mirrors
/
pytorch-image-models
mirror of https://github.com/huggingface/pytorch-image-models.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity
pytorch-image-models/README.md

657 lines
56 KiB
Markdown
Raw Normal View History

Update README.md, documentation, add TOC, sponsors, and links to timmdocs and paperswithcode
2021-03-11 02:23:51 +08:00
# PyTorch Image Models
- [Sponsors](#sponsors)
- [What's New](#whats-new)
- [Introduction](#introduction)
- [Models](#models)
- [Features](#features)
- [Results](#results)
- [Getting Started (Documentation)](#getting-started-documentation)
- [Train, Validation, Inference Scripts](#train-validation-inference-scripts)
- [Awesome PyTorch Resources](#awesome-pytorch-resources)
- [Licenses](#licenses)
- [Citing](#citing)
## Sponsors
Update README.md
2022-07-09 03:21:23 +08:00
Thanks to the following for hardware support:
* TPU Research Cloud (TRC) (https://sites.research.google/trc/about/)
Update README.md, documentation, add TOC, sponsors, and links to timmdocs and paperswithcode
2021-03-11 02:23:51 +08:00
* Nvidia (https://www.nvidia.com/en-us/)
Update README.md
2022-07-09 03:21:23 +08:00
And a big thanks to all GitHub sponsors who helped with some of my costs before I joined Hugging Face.
Create README.md
2019-04-15 06:10:52 +08:00
Update README.md with latest changes
2019-11-30 09:45:35 +08:00
## What's New
Add timm ConvNeXt 'atto' weights, change test resolution for FB ConvNeXt 224x224 weights, add support for different dw kernel_size
2022-08-16 08:56:08 +08:00
Update README, prep for version 0.9.0 release
2023-05-12 06:22:50 +08:00
❗Updates after Oct 10, 2022 are available in version >= 0.9❗
* Many changes since the last 0.6.x stable releases. They were previewed in 0.8.x dev releases but not everyone transitioned.
* `timm.models.layers` moved to `timm.layers`:
* `from timm.models.layers import name` will still work via deprecation mapping (but please transition to `timm.layers`).
* `import timm.models.layers.module` or `from timm.models.layers.module import name` needs to be changed now.
* Builder, helper, non-model modules in `timm.models` have a `_` prefix added, ie `timm.models.helpers` -> `timm.models._helpers`, there are temporary deprecation mapping files but those will be removed.
* All models now support `architecture.pretrained_tag` naming (ex `resnet50.rsb_a1`).
* The pretrained_tag is the specific weight variant (different head) for the architecture.
Update README.md
2023-05-12 06:32:01 +08:00
* Using only `architecture` defaults to the first weights in the default_cfgs for that model architecture.
Update README, prep for version 0.9.0 release
2023-05-12 06:22:50 +08:00
* In adding pretrained tags, many model names that existed to differentiate were renamed to use the tag (ex: `vit_base_patch16_224_in21k` -> `vit_base_patch16_224.augreg_in21k`). There are deprecation mappings for these.
Update README.md
2023-05-12 06:28:42 +08:00
* A number of models had their checkpoints remaped to match architecture changes needed to better support `features_only=True`, there are `checkpoint_filter_fn` methods in any model module that was remapped. These can be passed to `timm.models.load_checkpoint(..., filter_fn=timm.models.swin_transformer_v2.checkpoint_filter_fn)` to remap your existing checkpoint.
Update README, prep for version 0.9.0 release
2023-05-12 06:22:50 +08:00
* The Hugging Face Hub (https://huggingface.co/timm) is now the primary source for `timm` weights. Model cards include link to papers, original source, license.
* Previous 0.6.x can be cloned from [0.6.x](https://github.com/rwightman/pytorch-image-models/tree/0.6.x) branch or installed via pip with version.
Update readme with resizing addition
2023-08-29 13:02:07 +08:00
### Aug 28, 2023
* Add dynamic img size support to models in `vision_transformer.py`, `vision_transformer_hybrid.py`, `deit.py`, and `eva.py` w/o breaking backward compat.
* Add `dynamic_img_size=True` to args at model creation time to allow changing the grid size (interpolate abs and/or ROPE pos embed each forward pass).
* Add `dynamic_img_pad=True` to allow image sizes that aren't divisible by patch size (pad bottom right to patch size each forward pass).
* Enabling either dynamic mode will break FX tracing unless PatchEmbed module added as leaf.
* Existing method of resizing position embedding by passing different `img_size` (interpolate pretrained embed weights once) on creation still works.
* Existing method of changing `patch_size` (resize pretrained patch_embed weights once) on creation still works.
* Example validation cmd `python validate.py /imagenet --model vit_base_patch16_224 --amp --amp-dtype bfloat16 --img-size 255 --crop-pct 1.0 --model-kwargs dynamic_img_size=True dyamic_img_pad=True`
Update README, add --reparm to onnx_export
2023-08-26 02:04:34 +08:00
### Aug 25, 2023
* Many new models since last release
* FastViT - https://arxiv.org/abs/2303.14189
* MobileOne - https://arxiv.org/abs/2206.04040
* InceptionNeXt - https://arxiv.org/abs/2303.16900
* RepGhostNet - https://arxiv.org/abs/2211.06088 (thanks https://github.com/ChengpengChen)
* GhostNetV2 - https://arxiv.org/abs/2211.12905 (thanks https://github.com/yehuitang)
* EfficientViT (MSRA) - https://arxiv.org/abs/2305.07027 (thanks https://github.com/seefun)
* EfficientViT (MIT) - https://arxiv.org/abs/2205.14756 (thanks https://github.com/seefun)
* Add `--reparam` arg to `benchmark.py`, `onnx_export.py`, and `validate.py` to trigger layer reparameterization / fusion for models with any one of `reparameterize()`, `switch_to_deploy()` or `fuse()`
* Including FastViT, MobileOne, RepGhostNet, EfficientViT (MSRA), RepViT, RepVGG, and LeViT
* Preparing 0.9.6 'back to school' release
Update README.md
2023-08-30 03:16:11 +08:00
### Aug 11, 2023
* Swin, MaxViT, CoAtNet, and BEiT models support resizing of image/window size on creation with adaptation of pretrained weights
* Example validation cmd to test w/ non-square resize `python validate.py /imagenet --model swin_base_patch4_window7_224.ms_in22k_ft_in1k --amp --amp-dtype bfloat16 --input-size 3 256 320 --model-kwargs window_size=8,10 img_size=256,320`
Version 0.9.5 prep for release. README update
2023-08-04 06:16:46 +08:00
### Aug 3, 2023
* Add GluonCV weights for HRNet w18_small and w18_small_v2. Converted by [SeeFun](https://github.com/seefun)
* Fix `selecsls*` model naming regression
* Patch and position embedding for ViT/EVA works for bfloat16/float16 weights on load (or activations for on-the-fly resize)
* v0.9.5 release prep
Update README, version 0.9.4dev0
2023-07-28 08:07:31 +08:00
### July 27, 2023
* Added timm trained `seresnextaa201d_32x8d.sw_in12k_ft_in1k_384` weights (and `.sw_in12k` pretrain) with 87.3% top-1 on ImageNet-1k, best ImageNet ResNet family model I'm aware of.
* RepViT model and weights (https://arxiv.org/abs/2307.09283) added by [wangao](https://github.com/jameslahm)
* I-JEPA ViT feature weights (no classifier) added by [SeeFun](https://github.com/seefun)
* SAM-ViT (segment anything) feature weights (no classifier) added by [SeeFun](https://github.com/seefun)
* Add support for alternative feat extraction methods and -ve indices to EfficientNet
* Add NAdamW optimizer
* Misc fixes
Update README, prep for version 0.9.0 release
2023-05-12 06:22:50 +08:00
### May 11, 2023
* `timm` 0.9 released, transition from 0.8.xdev releases
Update README.md
2023-04-22 08:58:26 +08:00
Version 0.8.23dev0 and README update
2023-05-11 05:41:22 +08:00
### May 10, 2023
Update README, prep for version 0.9.0 release
2023-05-12 06:22:50 +08:00
* Hugging Face Hub downloading is now default, 1132 models on https://huggingface.co/timm, 1163 weights in `timm`
Version 0.8.23dev0 and README update
2023-05-11 05:41:22 +08:00
* DINOv2 vit feature backbone weights added thanks to [Leng Yue](https://github.com/leng-yue)
* FB MAE vit feature backbone weights added
* OpenCLIP DataComp-XL L/14 feat backbone weights added
* MetaFormer (poolformer-v2, caformer, convformer, updated poolformer (v1)) w/ weights added by [Fredo Guan](https://github.com/fffffgggg54)
* Experimental `get_intermediate_layers` function on vit/deit models for grabbing hidden states (inspired by DINO impl). This is WIP and may change significantly... feedback welcome.
* Model creation throws error if `pretrained=True` and no weights exist (instead of continuing with random initialization)
* Fix regression with inception / nasnet TF sourced weights with 1001 classes in original classifiers
Update README, prep for version 0.9.0 release
2023-05-12 06:22:50 +08:00
* bitsandbytes (https://github.com/TimDettmers/bitsandbytes) optimizers added to factory, use `bnb` prefix, ie `bnbadam8bit`
Version 0.8.23dev0 and README update
2023-05-11 05:41:22 +08:00
* Misc cleanup and fixes
* Final testing before switching to a 0.9 and bringing `timm` out of pre-release state
Update README.md
2023-04-28 07:20:18 +08:00
### April 27, 2023
* 97% of `timm` models uploaded to HF Hub and almost all updated to support multi-weight pretrained configs
* Minor cleanup and refactoring of another batch of models as multi-weight added. More fused_attn (F.sdpa) and features_only support, and torchscript fixes.
Update README.md
2023-04-22 08:58:26 +08:00
### April 21, 2023
* Gradient accumulation support added to train script and tested (`--grad-accum-steps`), thanks [Taeksang Kim](https://github.com/voidbag)
* More weights on HF Hub (cspnet, cait, volo, xcit, tresnet, hardcorenas, densenet, dpn, vovnet, xception_aligned)
* Added `--head-init-scale` and `--head-init-bias` to train.py to scale classiifer head and set fixed bias for fine-tune
* Remove all InplaceABN (`inplace_abn`) use, replaced use in tresnet with standard BatchNorm (modified weights accordingly).
Update README.md
2023-04-13 00:24:35 +08:00
### April 12, 2023
* Add ONNX export script, validate script, helpers that I've had kicking around for along time. Tweak 'same' padding for better export w/ recent ONNX + pytorch.
* Refactor dropout args for vit and vit-like models, separate drop_rate into `drop_rate` (classifier dropout), `proj_drop_rate` (block mlp / out projections), `pos_drop_rate` (position embedding drop), `attn_drop_rate` (attention dropout). Also add patch dropout (FLIP) to vit and eva models.
* fused F.scaled_dot_product_attention support to more vit models, add env var (TIMM_FUSED_ATTN) to control, and config interface to enable/disable
* Add EVA-CLIP backbones w/ image tower weights, all the way up to 4B param 'enormous' model, and 336x336 OpenAI ViT mode that was missed.
Update README.md
2023-01-07 13:38:40 +08:00
Update README
2023-04-06 07:37:07 +08:00
### April 5, 2023
* ALL ResNet models pushed to Hugging Face Hub with multi-weight support
* All past `timm` trained weights added with recipe based tags to differentiate
* All ResNet strikes back A1/A2/A3 (seed 0) and R50 example B/C1/C2/D weights available
* Add torchvision v2 recipe weights to existing torchvision originals
* See comparison table in https://huggingface.co/timm/seresnextaa101d_32x8d.sw_in12k_ft_in1k_288#model-comparison
* New ImageNet-12k + ImageNet-1k fine-tunes available for a few anti-aliased ResNet models
* `resnetaa50d.sw_in12k_ft_in1k` - 81.7 @ 224, 82.6 @ 288
* `resnetaa101d.sw_in12k_ft_in1k` - 83.5 @ 224, 84.1 @ 288
* `seresnextaa101d_32x8d.sw_in12k_ft_in1k` - 86.0 @ 224, 86.5 @ 288
* `seresnextaa101d_32x8d.sw_in12k_ft_in1k_288` - 86.5 @ 288, 86.7 @ 320
Update README
2023-03-31 23:42:51 +08:00
### March 31, 2023
Update README
2023-04-01 07:51:02 +08:00
* Add first ConvNext-XXLarge CLIP -> IN-1k fine-tune and IN-12k intermediate fine-tunes for convnext-base/large CLIP models.
| model |top1 |top5 |img_size|param_count|gmacs |macts |
|----------------------------------------------------------------------------------------------------------------------|------|------|--------|-----------|------|------|
| [convnext_xxlarge.clip_laion2b_soup_ft_in1k](https://huggingface.co/timm/convnext_xxlarge.clip_laion2b_soup_ft_in1k) |88.612|98.704|256 |846.47 |198.09|124.45|
| convnext_large_mlp.clip_laion2b_soup_ft_in12k_in1k_384 |88.312|98.578|384 |200.13 |101.11|126.74|
| convnext_large_mlp.clip_laion2b_soup_ft_in12k_in1k_320 |87.968|98.47 |320 |200.13 |70.21 |88.02 |
| convnext_base.clip_laion2b_augreg_ft_in12k_in1k_384 |87.138|98.212|384 |88.59 |45.21 |84.49 |
| convnext_base.clip_laion2b_augreg_ft_in12k_in1k |86.344|97.97 |256 |88.59 |20.09 |37.55 |
Update README.md
2023-04-01 03:24:10 +08:00
* Add EVA-02 MIM pretrained and fine-tuned weights, push to HF hub and update model cards for all EVA models. First model over 90% top-1 (99% top-5)! Check out the original code & weights at https://github.com/baaivision/EVA for more details on their work blending MIM, CLIP w/ many model, dataset, and train recipe tweaks.
Update README
2023-03-31 23:42:51 +08:00
| model |top1 |top5 |param_count|img_size|
|----------------------------------------------------|------|------|-----------|--------|
| [eva02_large_patch14_448.mim_m38m_ft_in22k_in1k](https://huggingface.co/timm/eva02_large_patch14_448.mim_m38m_ft_in1k) |90.054|99.042|305.08 |448 |
| eva02_large_patch14_448.mim_in22k_ft_in22k_in1k |89.946|99.01 |305.08 |448 |
| eva_giant_patch14_560.m30m_ft_in22k_in1k |89.792|98.992|1014.45 |560 |
| eva02_large_patch14_448.mim_in22k_ft_in1k |89.626|98.954|305.08 |448 |
| eva02_large_patch14_448.mim_m38m_ft_in1k |89.57 |98.918|305.08 |448 |
| eva_giant_patch14_336.m30m_ft_in22k_in1k |89.56 |98.956|1013.01 |336 |
| eva_giant_patch14_336.clip_ft_in1k |89.466|98.82 |1013.01 |336 |
| eva_large_patch14_336.in22k_ft_in22k_in1k |89.214|98.854|304.53 |336 |
| eva_giant_patch14_224.clip_ft_in1k |88.882|98.678|1012.56 |224 |
| eva02_base_patch14_448.mim_in22k_ft_in22k_in1k |88.692|98.722|87.12 |448 |
| eva_large_patch14_336.in22k_ft_in1k |88.652|98.722|304.53 |336 |
| eva_large_patch14_196.in22k_ft_in22k_in1k |88.592|98.656|304.14 |196 |
| eva02_base_patch14_448.mim_in22k_ft_in1k |88.23 |98.564|87.12 |448 |
| eva_large_patch14_196.in22k_ft_in1k |87.934|98.504|304.14 |196 |
| eva02_small_patch14_336.mim_in22k_ft_in1k |85.74 |97.614|22.13 |336 |
| eva02_tiny_patch14_336.mim_in22k_ft_in1k |80.658|95.524|5.76 |336 |
* Multi-weight and HF hub for DeiT and MLP-Mixer based models
Update README
2023-03-23 06:40:10 +08:00
### March 22, 2023
* More weights pushed to HF hub along with multi-weight support, including: `regnet.py`, `rexnet.py`, `byobnet.py`, `resnetv2.py`, `swin_transformer.py`, `swin_transformer_v2.py`, `swin_transformer_v2_cr.py`
* Swin Transformer models support feature extraction (NCHW feat maps for `swinv2_cr_*`, and NHWC for all others) and spatial embedding outputs.
* FocalNet (from https://github.com/microsoft/FocalNet) models and weights added with significant refactoring, feature extraction, no fixed resolution / sizing constraint
* RegNet weights increased with HF hub push, SWAG, SEER, and torchvision v2 weights. SEER is pretty poor wrt to performance for model size, but possibly useful.
* More ImageNet-12k pretrained and 1k fine-tuned `timm` weights:
* `rexnetr_200.sw_in12k_ft_in1k` - 82.6 @ 224, 83.2 @ 288
* `rexnetr_300.sw_in12k_ft_in1k` - 84.0 @ 224, 84.5 @ 288
* `regnety_120.sw_in12k_ft_in1k` - 85.0 @ 224, 85.4 @ 288
* `regnety_160.lion_in12k_ft_in1k` - 85.6 @ 224, 86.0 @ 288
* `regnety_160.sw_in12k_ft_in1k` - 85.6 @ 224, 86.0 @ 288 (compare to SWAG PT + 1k FT this is same BUT much lower res, blows SEER FT away)
* Model name deprecation + remapping functionality added (a milestone for bringing 0.8.x out of pre-release). Mappings being added...
* Minor bug fixes and improvements.
