Let's see how these equations set up for layer 1 and layer 2. If the training example set, X is (x1, x2, x3) for the preceding network.
Let's see how the equation comes along for layer 1:
The activation function for layer 1 is as follows:
The input can also be expressed as follows:
For layer 2, the input will be as follows:
The activation function that's applied here is as follows:
Similarly, for layer 3, the input that's applied is as follows:
The activation function for the third layer is as follows:
Finally, here's the input for the last layer:
This is its activation:
Hence, the generalized forward propagation equation turns out to be as follows:
