DDE-1/Time-series-analysis-II.md

@@ -33,7 +33,7 @@ or, we can also have another activation function here as well:
 y_t = ReLU(W^{yh}_{16}h_t)
 ```
 
-Imagine that we have an input X of the shape (1,3,4) where the second dimension is the time. If we use our model for all time steps, we get the following:
+Let's look at a simpler case where the second activation function is linear. Imagine that we have an input X of the shape (1,3,4) where the second dimension is the time. If we use our model for all time steps, we get the following:
 
 ```math
 y_1 = W^{yh}_{16} tanh(W^{xh}_{64}X_1 + W^{hh}_{66}h_{0})
@@ -90,8 +90,14 @@ Recurrent cells can also be used to perform sequence to sequence analysis, also
 <img src="uploads/d229e9dd599633916fbaabdd06e61e76/rnn_3.png"  width="600">
 </div>
 
+**Many to one**
+
+We may also be interested in guessing one output (label) given a sequence of data. For example, we can pass the last 24 hours of the load demand and expect the model to predict the next expected demand. 
+
+<div align="center">
+<img src="uploads/1c0c6bbb21d5a1237a52b9369fc4e9bf/rnn_4.png"  width="600">
+</div>
 
-...
 
 ### Multilayered RNN Case study: Load demand forecasting