Add convnext_xxlarge CLIP image tower weights, version 0.8.15dev0
2023-02-27 13:51:48 +08:00
### Feb 26, 2023
* Add ConvNeXt-XXLarge CLIP pretrained image tower weights for fine-tune & features (fine-tuning TBD) -- see [model card](https://huggingface.co/laion/CLIP-convnext_xxlarge-laion2B-s34B-b82K-augreg-soup)
* Update `convnext_xxlarge` default LayerNorm eps to 1e-5 (for CLIP weights, improved stability)
* 0.8.15dev0
Update README.md
2023-02-21 02:26:09 +08:00
### Feb 20, 2023
* Add 320x320 `convnext_large_mlp.clip_laion2b_ft_320` and `convnext_lage_mlp.clip_laion2b_ft_soup_320` CLIP image tower weights for features & fine-tune
* 0.8.13dev0 pypi release for latest changes w/ move to huggingface org
Update README
2023-02-17 09:20:27 +08:00
### Feb 16, 2023
* `safetensor` checkpoint support added
* Add ideas from 'Scaling Vision Transformers to 22 B. Params' (https://arxiv.org/abs/2302.05442) -- qk norm, RmsNorm, parallel block
* Add F.scaled_dot_product_attention support (PyTorch 2.0 only) to `vit_*`, `vit_relpos*`, `coatnet` / `maxxvit` (to start)
* Lion optimizer (w/ multi-tensor option) added (https://arxiv.org/abs/2302.06675)
* gradient checkpointing works with `features_only=True`
Update README and version
2023-02-07 15:43:14 +08:00
### Feb 7, 2023
* New inference benchmark numbers added in [results](results/) folder.
* Add convnext LAION CLIP trained weights and initial set of in1k fine-tunes
* `convnext_base.clip_laion2b_augreg_ft_in1k` - 86.2% @ 256x256
* `convnext_base.clip_laiona_augreg_ft_in1k_384` - 86.5% @ 384x384
* `convnext_large_mlp.clip_laion2b_augreg_ft_in1k` - 87.3% @ 256x256
Fix README
2023-02-07 15:46:26 +08:00
* `convnext_large_mlp.clip_laion2b_augreg_ft_in1k_384` - 87.9% @ 384x384
Update README and version
2023-02-07 15:43:14 +08:00
* Add DaViT models. Supports `features_only=True`. Adapted from https://github.com/dingmyu/davit by [Fredo](https://github.com/fffffgggg54).
* Use a common NormMlpClassifierHead across MaxViT, ConvNeXt, DaViT
* Add EfficientFormer-V2 model, update EfficientFormer, and refactor LeViT (closely related architectures). Weights on HF hub.
* New EfficientFormer-V2 arch, significant refactor from original at (https://github.com/snap-research/EfficientFormer). Supports `features_only=True`.
* Minor updates to EfficientFormer.
* Refactor LeViT models to stages, add `features_only=True` support to new `conv` variants, weight remap required.
Fix README
2023-02-07 15:46:26 +08:00
* Move ImageNet meta-data (synsets, indices) from `/results` to [`timm/data/_info`](timm/data/_info/).
Update README and version
2023-02-07 15:43:14 +08:00
* Add ImageNetInfo / DatasetInfo classes to provide labelling for various ImageNet classifier layouts in `timm`
* Update `inference.py` to use, try: `python inference.py /folder/to/images --model convnext_small.in12k --label-type detail --topk 5`
* Ready for 0.8.10 pypi pre-release (final testing).
Update README
2023-01-21 07:00:49 +08:00
### Jan 20, 2023
* Add two convnext 12k -> 1k fine-tunes at 384x384
* `convnext_tiny.in12k_ft_in1k_384` - 85.1 @ 384
* `convnext_small.in12k_ft_in1k_384` - 86.2 @ 384
* Push all MaxxViT weights to HF hub, and add new ImageNet-12k -> 1k fine-tunes for `rw` base MaxViT and CoAtNet 1/2 models
|model |top1 |top5 |samples / sec |Params (M) |GMAC |Act (M)|
|------------------------------------------------------------------------------------------------------------------------|----:|----:|--------------:|--------------:|-----:|------:|
|[maxvit_xlarge_tf_512.in21k_ft_in1k](https://huggingface.co/timm/maxvit_xlarge_tf_512.in21k_ft_in1k) |88.53|98.64| 21.76| 475.77|534.14|1413.22|
|[maxvit_xlarge_tf_384.in21k_ft_in1k](https://huggingface.co/timm/maxvit_xlarge_tf_384.in21k_ft_in1k) |88.32|98.54| 42.53| 475.32|292.78| 668.76|
|[maxvit_base_tf_512.in21k_ft_in1k](https://huggingface.co/timm/maxvit_base_tf_512.in21k_ft_in1k) |88.20|98.53| 50.87| 119.88|138.02| 703.99|
|[maxvit_large_tf_512.in21k_ft_in1k](https://huggingface.co/timm/maxvit_large_tf_512.in21k_ft_in1k) |88.04|98.40| 36.42| 212.33|244.75| 942.15|
|[maxvit_large_tf_384.in21k_ft_in1k](https://huggingface.co/timm/maxvit_large_tf_384.in21k_ft_in1k) |87.98|98.56| 71.75| 212.03|132.55| 445.84|
|[maxvit_base_tf_384.in21k_ft_in1k](https://huggingface.co/timm/maxvit_base_tf_384.in21k_ft_in1k) |87.92|98.54| 104.71| 119.65| 73.80| 332.90|
