-
Thanks for your sharing! As it grid_sample is not mentioned in README, I wonder whether it works for openVINO custom layer?
-
https://github.com/pytorch/pytorch/blob/bdf360f9f233211b0f84bfb461700d32f69545f7/torch/nn/functional.py#L1549
is there should be:
ret = (y_hard - y_soft).detach() + y_soft ？
cc @albanD @mruberry
-
Hi, thanks for this great repo!
Quick question, should this line be `hard=False` instead? Otherwise, how does the gradient flow through?
https://github.com/shariqiqbal2810/maddpg-pytorch/blob/40…
-
> A token is any symbol from a discrete vocabulary; for humans, each English letter is a token from a 26-letter alphabet. DALL·E’s vocabulary has tokens for both text and image concepts. Specificall…
-
Hello!
I've been using `DeepSpeed sparse attention` pretty well, but I got this error since [tag/0.0.59](https://github.com/lucidrains/DALLE-pytorch/releases/tag/0.0.59) release.
```bash
Tracebac…
-
I have git cloned the repository and run `./setup.py install` and `./setup-dataset.sh`, but then I realized `train_all.sh` was not present. Later I found it in the 4.1.3 release. Do I need to set up o…
-
Can you explain what `recons` stands for? Please not use acronyms or shorthand without annotation. Thank you!
-
## 🐛 Bug
## To Reproduce
Steps to reproduce the behavior:
1. simply invoke the gumbel_softmax function many times in fp16 mode. Occainonly, it may encounter an nan problem.
I debug in…
-
## ❓ Questions and Help
I'm trying to install fairseq but i can't.
Whenever I execute this code `pip3 install fairseq`, I get this overwhelmingly long error.
Collecting fairseq
Using cac…
-
In function select2withP of AutoDL-Projects\lib\models\shape_searchs\SoftSelect.py, the implementation of Gumbel softmax is below:
gumbels = -torch.empty_like(logits).exponential_().log()
I don'…