Discriminator cond_dim if statement being bypassed about rgan

[corcra]

Oh wow, thank you so much! I think this may be the bug I have been failing to find for ages... I will update the code right away.

Also, in case you/anyone is worried: we didn't actually use this code for the conditional experiments in the paper, because it didn't work (for reasons I didn't know, until now, hopefully), we were using a different (messier, but seemingly less buggy) version of the code written by @cresteban.

[ajhuofa]

No worries, Happy to be able to help.
I'm not sure if its possible to use batch mean and conditional inputs at the same time, but if so, then X in the batch_means if statement would need to be changed to inputs as defined by the cond_dim if statement.

[dmortem]

@corcra
Hi, what are the differences on the codes between this version and the version for the conditional experiments in the paper?
Thank you!