carpedm20 / DiscoGAN-pytorch

PyTorch implementation of "Learning to Discover Cross-Domain Relations with Generative Adversarial Networks"

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

[!] Sampled dataset from A and B have different # of data. Try resampling...

EliasVansteenkiste opened this issue · comments

I get the following warning message;

[!] Sampled dataset from A and B have different # of data. Try resampling...

From what I understand it is not necessary that both data sets are equal in size?
Both datasets have pictures with 256x256 pixels.

I have same problem...

I don't remember it clearly, but it could be a problem of png. An image of png has 4 color channels but the current code may only support 3 color channel, I guess?

I encountered this issue as well, and the training is also interrupted after resampling.

The training seems to continue but that iteration gets skipped, any solution to the problem?

Declining the batch size < the max number of your dataset.