yoshitomo-matsubara / torchdistill

A coding-free framework built on PyTorch for reproducible deep learning studies. 🏆25 knowledge distillation methods presented at CVPR, ICLR, ECCV, NeurIPS, ICCV, etc are implemented so far. 🎁 Trained models, training logs and configurations are available for ensuring the reproducibiliy and benchmark.
https://yoshitomo-matsubara.net/torchdistill/
MIT License
1.37k stars 132 forks source link

[BUG] Not supported to Nvidia 4090 #367

Closed allent4n closed 1 year ago

allent4n commented 1 year ago

Please use Discussions to ask questions.

Describe the bug When I install torchdistill==0.3.3 and torchvision==0.13.1, it did not support my cuda version (12.0). I assume the problem is the torch library is supporting my cuda version.

Environment (please complete the following information):

Additional context Is it possible if you can update some of the libraries so that those who are using the new GPU can also interact with your repo directly? Thank you.

yoshitomo-matsubara commented 1 year ago

Hi @allent4n

To use torchdistill with some recent GPUs right now, you have two options:

As you can see in commit logs after v0.3.3, the next release of torchdistill is almost ready, but the documentation is work in progress and I cannot complete it soon because it's a very busy season for submitting/reviewing papers.

If either the 1st or 2nd options does not work for you, feel free to reopen this issue and share more details.