8M
zLCA3iQ;_31GeI*ojG3t9p+ZjjAJY|$k%-I59IDmOVGKH4$s>fUGogf>!l9`eI+8)t
zl&lx>C=*J`qd9bphK^-WgObyQ6eg6C$8pHh(D4j9LCF(^oMA#~c@l?aYUpGJ%~J9d
zAx|}-jBMo4X&P!`&}=1}g`8tTS=qv&xf+_sp!rHJ5b|^r%E^TsYSqvg3_4TEMM5q%
zp}dqFA`LBJkWzA~kjqS{AkX5^*%~^BLCcj~A>_9VC>fXMa_HL{`VNE6Q}TQvFEF8m
zypTf|X=o*bRw?;iAul$eq`ZVfmul!T23@Y?6+&KVLMhqCp{q1>HG{sV<TXNm--Ocg
z2OPRqL)S5AwUXBhxyFPt@&*pwsG*w}bhDB_6!I1m%F0_gbeo25XV6+De<b7`CX|zR
za_Gkzx{E<~EBO;4?=hjg{3(alY3N=C{Y=UGg#5V)7342C^h*uh&!7jC{FRXH29%1+
zUvua;8u~4R9#nF@kPn$qLO#r)M>O;(gElDnn2?W~P*Og@p(i!;I|e<a<nM)i+JsW_
z84f+Gq30O%ypk^n`JxG><x3oTSwkBc<SY4#kef^>BVXmvYa03kgI-ti4I%$%LRq<)
zLvL#6Ee8Eb$t^<u*@SX(D~I0J&^rwJi;{m8@?8_k%fE5x?;6_1p!bw~U&w!$P(gmc
zp$|3m5raNf@)IFHHK24{Zs*Wv8v2|;Unu#dkYAZlLjIFOUu)=J4Enc{{}B|!_bn!u
zq%K%vD0M{*nz|u>BS5>LRMZ_&P!B{?CPh85#$|e;)-t;yGnopMigrU3)Em*1Nz?9F
z<1%}o)-rn{Gnqap74=0F)DO{=$xwf+ahU<Awai|~OlEJCiuOSiv@fD5lcj-J<1+i9
z)-wAeGnq=1iUuJHszNknax@rgTxJMrEpq@elNpLq(J(|o2O^p>c^Zy2E^`oSEi(d{
z$yB3MG!jwJ!HA|zfeyhMml=gx%Zx^5GGkCG8jC0>j%dha;*`J|mr0`5GAU#xlSZj1
zgD5DAXv!oghczyfN3CTF$V_G&N=4%l1x-LSWs+2bH7+v|wU#*)naNB-sc15ypjt#z
zCPjx~jmsR4TFV@P%w+0NDw={QXey#9lcpoF#$~3V)-v_TOy($*ijGDUbPS>?lc8g=
z#$_5%YnkcDOh%wobR41}57Cs#((zd1GAE$cGAAN48T~_zqLUB>%|tY1a&$7*xXdin
zTILjFCUYuEMU9ApPD3<h^3;SiE;AdomT5+2GILNWYC#k<7txd{&^)Yhnfa)-%mQR4
zb2>^z3lRmiA{sK;IGuqtE^{VoEwc!j$t*^xNFoX%L{la~OR&ae6lyKA6q(5^L#gO2
zL_udGnlec`2WwnrIchDl0-4Et3#FoS5e0o4(UeKicd*7~&O@zb&PQf47ob#hA)=s*
z5KWmht;8CaS%q55d>5I?T#QoDC5VD9MKonHbQ#vT%;l)H%oWH?=1P=`+7Ja@g=orT
z>1wQTneU<2GS?t8neU@i^aDge*CLuSIl2yOTxK<DEpt6GlUajO(G7@#ZbUR?@^ll{
zxXjI{wagEZnanLH72S#`=r%-Cra-r2jmxY>tz~|M%w+CBspw8bK|e+`WO8x33u|2F
zZq!=lC&*0Z9+ZlHiYRCuqA8Q0d$Go4eui4h+=t9$evVSnFAxR&648`N(*0QDG7q5E
zGQUD*GVLf8{TflwZxBtH6#W)!T;@U4T4p^mlX(cGqK6R$J%VV;r0G$tahVOMwajD4
zOy+Tvik?6e^dzDwlcC>XjmtcRTFd+%naMnjQqePrf}TY*WwP`f*0{{`sI|-s$V}!%
zl!{(L6!bEpDU+j(SmQE2YAy2$GLzYaQqil3f?h*3W%Bd~tZ|vwQEQnukeSRMQ7YPu
zDCkW@Q>H*~VU5fD3AL8lg3M(8j8f57L_u#O8Z!Afy@NF_^B2@w=C8<1=3SJE{)Q;%
