search

LynnHo / AttGAN-Tensorflow

AttGAN: Facial Attribute Editing by Only Changing What You Want (IEEE TIP 2019)
MIT License
605 stars 135 forks source link

What is the weight_decay value used? #38

Closed look4pritam closed 4 years ago

look4pritam commented 4 years ago

Default value for weight_decay is 0.0. What is the weight_decay value used for training the model?

look4pritam commented 4 years ago

What are the hyper parameters used for training the model?

LynnHo commented 4 years ago

@look4pritam It's just zero as you see, i.e., no weight_decay is applied to the model.

look4pritam commented 4 years ago

I was trying to reproduce your results using TensorFlow-2 version. Here generator loss is decreasing from 70. to 7. but discriminator loss is oscillating around 10.41. I am able to recreate input image with good quality, modified attributes are not getting applied. What may be the error?

look4pritam commented 4 years ago

You can find implementation at https://github.com/look4pritam/TensorFlowExamples/blob/master/GAN/AttGAN/AttGAN-Attributes.ipynb

LynnHo commented 4 years ago

@look4pritam I think you should print the losses separately, and find out which loss is abnormal so that you can check the corresponding part. image image

look4pritam commented 4 years ago

Thanks for prompt reply.

Are 60 epochs sufficient? Reconstructed image is good. Discriminator output is bad. I think more epochs are needed. But as suggested by you, I will check each losses to fix the problem.

Thanks and Regards

On Fri, May 22, 2020, 19:37 LynnHo notifications@github.com wrote:

@look4pritam https://github.com/look4pritam I think you should print the losses separately, and find out which loss is abnormal so that you can check the corresponding part. [image: image] https://user-images.githubusercontent.com/16457298/82676172-846d8500-9c78-11ea-9992-2aca4a6619c1.png [image: image] https://user-images.githubusercontent.com/16457298/82676187-8afbfc80-9c78-11ea-8cc1-13586650f37e.png

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/LynnHo/AttGAN-Tensorflow/issues/38#issuecomment-632710334, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABJOVYWRR3JDVZXTHOUICK3RS2BKJANCNFSM4NHYOWKQ .

LynnHo commented 4 years ago

@look4pritam 60 epochs should work. Actually, you can see the editing effect in about 10 epochs.

look4pritam commented 4 years ago

Thanks a lot, it will save time. I can train for epochs just more than 10 and see for attribute changing.

Before that, I wiil find out problematic loss part as per your suggestion.

Thanks a lot for your help.

On Fri, May 22, 2020, 20:00 LynnHo notifications@github.com wrote:

@look4pritam https://github.com/look4pritam 60 epochs should work. Actually, you can see the editing effect in about 10 epochs.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/LynnHo/AttGAN-Tensorflow/issues/38#issuecomment-632722312, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABJOVYQ5TWTSE6M54XUD6NTRS2D75ANCNFSM4NHYOWKQ .

look4pritam commented 4 years ago

Thank you very much for your reply. Now I am able to train the model.