shizhediao
shizhediao
what if the source embedding is 200000*300 while the target embedding is 199999*300, could it be trained correctly by this code? thanks!
when running train_discriminator.py ,occuring cudnn RNN backward can only be called in training mode
> Replace `x_gen_attr` with `x_gen_attr.detach()` in `train_discriminator.py`. i.e in `model.forward_encoder_embed` and `model.forward_discriminator_embed` It works, thanks very much!
@lovefc 请问数据集可以开源分享吗，谢谢
Is there an update on this issue? Thanks!
Hi Scott, Is there any progress on this?
> did you have some issues in mind? Actually no. I have fixed a lot of issues and I think the training process of the current repo is going well....
Hi, I implemented three TPU enabled PyTorch training repos for BigGAN-PyTorch, all of which are based on this repo. [BigGAN-PyTorch-TPU-Single](https://github.com/shizhediao/BigGAN-PyTorch-TPU-Single): Training BigGAN with a single TPU. [BigGAN-PyTorch-TPU-Parallel](https://github.com/shizhediao/BigGAN-PyTorch-TPU-Parallel): Parallel version (multiple-thread)...
After taking this fix, when doing multi-nodes multi-gpu training, the speed is so slow. Anyone could help?
> Sure @peteriz . I will try this over the weekend. Hi did you manage to solve this issue? Thanks!