Closed michael2221807 closed 3 years ago
Sorry about the previous issue. On Tensorboard, all I got are some masked images instead of the coloured images. I am not sure what I did wrong. Thanks for helping:) Something like this instead of the one you showed in your instruction.
Hello. Could you provide more details?
On Wed, Apr 14, 2021, 12:13 PM michael2221807 @.***> wrote:
Sorry about the previous issue. On Tensorboard, all I got are some masked images instead of the coloured images. I am not sure what I did wrong. Thanks for helping:) [image: 0_499_fake] https://user-images.githubusercontent.com/16907073/114657646-17952a80-9d23-11eb-8f74-cd1dfe5a5809.jpg Something like this instead of the one you showed in your instruction.
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/thaoshibe/BeautyGAN-PyTorch-reimplementation/issues/1#issuecomment-819232787, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFIJON7SCZXLPPEK2AKUM3TTIUP7LANCNFSM4222G23A .
I used the train.py provided to train and I see the Tensorboard output images to be all mask images. I wonder if anything I did wrong, bc I am new to the machine learning staff:) THX for replying!
Also, I used the test.py provided by the original author to generate output by the trained model and the outputs are also masked images.
Sorry for the bother. I have resolved the issue, this is due to the difference between Linux and Windows file paths. I changed the / to \ in the makeup data loader and now images file can successfully be read.
Hi, I used your code to train a new model, however, I don't know how should I use it to generate fake images. Thanks a lot for helping.