|[maxvit_rmlp_base_rw_384.sw_in12k_ft_in1k](https://huggingface.co/timm/maxvit_rmlp_base_rw_384.sw_in12k_ft_in1k) |87.81|98.37| 106.55| 116.14| 70.97| 318.95|
|[maxxvitv2_rmlp_base_rw_384.sw_in12k_ft_in1k](https://huggingface.co/timm/maxxvitv2_rmlp_base_rw_384.sw_in12k_ft_in1k) |87.47|98.37| 149.49| 116.09| 72.98| 213.74|
|[coatnet_rmlp_2_rw_384.sw_in12k_ft_in1k](https://huggingface.co/timm/coatnet_rmlp_2_rw_384.sw_in12k_ft_in1k) |87.39|98.31| 160.80| 73.88| 47.69| 209.43|
|[maxvit_rmlp_base_rw_224.sw_in12k_ft_in1k](https://huggingface.co/timm/maxvit_rmlp_base_rw_224.sw_in12k_ft_in1k) |86.89|98.02| 375.86| 116.14| 23.15| 92.64|
|[maxxvitv2_rmlp_base_rw_224.sw_in12k_ft_in1k](https://huggingface.co/timm/maxxvitv2_rmlp_base_rw_224.sw_in12k_ft_in1k) |86.64|98.02| 501.03| 116.09| 24.20| 62.77|
|[maxvit_base_tf_512.in1k](https://huggingface.co/timm/maxvit_base_tf_512.in1k) |86.60|97.92| 50.75| 119.88|138.02| 703.99|
|[coatnet_2_rw_224.sw_in12k_ft_in1k](https://huggingface.co/timm/coatnet_2_rw_224.sw_in12k_ft_in1k) |86.57|97.89| 631.88| 73.87| 15.09| 49.22|
|[maxvit_large_tf_512.in1k](https://huggingface.co/timm/maxvit_large_tf_512.in1k) |86.52|97.88| 36.04| 212.33|244.75| 942.15|
|[coatnet_rmlp_2_rw_224.sw_in12k_ft_in1k](https://huggingface.co/timm/coatnet_rmlp_2_rw_224.sw_in12k_ft_in1k) |86.49|97.90| 620.58| 73.88| 15.18| 54.78|
|[maxvit_base_tf_384.in1k](https://huggingface.co/timm/maxvit_base_tf_384.in1k) |86.29|97.80| 101.09| 119.65| 73.80| 332.90|
|[maxvit_large_tf_384.in1k](https://huggingface.co/timm/maxvit_large_tf_384.in1k) |86.23|97.69| 70.56| 212.03|132.55| 445.84|
|[maxvit_small_tf_512.in1k](https://huggingface.co/timm/maxvit_small_tf_512.in1k) |86.10|97.76| 88.63| 69.13| 67.26| 383.77|
|[maxvit_tiny_tf_512.in1k](https://huggingface.co/timm/maxvit_tiny_tf_512.in1k) |85.67|97.58| 144.25| 31.05| 33.49| 257.59|
|[maxvit_small_tf_384.in1k](https://huggingface.co/timm/maxvit_small_tf_384.in1k) |85.54|97.46| 188.35| 69.02| 35.87| 183.65|
|[maxvit_tiny_tf_384.in1k](https://huggingface.co/timm/maxvit_tiny_tf_384.in1k) |85.11|97.38| 293.46| 30.98| 17.53| 123.42|
|[maxvit_large_tf_224.in1k](https://huggingface.co/timm/maxvit_large_tf_224.in1k) |84.93|96.97| 247.71| 211.79| 43.68| 127.35|
|[coatnet_rmlp_1_rw2_224.sw_in12k_ft_in1k](https://huggingface.co/timm/coatnet_rmlp_1_rw2_224.sw_in12k_ft_in1k) |84.90|96.96| 1025.45| 41.72| 8.11| 40.13|
|[maxvit_base_tf_224.in1k](https://huggingface.co/timm/maxvit_base_tf_224.in1k) |84.85|96.99| 358.25| 119.47| 24.04| 95.01|
|[maxxvit_rmlp_small_rw_256.sw_in1k](https://huggingface.co/timm/maxxvit_rmlp_small_rw_256.sw_in1k) |84.63|97.06| 575.53| 66.01| 14.67| 58.38|
|[coatnet_rmlp_2_rw_224.sw_in1k](https://huggingface.co/timm/coatnet_rmlp_2_rw_224.sw_in1k) |84.61|96.74| 625.81| 73.88| 15.18| 54.78|
|[maxvit_rmlp_small_rw_224.sw_in1k](https://huggingface.co/timm/maxvit_rmlp_small_rw_224.sw_in1k) |84.49|96.76| 693.82| 64.90| 10.75| 49.30|
|[maxvit_small_tf_224.in1k](https://huggingface.co/timm/maxvit_small_tf_224.in1k) |84.43|96.83| 647.96| 68.93| 11.66| 53.17|
|[maxvit_rmlp_tiny_rw_256.sw_in1k](https://huggingface.co/timm/maxvit_rmlp_tiny_rw_256.sw_in1k) |84.23|96.78| 807.21| 29.15| 6.77| 46.92|
|[coatnet_1_rw_224.sw_in1k](https://huggingface.co/timm/coatnet_1_rw_224.sw_in1k) |83.62|96.38| 989.59| 41.72| 8.04| 34.60|
|[maxvit_tiny_rw_224.sw_in1k](https://huggingface.co/timm/maxvit_tiny_rw_224.sw_in1k) |83.50|96.50| 1100.53| 29.06| 5.11| 33.11|
|[maxvit_tiny_tf_224.in1k](https://huggingface.co/timm/maxvit_tiny_tf_224.in1k) |83.41|96.59| 1004.94| 30.92| 5.60| 35.78|
|[coatnet_rmlp_1_rw_224.sw_in1k](https://huggingface.co/timm/coatnet_rmlp_1_rw_224.sw_in1k) |83.36|96.45| 1093.03| 41.69| 7.85| 35.47|
|[maxxvitv2_nano_rw_256.sw_in1k](https://huggingface.co/timm/maxxvitv2_nano_rw_256.sw_in1k) |83.11|96.33| 1276.88| 23.70| 6.26| 23.05|
|[maxxvit_rmlp_nano_rw_256.sw_in1k](https://huggingface.co/timm/maxxvit_rmlp_nano_rw_256.sw_in1k) |83.03|96.34| 1341.24| 16.78| 4.37| 26.05|
|[maxvit_rmlp_nano_rw_256.sw_in1k](https://huggingface.co/timm/maxvit_rmlp_nano_rw_256.sw_in1k) |82.96|96.26| 1283.24| 15.50| 4.47| 31.92|
|[maxvit_nano_rw_256.sw_in1k](https://huggingface.co/timm/maxvit_nano_rw_256.sw_in1k) |82.93|96.23| 1218.17| 15.45| 4.46| 30.28|
|[coatnet_bn_0_rw_224.sw_in1k](https://huggingface.co/timm/coatnet_bn_0_rw_224.sw_in1k) |82.39|96.19| 1600.14| 27.44| 4.67| 22.04|
