How do I get weights and biases from each iteration of my model?

Question

In my neural network, I need to get weights and biases from the model in each iteration. for example I run 5 epochs - the weights and biases that I get at the end are the weights and biases for the 5th epoch but I want to get the weights and biases of all epoch not only the last one. Here is my code that gives weights of last epoch:

model.layers[0].get_weights()[0]
model.layers[0].get_weights()[1]

Answer 1

You can loop over epochs and save the weights and biases of each epoch in a list:

weights = []
biases = []
for epoch in range(num_epochs):
   
    # training for each epoch
    model.fit(x_train, y_train, ...) 

    # add the weights and biases of the current epoch to the lists
    weights.append(model.layers[0].get_weights()[0])
    biases.append(model.layers[0].get_weights()[1])