z?}(;Mg0^9e%e;qL%e;@wWd4Ct(FcfvK14KSlJpVQxXj0>wah2TOy*ORinb#P`V7&O
zNzvz6<1$~M)-qorGnubYD*7j)psx{4nKb<iYh32vsI|<0keN&jzjsp91yN8}L{lb1
z-LS@Gc0sLWx+6209w-&{L=@Bu(Ui&3u2|zT6{xk$ZpcigH%djjBMRCB(Ui&2o>=2D
zeNbzezQ|0bA4*025d{rEG-dL%7uLAU-l(<AKFCaFUzCalA`03M(Ud9B{#fHOm8iAM
zAY>*}g;LRAL_tFk4Vglm4!|0h8H!rV3`1rz2clFo98u6gh^9<}MqrK0RHN21BaxZR
z!6+3Sf+%PdqA8Q4(OBa$V^C|EvB*p&j#5zqQBV@mlu1zvYg{IcTFYdRnM@X?q8y^2
zJfbO+rUKTu%sA9qW;`;JnSfGJ4Wgilh^9=24#gUmnS@%)Oh#rhwI~%GhA8N8L{lb9
zM_`T1)S=cgQ;?a=RFsO2L=-d)(Ui$iJ=VC)QK+@d(a22Z7?g^RMHJM4Xv*YiI@Y+1
zK&@qtLuN7_N=3&b3OWJNlqt}OSmQD?P-~f!keSR(l!{J96f_I5WlTeT&t6@JR#pxh
z7>ntThO(Dj`o}8SC70RfsFp?g5=UL_zQ#2dcKKyy>jI@(TeQW>nymx%RTKSD-l9)z
z0<+Lpb;ic)YV&3FxmVb~WzEZzwfp+^TU&7<DW*R<a-Uf0o~WxG7ukL0%uTmf)oTB4
zPi(rsbN{}v8p|JtUx@w$-kW}9(SP=bli!c#j}zeFKXuC$MgOfE$3Gg)Kk4MZ*Sf&J
z>G?GmM)OZO`43pvRy_Vy&-B?fnt$5KfBAr`i~jfS+4Oug|BRFWmgUzJ{cn4}(U0bz
zb@E@k__Ct^&=~_ikLI6q@|QKi>tFTs`mdt-=bijTd+`2$d+1vaMDs5=`Olpj_*Wl)
z{xO~V_la5a7jJ);`MY8J%;M|+)aFaB?c6_P{wCe#Z!8wP{s*qOqI)!dxA{A(?b71$
zKiNFBe>8u$`P*I<%-_2{dg9Ay{%-TPZ9{PULl<sqjOOn)f45g(Qat~QMkk`pzogsz
zT{mTA(SP&YZX=={zuWxXust~b0rfXdi01D$f7`3BEgpYr+$Rmu{N3j7iUF&O{ugc>
zd}lO&xA{A(?fRm>H}p>G+&^Ug;_c@$e;3!xE&9Lt&idy%_YaxBDYyB%*Sa-D|Ieol
zToujVZT_xZ9QY?!JhC>LzuWwkHGzM{s9q7zzuWws)fV_)CoerI+VQ*1->F*y|D`9k
zKOW8BZT?QIxS`m8nL8$BqxrkdUw41E`CIX7aQ?TPde|$`j^Azmy1)Kz^LNF7;Psz$
zUh?wJk3VGoqWxXwZ+q3!V*gdHxn+3g{vq=>?KXd7vEcaSJ;M%(=I=ItN6rfRui@^s
zmqzn<o4<4C2L9(>F!r=){%-U4vF(BX)-zssCz`+8{B5ra{107P_jNRXxA~h&1^xH=
zEmwUI&EIYQy7#}^{B`$to4+f!1+V|vZ}&Pn+WB{zzo}I4{qNEZcWmz5KV<%*{aohn
zMcd9Q&ff?3m{Zlcf5`mJxXoYZ?>{qc^LNel!Ry~PxM#m;$L}_Oo%=82Hh-PJ{$$+d
z@0R7k`QLWJP0@b;lyRHCpH&C*SA4wqp=hta+x!*n!SUa>U9OJi?>2v(zy4(0=I?vc
zgY&;^?@wQkcKmMh_qk^S|0}oOm+Ra=Wd7picbUI!8!jx)--R!Zys~rukolW+o4;)v
zg5yt4P4<rF?>2wk{oUs8+Qq@~FPZkn`e?`RI)4NIzBfF4O*DVE`77E3{|_HO{rqVD
zZu8gu{JYKHEz5)BzkA-Q%4o;$Hh*We1^$~RT|Ft9zuWwEKYq9QyJP(qGJkRUUFPqA
zb*qZ=_l3nTzZU8I=iKJ+jvar<{Li_~U+4bI?d18J+sX4c=Qe+x=U>ik{yM+^b8hq3
z`TM_|+x&I+cbmV1R|fNcWAe^uzyHs<&EJu;0{{Qt_a7ni7l-dMf45g(T%5mSDi4l!