|[coatnet_0_rw_224.sw_in1k](https://huggingface.co/timm/coatnet_0_rw_224.sw_in1k) |82.39|95.84| 1831.21| 27.44| 4.43| 18.73|
|[coatnet_rmlp_nano_rw_224.sw_in1k](https://huggingface.co/timm/coatnet_rmlp_nano_rw_224.sw_in1k) |82.05|95.87| 2109.09| 15.15| 2.62| 20.34|
|[coatnext_nano_rw_224.sw_in1k](https://huggingface.co/timm/coatnext_nano_rw_224.sw_in1k) |81.95|95.92| 2525.52| 14.70| 2.47| 12.80|
|[coatnet_nano_rw_224.sw_in1k](https://huggingface.co/timm/coatnet_nano_rw_224.sw_in1k) |81.70|95.64| 2344.52| 15.14| 2.41| 15.41|
|[maxvit_rmlp_pico_rw_256.sw_in1k](https://huggingface.co/timm/maxvit_rmlp_pico_rw_256.sw_in1k) |80.53|95.21| 1594.71| 7.52| 1.85| 24.86|
Add tiny/small in12k pretrained and fine-tuned ConvNeXt models
2023-01-12 06:50:39 +08:00
### Jan 11, 2023
* Update ConvNeXt ImageNet-12k pretrain series w/ two new fine-tuned weights (and pre FT `.in12k` tags)
* `convnext_nano.in12k_ft_in1k` - 82.3 @ 224, 82.9 @ 288 (previously released)
* `convnext_tiny.in12k_ft_in1k` - 84.2 @ 224, 84.5 @ 288
* `convnext_small.in12k_ft_in1k` - 85.2 @ 224, 85.3 @ 288
Update README.md
2023-01-07 13:38:40 +08:00
### Jan 6, 2023
* Finally got around to adding `--model-kwargs` and `--opt-kwargs` to scripts to pass through rare args directly to model classes from cmd line
* `train.py /imagenet --model resnet50 --amp --model-kwargs output_stride=16 act_layer=silu`
Update README.md
2023-01-07 13:39:25 +08:00
* `train.py /imagenet --model vit_base_patch16_clip_224 --img-size 240 --amp --model-kwargs img_size=240 patch_size=12`
Fix typo
2023-03-09 10:42:54 +08:00
* Cleanup some popular models to better support arg passthrough / merge with model configs, more to go.
Update README.md
2022-12-10 03:13:37 +08:00
Add ConvNeXt-V2 support (model additions and weights) (#1614) * Add ConvNeXt-V2 support (model additions and weights) * ConvNeXt-V2 weights on HF Hub, tweaking some tests * Update README, fixing convnextv2 tests
2023-01-05 23:53:32 +08:00
### Jan 5, 2023
* ConvNeXt-V2 models and weights added to existing `convnext.py`
* Paper: [ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders](http://arxiv.org/abs/2301.00808)
* Reference impl: https://github.com/facebookresearch/ConvNeXt-V2 (NOTE: weights currently CC-BY-NC)
Update README
2022-12-24 07:19:54 +08:00
### Dec 23, 2022 🎄☃
* Add FlexiViT models and weights from https://github.com/google-research/big_vision (check out paper at https://arxiv.org/abs/2212.08013)
* NOTE currently resizing is static on model creation, on-the-fly dynamic / train patch size sampling is a WIP
* Many more models updated to multi-weight and downloadable via HF hub now (convnext, efficientnet, mobilenet, vision_transformer*, beit)
* More model pretrained tag and adjustments, some model names changed (working on deprecation translations, consider main branch DEV branch right now, use 0.6.x for stable use)
* More ImageNet-12k (subset of 22k) pretrain models popping up:
* `efficientnet_b5.in12k_ft_in1k` - 85.9 @ 448x448
Update README.md
2022-12-24 07:20:43 +08:00
* `vit_medium_patch16_gap_384.in12k_ft_in1k` - 85.5 @ 384x384
Update README
2022-12-24 07:19:54 +08:00
* `vit_medium_patch16_gap_256.in12k_ft_in1k` - 84.5 @ 256x256
* `convnext_nano.in12k_ft_in1k` - 82.9 @ 288x288
Update README.md
2022-12-10 03:14:35 +08:00
### Dec 8, 2022
Add EVA-large models
2022-12-09 08:20:49 +08:00
* Add 'EVA l' to `vision_transformer.py`, MAE style ViT-L/14 MIM pretrain w/ EVA-CLIP targets, FT on ImageNet-1k (w/ ImageNet-22k intermediate for some)
* original source: https://github.com/baaivision/EVA
| model | top1 | param_count | gmac | macts | hub |
|:------------------------------------------|-----:|------------:|------:|------:|:----------------------------------------|
| eva_large_patch14_336.in22k_ft_in22k_in1k | 89.2 | 304.5 | 191.1 | 270.2 | [link](https://huggingface.co/BAAI/EVA) |
| eva_large_patch14_336.in22k_ft_in1k | 88.7 | 304.5 | 191.1 | 270.2 | [link](https://huggingface.co/BAAI/EVA) |
| eva_large_patch14_196.in22k_ft_in22k_in1k | 88.6 | 304.1 | 61.6 | 63.5 | [link](https://huggingface.co/BAAI/EVA) |
| eva_large_patch14_196.in22k_ft_in1k | 87.9 | 304.1 | 61.6 | 63.5 | [link](https://huggingface.co/BAAI/EVA) |
Update README.md
2022-12-10 03:14:35 +08:00
### Dec 6, 2022
Fix typo
2023-03-09 10:42:54 +08:00
* Add 'EVA g', BEiT style ViT-g/14 model weights w/ both MIM pretrain and CLIP pretrain to `beit.py`.
