Hi all, I am trying to make an autoencoder for multi-channel input. For example, instead of an input image of shape (1,H,W) my input is of shape (15,H,W).
The autoencoders I found on the internet worked well for 1 channel but (very) bad for more than 1 (15 in my case). Did someone try something similar before? Many Thanks, Ioannis -- --- You received this message because you are subscribed to the Google Groups "theano-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
