I've tried to copy the dogs vs cats code on my way, but the DNN says always it's a cat. While it's training the accuracy goes up to 85%. I don't know if my hardware to weak (MacBook Air) or if i made a mistake. the network has 6 conv_2d layers and 2 fully_connected layers (with size 64 and 32). the data loading function is exactly the same like in the tutorial.
thanks for every answer
You must be logged in to post. Please login or register an account.
Probably a mistake is being made somewhere. If your network isn't big enough, that could also be a problem.
-Harrison 7 years ago
You must be logged in to post. Please login or register an account.