Add EVA FT results, hopefully fix BEiT test failures
2022-12-07 15:14:59 +08:00
* original source: https://github.com/baaivision/EVA
* paper: https://arxiv.org/abs/2211.07636
| model | top1 | param_count | gmac | macts | hub |
|:-----------------------------------------|-------:|--------------:|-------:|--------:|:----------------------------------------|
| eva_giant_patch14_560.m30m_ft_in22k_in1k | 89.8 | 1014.4 | 1906.8 | 2577.2 | [link](https://huggingface.co/BAAI/EVA) |
| eva_giant_patch14_336.m30m_ft_in22k_in1k | 89.6 | 1013 | 620.6 | 550.7 | [link](https://huggingface.co/BAAI/EVA) |
| eva_giant_patch14_336.clip_ft_in1k | 89.4 | 1013 | 620.6 | 550.7 | [link](https://huggingface.co/BAAI/EVA) |
| eva_giant_patch14_224.clip_ft_in1k | 89.1 | 1012.6 | 267.2 | 192.6 | [link](https://huggingface.co/BAAI/EVA) |
Add eva models to beit.py
2022-12-07 08:45:11 +08:00
Update README.md
2022-12-10 03:14:35 +08:00
### Dec 5, 2022
Update multi-weight release README
2022-12-06 02:16:47 +08:00
Update README.md
2022-12-06 02:20:08 +08:00
* Pre-release (`0.8.0dev0`) of multi-weight support (`model_arch.pretrained_tag`). Install with `pip install --pre timm`
Update multi-weight release README
2022-12-06 02:16:47 +08:00
* vision_transformer, maxvit, convnext are the first three model impl w/ support
* model names are changing with this (previous _21k, etc. fn will merge), still sorting out deprecation handling
* bugs are likely, but I need feedback so please try it out
* if stability is needed, please use 0.6.x pypi releases or clone from [0.6.x branch](https://github.com/rwightman/pytorch-image-models/tree/0.6.x)
* Support for PyTorch 2.0 compile is added in train/validate/inference/benchmark, use `--torchcompile` argument
* Inference script allows more control over output, select k for top-class index + prob json, csv or parquet output
* Add a full set of fine-tuned CLIP image tower weights from both LAION-2B and original OpenAI CLIP models
| model | top1 | param_count | gmac | macts | hub |
|:-------------------------------------------------|-------:|--------------:|-------:|--------:|:-------------------------------------------------------------------------------------|
| vit_huge_patch14_clip_336.laion2b_ft_in12k_in1k | 88.6 | 632.5 | 391 | 407.5 | [link](https://huggingface.co/timm/vit_huge_patch14_clip_336.laion2b_ft_in12k_in1k) |
| vit_large_patch14_clip_336.openai_ft_in12k_in1k | 88.3 | 304.5 | 191.1 | 270.2 | [link](https://huggingface.co/timm/vit_large_patch14_clip_336.openai_ft_in12k_in1k) |
| vit_huge_patch14_clip_224.laion2b_ft_in12k_in1k | 88.2 | 632 | 167.4 | 139.4 | [link](https://huggingface.co/timm/vit_huge_patch14_clip_224.laion2b_ft_in12k_in1k) |
| vit_large_patch14_clip_336.laion2b_ft_in12k_in1k | 88.2 | 304.5 | 191.1 | 270.2 | [link](https://huggingface.co/timm/vit_large_patch14_clip_336.laion2b_ft_in12k_in1k) |
| vit_large_patch14_clip_224.openai_ft_in12k_in1k | 88.2 | 304.2 | 81.1 | 88.8 | [link](https://huggingface.co/timm/vit_large_patch14_clip_224.openai_ft_in12k_in1k) |
| vit_large_patch14_clip_224.laion2b_ft_in12k_in1k | 87.9 | 304.2 | 81.1 | 88.8 | [link](https://huggingface.co/timm/vit_large_patch14_clip_224.laion2b_ft_in12k_in1k) |
| vit_large_patch14_clip_224.openai_ft_in1k | 87.9 | 304.2 | 81.1 | 88.8 | [link](https://huggingface.co/timm/vit_large_patch14_clip_224.openai_ft_in1k) |
| vit_large_patch14_clip_336.laion2b_ft_in1k | 87.9 | 304.5 | 191.1 | 270.2 | [link](https://huggingface.co/timm/vit_large_patch14_clip_336.laion2b_ft_in1k) |
| vit_huge_patch14_clip_224.laion2b_ft_in1k | 87.6 | 632 | 167.4 | 139.4 | [link](https://huggingface.co/timm/vit_huge_patch14_clip_224.laion2b_ft_in1k) |
| vit_large_patch14_clip_224.laion2b_ft_in1k | 87.3 | 304.2 | 81.1 | 88.8 | [link](https://huggingface.co/timm/vit_large_patch14_clip_224.laion2b_ft_in1k) |
| vit_base_patch16_clip_384.laion2b_ft_in12k_in1k | 87.2 | 86.9 | 55.5 | 101.6 | [link](https://huggingface.co/timm/vit_base_patch16_clip_384.laion2b_ft_in12k_in1k) |
| vit_base_patch16_clip_384.openai_ft_in12k_in1k | 87 | 86.9 | 55.5 | 101.6 | [link](https://huggingface.co/timm/vit_base_patch16_clip_384.openai_ft_in12k_in1k) |
| vit_base_patch16_clip_384.laion2b_ft_in1k | 86.6 | 86.9 | 55.5 | 101.6 | [link](https://huggingface.co/timm/vit_base_patch16_clip_384.laion2b_ft_in1k) |
| vit_base_patch16_clip_384.openai_ft_in1k | 86.2 | 86.9 | 55.5 | 101.6 | [link](https://huggingface.co/timm/vit_base_patch16_clip_384.openai_ft_in1k) |
| vit_base_patch16_clip_224.laion2b_ft_in12k_in1k | 86.2 | 86.6 | 17.6 | 23.9 | [link](https://huggingface.co/timm/vit_base_patch16_clip_224.laion2b_ft_in12k_in1k) |
| vit_base_patch16_clip_224.openai_ft_in12k_in1k | 85.9 | 86.6 | 17.6 | 23.9 | [link](https://huggingface.co/timm/vit_base_patch16_clip_224.openai_ft_in12k_in1k) |
