1

I'm trying to stack vgg16 and lstm layers. Any ideas how to it make it work?

Pss, LSTM requires time distributed over all other layers, which don't return sequences.

Thanks for considering my problem

Michael
  • 66
  • 6
  • 2
    You must first decide which dimension you want to be considered "time steps". There are plenty of possible combinations of how to do that, but it depends on what you want to achieve. What do you want your LSTMs to consider "a sequence"? Width? Height? Channels? A group of images (how many)? Each pixel?..... – Daniel Möller Nov 25 '17 at 16:16
  • https://stackoverflow.com/questions/47470385/use-kerastensorflow-to-build-a-conv2dlstm-model/47472272?noredirect=1#comment81908096_47472272 – Daniel Möller Nov 25 '17 at 16:41
  • @Daniel Thanks for your comments! I think it will help – Michael Nov 25 '17 at 20:29

0 Answers0