pytorch-image-models
https://github.com/huggingface/pytorch-image-models
Python
PyTorch image models, scripts, pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (ViT), MobileNet-V3/V2, RegNet, DPN, CSPNet, Swin Transformer, MaxViT, CoAtNet, ConvNeXt, and more
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Python not yet supported1 Subscribers
Add a CodeTriage badge to pytorch-image-models
Help out
- Issues
- Fix attention bias caches after dtype conversion
- Fix convit GPSA rel_indices staying on stale device/dtype after .to()
- Add MobileDet-CPU/DSP/EdgeTPU/GPU backbones
- fix(distributed): bind the device for `npu:<index>` in init_distributed_device_so()
- Refactor RoPE helpers, improve precision, and extend coordinate augme…
- init_distributed_device_so() never binds the device for non-CUDA accelerators (Ascend NPU): `--device npu:N` silently runs on NPU 0
- Fix apply_rot_embed*/RotaryEmbedding{Cat,MRope} silently upcasting fp16/bf16 to fp32
- multi-label classification
- Fix/model input config
- Support non-RGB pretrained input channel transfer
- Docs
- Python not yet supported