Network layers
The input layer consists of the features that we are passing to our neural network. If we had, say, a 10 x 10 pixel image as our input, we would have 100 input units. Nothing is actually done in the input layer, but it is the connection between the input and hidden layers that is important.
Our input layer connections perform a linear transformation on the input vectors, and sends the results of that transformation to the hidden layer, through which the results are transformed through the activation function. Once we perform this computation, we pass the results onto the hidden layer. Hidden layers are where our activation functions live, and our network can have any number of them. Hidden layers are so called such because ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access