z{U`4>e`QT@{PlzSN4x%=cbmV?{g-!}zwZ8S^LO2p;QV(zsrHjdUw`v%^Vj|O-R7_R
z@$cmMo8QUvH}5uo-Os<<{M|ACkogOHm-+iyb+G?w<&<}Ci}e00?Bw}daGSr*^S|IW
zf8G7v=C8ZI+x&I+cbmWN{%-Tv-QR8gy8FA$Uw41E`Rn4ptc8}~4{15=KTcKXyDwh4
zd7i#;M&BgY!QCH2VsUo&#hBvFCresr0shF?@+SQ!y&4zHsjD5eLpNh2Qt@0Om(8To
z*+M*-Pvp}1cp{t3<a4P^B9TdEvZ-_?m(A;EIrXCvM_1z+>W_wceJ|fPfBB5xpwmzP
z5949~9kf!O-+j~HV3**QpKtj8-PBVmz;B@GV+d?A1cF<3N@;%ki%EO?ar+{jD5d$$
zD}4=bKZK5#;{4td6ZiJF%`JC+>xhZ_A$YRXo!|Lk;$Efgo+`!pjT<KJhv4Z_oZok0
z;@%+JJyVMF+a^rh55cpgIKS(`#JwupJy(kJn-)ym55e&sGT$@&o&yv22HWn1Qk>s1
zVD_wU<JQa0H~g2F53$UXR?qVN`X=s&;8xG_?e-?_9bmg#PoD3VH*r4%w|bUub~kZv
zsO@g`EZ?hc;(iEj^(^0NZsOiB+uiC}zEj-9{Se&hS-#QR#JvM;cdKXlK5Y~CLvX8S
z`8I2_XT9N;yY;&9P0}UiA+*)Ae2=q<dk5L(R?qS+$|mlI;8xG_9mXc^jj-LVW9J)$
zP23N`t)AukdrjQKm-BDFz81g0rL3MUkNY9G)w6teZHfCx+uiEf4vswpw|bWEZ7p#>
z*mk#imTyfpd)Bvc>%{Yor6uNvSmtS~XZgNR6Zb=Kt7rMPP80V=+3r@)@?DuG?uX!3
z&+<)`Chm>4-K|r{_e7ewAA(yw+rgP1W4l{DTORjAaI0tehC=hndt+^Pt7rLsKoj>v
zaI0tec0RLby}0FWow@RZ(hs4np5=S#%;)YUY;&t;%bz(v1h;y&{5!@=+U{1*@{MTb
zllMb#t7rK>G86Yww!3xa`8F^U_d{^2XUo51ytM6Z^(^0nRpK6kTRqG7P?@-wvE8kn
z<y)Z4p7m|qI&{8)sl+^MnP;t@Esy&lxYe_K`%sB{&UUwYmS24{aX$pNdX{hAF>x<%
zyIVb5{>l3xxYe_K>r9Dz!FIP^IKK16#QhN5>e&w5$Jy>y&z3)Reh6;$EZ;U__N+JF
za<`s)2j(HP)wAW#oHxNX-+`X>L+~BwS+B-+w|cgNrVGKXp6%depJ=;VJ=?*g3&E|P
zEsuML+V0lt&M)?tw0sC|^=$cR;Z3sLt)AtV+RdKzZQL^F7s}1&?oGDL^H$H6KXZNv
zZuM+=+^e<St)AsqnawBfhu~JvcHn-P?QZpK2ks&G4)m;dxb40JJ?n?yR?n7y^4<})
zyVW-3aX$pNdX`_FG<(*ov)nDZe9!tJwAHiaac_!kZuM+=+z-L6o-L1iQ*C#vXUpS$
z2yXRkdE7hFcDH)AJno0!R?n8ly=k_))wAVsKLodWwmk0D+wNA^@rybo!^Xh>dl{+p
z*Sq{`3x1`_9^xqO9)2CDp}t2C{D@wBRCU9T%m01MFUhyy;)s5H0Zsh?*Hh=!b9Icx
Vy5OUszIzY-@A>#oe}3?}{|90}f0qCN
literal 0
HcmV?d00001
--
GitLab