what I did wrong here, the code run but it said not all test passed !!
Hi @Mubarak_Ibrahim ,
When adding the regulation term to the 3 hidden layers, the weights used for layer3 and layer1 are from the wrong layer. Remember, this is backward propagation, starting from the last hidden layer back to the first hidden layer. So for calculating dW3, W3 should be used, but in your case, W1 is used. Likewise, for the calculation of dW1.