asarigun / transgan Goto Github PK

Maybe you can tell how to train it,and maybe you can train it with celeb

I am wondering why there is no parameter to control the probability of argumentation. Does this mean that the discriminator has not seen the real data (non-argumented image)?

Can you resume training at breakpoints?

How to train multiple Gpus? A direct use of nn.DataParallel(model) does not seem feasible.

Hello, have you finally trained to the fid of the final result of the paper? Is there a pre-trained model?

Is there any other way to get your pre-trained model?
Thanks.

I want to use the checkpoint saved to generate my own results from a testing dataset and use those images later to calculate my own evaluation metrics. Please help

Hello,

I saw your comment in VITA-Group's implementation of TransGAN and started looking at your implementation here.

Without modifying anything and attempting to run "python train.py" results in CUDA out of memory; I believe the GPU I'm using cannot handle the model size/training images that you've specified. I tried editing the batch size on lines 35 and 36 of train.py (--gener_batch_size, changing default from 64 to 32, etc.), but I get a RuntimeError of:

Output 0 of UnbindBackward is a view and is being modified inplace. This view is the output of a function that returns multiple views. Such fuctions do not allow the otutput views to be modified inplace. You should replace the inplace operation by an out-of-place one.

My two questions are:

1. How would you suggest modifying the training parameters to deal with GPU running out of memory? and,
2. Is there a better way to edit the batch size, and what else do I need to change in order for the code to not break when the batch size is changed?

Thanks!

Hello and thanks for the implementation.
I'm trying to train this model on a different datset, but to do so I need a custom fid_stats file for my dataset. How can I create it ?

FID score: nan
WARNING:root:NaN or Inf found in input tensor.
FID score: nan - best ID score: 10000.0 || @ epoch 3.

hello,I would like to humbly ask you what is the difference beetween TransGAN-main and TransGAN-master?can Trans-main reproduce similar results of the original paper? The results obtained by using CIFAR in TransGAN-main are quite different from those in the paper,and WGAN-EP loss concussion,so I want to ask you.

Thank you for your contribution. But in the training processing, FID score is Nan. I want to known whether it is appropriate. Should I make some chance to solve this problem?