ml for intrusion detection

code/src/machine_learning.py

+import matplotlib
+matplotlib.use('TkAgg')
+
+import time
+import numpy as np
+import pandas as pd
+import pyarrow as pa
+from math import *
+import matplotlib.pyplot as plt
+from PIL import Image
+import seaborn as sns
+import itertools
+import io
+import plotly.offline as py
+from IPython import get_ipython
+
+import plotly.graph_objs as go
+from plotly.subplots import make_subplots
+import plotly.figure_factory as ff
+import warnings
+
+from sklearn.preprocessing import StandardScaler
+from sklearn.preprocessing import LabelEncoder
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import MinMaxScaler
+from sklearn.naive_bayes import GaussianNB
+from sklearn.metrics import accuracy_score
+
+try:
+    get_ipython()
+    py.init_notebook_mode(connected=True)
+except NameError:
+    pass
+warnings.filterwarnings('ignore')
+
+
+Training = pd.read_parquet('intrusion_dataset.parquet')
+
+####################################################################################
+# Checking what is in the dataset
+# def data_overview(df, message):
+#     print(f'{message}:\n')
+#     print("Rows: ", df.shape[0])
+#     print("\nNumber of featerus: ", df.shape[1])
+#     print("\nFeatures: ", df.columns.to_list())
+#     print("\nMissing values: ", df.isnull().sum().values.sum())
+#     print("\nUnique values: ", df.nunique())
+#
+#
+# data_overview(Training, 'Intrusion Detection Dataset')
+####################################################################################
+
+####################################################################################
+# This code shall be used to find correlation and delete them
+# but im lazy to do it so if someone of you want to do it, go ahead
+# and do it
+#
+# Training = Training.dropna(axis="columns")
+# nTraining = Training[[col for col in Training.columns if Training[col].nunique() > 1 and
+#                      pd.api.types.is_numeric_dtype(Training[col])]]
+#
+# corr = nTraining.corr()
+# plt.figure(figsize=(15, 12))
+# sns.heatmap(corr)
+# plt.show()
+####################################################################################
+
+
+
+# Preparing of the data
+attack_cat = Training.get('attack_cat').drop_duplicates()
+
+Training = Training.drop(['label'], axis=1)
+
+y = Training[["attack_cat"]]
+x = Training.drop(["attack_cat", ], axis=1)
+
+non_numeric_columns = Training.select_dtypes(include=['object']).columns
+
+# Convert non-numeric columns to numeric using LabelEncoder
+label_encoders = {}
+for col in non_numeric_columns:
+    le = LabelEncoder()
+    Training[col] = le.fit_transform(Training[col])
+    label_encoders[col] = le
+
+sc = StandardScaler()
+x = Training.select_dtypes(include=['float64', 'int64'])
+
+sc = MinMaxScaler()
+x = sc.fit_transform(x)
+
+x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.33, random_state=42)
+print(x_train.shape, x_test.shape, y_train.shape, y_test.shape)
+
+# Naive Bayes
+clfg = GaussianNB()
+start = time.time()
+clfg.fit(x_train, y_train.values.ravel())
+end = time.time()
+print(f"Training time: {end - start}")
+
+start = time.time()
+y_pred = clfg.predict(x_test)
+end = time.time()
+print(f"Prediction time: {end - start}")
+accuracy = accuracy_score(y_test, y_pred)
+print(f"Accuracy: {accuracy}")