Arian Tashakkor

Hi @SkafteNicki, sorry for the late response as I am just coming back from a vacation. I understand that most of the implementations in `torchmetrics` are meant to mirror their...

Hi, unfortunately I didn't manage to get it to work yet but essentially what I did was to manually append an `img_channels` to the `DATA` field of configuration that is...

Due to certain constraints on my training configuration, I was not able to evaluate the results using InceptionV3_TF and have fallen back on the Torch version which I'm assuming gives...

It might be possible that applying DRA can guide the results to our expectations as per the claims of the authors about how DRA can alleviate the stability and mode...

I just finished training both models on Tiny ImageNet and the differences are even more stark. Although plain ReACGAN collapsed after about 100k steps, it still achieved far better results...

Are you referring to `x_self_cond`? Because I can't find a label condition under `Unet1D`. If so, it appears according to [line 348 here](https://github.com/lucidrains/denoising-diffusion-pytorch/blob/main/denoising_diffusion_pytorch/denoising_diffusion_pytorch_1d.py#L348), if `x_self_cond` is not `None`, its shape...

Thanks for the effort you made to improve the implementation of CW in this library. I had one suggestion, and correct me if it is not feasible to implement, but...

@arxyzan I think this is what you meant I should be doing 😅. If there's anything further wrong with the PR, do let me know please.

@arxyzan My pleasure! Sure, I can try it on other models, I'll add them to `test_safetensors.py` and update the PR. I'm also psyched about seeing counters go up on HF...

@arxyzan I added a more comprehensive test suite for `safetensors` and tried to test it out on Colab since I don't have access to premium internet at the moment. It...