In the week 2 programming assignment, W1 and W2 are referenced in the update_parameters_with_gd function. So these numbers indicate the layer of the neural network. But i was under the impression that the parameters must be updated after each training example. Also how can we know the number of neurons in the hidden layer?
W1 and W2 are the weights of the first and second hidden layers, respectively. Parameters are updated after each iteration. And, check this line of code from the update_parameters_with_gd function. What does it say?
L = len(parameters) // 2 # number of layers in the neural networks
yes i saw that line. I’m just confused about the dimensions of the matrices. If you could help me with that it would solve even this issue of mine :
Just a run through of all the matrices involved from Input to the end would suffice. Thanks in advance.
I replied to your original post here.