pytorch-image-models
pytorch-image-models copied to clipboard
huggingface
The largest collection of PyTorch image encoders / backbones. Including train, eval, inference, export scripts, and pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (ViT)...
**Is your feature request related to a problem? Please describe.** TorchGeo provides a number of model weights pre-trained on non-RGB imagery (e.g., Sentinel-2, 13 channels). Oftentimes, when dealing with time-series...
Here: https://github.com/huggingface/pytorch-image-models/blob/a49b020effef7d3e8d6788adb32ab3faddf574a8/timm/models/swin_transformer_v2.py#L380-L383 the shifted window attention is applying padding after shifting the values. https://github.com/huggingface/pytorch-image-models/blob/a49b020effef7d3e8d6788adb32ab3faddf574a8/timm/models/swin_transformer_v2.py#L292 assumes that the window has been rolled by the `shift_size`. However, the applied padding would offset...
Hi @rwightman this PR resolved #2364 , please check. ## Result test T2T-ViT model and weight on ImageNet val dataset | Model | Acc@1 | Acc@5 | FLOPs#G | MACs#G...
**Is your feature request related to a problem? Please describe.** I think we can refactor the ReadMe and make it a bit simple and decluttered. Firstly, we could add the...
Hello: I recently tried to run "inference.py" script on my custom dataset with custom labels (only 3). I trained the model EfficientNet and pushed it to the HuggingFace platform. Now,...
Implement support for providing the labels with a json file for the "Folder" dataset. Support for other file formats could also be added easily. I train on pseudo-labels and having...
**Describe the bug** These models return `feature_info.channels()` is not match - `ecaresnet50d_pruned` - `ecaresnet101d_pruned` - `efficientnet_b1_pruned` - `efficientnet_b2_pruned` - `efficientnet_b3_pruned` **To Reproduce** ```python import timm import torch model_list = [...
add [t2t_vit](https://github.com/yitu-opensource/T2T-ViT)
**Is your feature request related to a problem? Please describe.** RWKV and MAMBA are new families of models that are efficient. It would be great if TIMM lib have this...
Metadata
Owner
Metadata
The largest collection of PyTorch image encoders / backbones. Including train, eval, inference, export scripts, and pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (ViT)...