deep-learning-with-python-notebooks icon indicating copy to clipboard operation
deep-learning-with-python-notebooks copied to clipboard
verified publisher icon fchollet

6.4 Implementing a 1D convnet for the IMDB dataset

Open yurisolovev opened this issue 6 years ago • 1 comments

In the Dense layer, a sigmoidal activation function must be added, since we solve the classification problem, right?

yurisolovev avatar Nov 22 '19 12:11 yurisolovev

Yep, I believe this is a mistake in the book. Until I added the sigmoid activation the loss wasn't changing with the epochs.

oren0e avatar Jul 07 '20 13:07 oren0e