Nicolas Hug

Hi @manavkulshrestha , uint8 tensors are assumed to be in `[0, 255]`. When converting uint8 1 to torch.long the operation is basically `1 / 256 * (torch.iinfo(torch.long).max + 1)` which...

`img.to(torch.long)` should do it.

Thanks for the proposal @dhruvbird . Unfortunately as @abhi-glitchhg correcly noted above, hosting the datasets ourselves isn't an option because of licensing / copyright issues. Some datasets just plainly don't...

Hi @skier233 , Thanks for opening this issue. We have recently started https://github.com/pytorch/torchcodec which is where we want to consolidate the video decoding capabilities of pytorch. The library is currently...

Hi @heth27 and thank you for the feature request. Torchvision doesn't really have a holistic support for 3D data in general, so I'm not sure procrustes alignement would be in...

> If this is better suited for e.g. torchmetrics ([lightning.ai/docs/torchmetrics/stable](https://lightning.ai/docs/torchmetrics/stable/)) this would also be good to know It might be in scope for torchmetrics, although note that this isn't owned...

Hi @sanghunpark , thanks for the report. What you're observing is the result of a compromise between backward compatibility, and extending the v2 transforms to support new use-cases. We cannot...

Thanks for the feature request @kazemSafari . Is there an example of a CUDA implementation of such a util already? For the moment I'm a bit concerned that the complexity...

Hi @trawler0 , thank you for feature request. Sure, I think this can be in scope of our augmentation strategies, even if ultimately the implementation will just be a `Compose`...

Thanks for the proposal @lqrhy3, happy to consider a PR. Sometimes, we have to use weird / wrong type check and annotation because of torchscript support. Not sure if that...