| vit_base_patch32_clip_448.laion2b_ft_in12k_in1k | 85.8 | 88.3 | 17.9 | 23.9 | [link](https://huggingface.co/timm/vit_base_patch32_clip_448.laion2b_ft_in12k_in1k) |
| vit_base_patch16_clip_224.laion2b_ft_in1k | 85.5 | 86.6 | 17.6 | 23.9 | [link](https://huggingface.co/timm/vit_base_patch16_clip_224.laion2b_ft_in1k) |
| vit_base_patch32_clip_384.laion2b_ft_in12k_in1k | 85.4 | 88.3 | 13.1 | 16.5 | [link](https://huggingface.co/timm/vit_base_patch32_clip_384.laion2b_ft_in12k_in1k) |
| vit_base_patch16_clip_224.openai_ft_in1k | 85.3 | 86.6 | 17.6 | 23.9 | [link](https://huggingface.co/timm/vit_base_patch16_clip_224.openai_ft_in1k) |
| vit_base_patch32_clip_384.openai_ft_in12k_in1k | 85.2 | 88.3 | 13.1 | 16.5 | [link](https://huggingface.co/timm/vit_base_patch32_clip_384.openai_ft_in12k_in1k) |
| vit_base_patch32_clip_224.laion2b_ft_in12k_in1k | 83.3 | 88.2 | 4.4 | 5 | [link](https://huggingface.co/timm/vit_base_patch32_clip_224.laion2b_ft_in12k_in1k) |
| vit_base_patch32_clip_224.laion2b_ft_in1k | 82.6 | 88.2 | 4.4 | 5 | [link](https://huggingface.co/timm/vit_base_patch32_clip_224.laion2b_ft_in1k) |
| vit_base_patch32_clip_224.openai_ft_in1k | 81.9 | 88.2 | 4.4 | 5 | [link](https://huggingface.co/timm/vit_base_patch32_clip_224.openai_ft_in1k) |
* Port of MaxViT Tensorflow Weights from official impl at https://github.com/google-research/maxvit
* There was larger than expected drops for the upscaled 384/512 in21k fine-tune weights, possible detail missing, but the 21k FT did seem sensitive to small preprocessing
| model | top1 | param_count | gmac | macts | hub |
|:-----------------------------------|-------:|--------------:|-------:|--------:|:-----------------------------------------------------------------------|
| maxvit_xlarge_tf_512.in21k_ft_in1k | 88.5 | 475.8 | 534.1 | 1413.2 | [link](https://huggingface.co/timm/maxvit_xlarge_tf_512.in21k_ft_in1k) |
| maxvit_xlarge_tf_384.in21k_ft_in1k | 88.3 | 475.3 | 292.8 | 668.8 | [link](https://huggingface.co/timm/maxvit_xlarge_tf_384.in21k_ft_in1k) |
| maxvit_base_tf_512.in21k_ft_in1k | 88.2 | 119.9 | 138 | 704 | [link](https://huggingface.co/timm/maxvit_base_tf_512.in21k_ft_in1k) |
| maxvit_large_tf_512.in21k_ft_in1k | 88 | 212.3 | 244.8 | 942.2 | [link](https://huggingface.co/timm/maxvit_large_tf_512.in21k_ft_in1k) |
| maxvit_large_tf_384.in21k_ft_in1k | 88 | 212 | 132.6 | 445.8 | [link](https://huggingface.co/timm/maxvit_large_tf_384.in21k_ft_in1k) |
| maxvit_base_tf_384.in21k_ft_in1k | 87.9 | 119.6 | 73.8 | 332.9 | [link](https://huggingface.co/timm/maxvit_base_tf_384.in21k_ft_in1k) |
| maxvit_base_tf_512.in1k | 86.6 | 119.9 | 138 | 704 | [link](https://huggingface.co/timm/maxvit_base_tf_512.in1k) |
| maxvit_large_tf_512.in1k | 86.5 | 212.3 | 244.8 | 942.2 | [link](https://huggingface.co/timm/maxvit_large_tf_512.in1k) |
| maxvit_base_tf_384.in1k | 86.3 | 119.6 | 73.8 | 332.9 | [link](https://huggingface.co/timm/maxvit_base_tf_384.in1k) |
| maxvit_large_tf_384.in1k | 86.2 | 212 | 132.6 | 445.8 | [link](https://huggingface.co/timm/maxvit_large_tf_384.in1k) |
| maxvit_small_tf_512.in1k | 86.1 | 69.1 | 67.3 | 383.8 | [link](https://huggingface.co/timm/maxvit_small_tf_512.in1k) |
| maxvit_tiny_tf_512.in1k | 85.7 | 31 | 33.5 | 257.6 | [link](https://huggingface.co/timm/maxvit_tiny_tf_512.in1k) |
| maxvit_small_tf_384.in1k | 85.5 | 69 | 35.9 | 183.6 | [link](https://huggingface.co/timm/maxvit_small_tf_384.in1k) |
| maxvit_tiny_tf_384.in1k | 85.1 | 31 | 17.5 | 123.4 | [link](https://huggingface.co/timm/maxvit_tiny_tf_384.in1k) |
| maxvit_large_tf_224.in1k | 84.9 | 211.8 | 43.7 | 127.4 | [link](https://huggingface.co/timm/maxvit_large_tf_224.in1k) |
| maxvit_base_tf_224.in1k | 84.9 | 119.5 | 24 | 95 | [link](https://huggingface.co/timm/maxvit_base_tf_224.in1k) |
| maxvit_small_tf_224.in1k | 84.4 | 68.9 | 11.7 | 53.2 | [link](https://huggingface.co/timm/maxvit_small_tf_224.in1k) |
| maxvit_tiny_tf_224.in1k | 83.4 | 30.9 | 5.6 | 35.8 | [link](https://huggingface.co/timm/maxvit_tiny_tf_224.in1k) |
Update README.md
2022-10-16 00:33:34 +08:00
### Oct 15, 2022
* Train and validation script enhancements
* Non-GPU (ie CPU) device support
* SLURM compatibility for train script
* HF datasets support (via ReaderHfds)
* TFDS/WDS dataloading improvements (sample padding/wrap for distributed use fixed wrt sample count estimate)
* in_chans !=3 support for scripts / loader
* Adan optimizer
* Can enable per-step LR scheduling via args
* Dataset 'parsers' renamed to 'readers', more descriptive of purpose
* AMP args changed, APEX via `--amp-impl apex`, bfloat16 supportedf via `--amp-dtype bfloat16`
* main branch switched to 0.7.x version, 0.6x forked for stable release of weight only adds
* master -> main branch rename
Add more maxxvit weights includ ConvNeXt conv block based experiments.
2022-10-11 12:49:18 +08:00
### Oct 10, 2022
* More weights in `maxxvit` series, incl first ConvNeXt block based `coatnext` and `maxxvit` experiments:
* `coatnext_nano_rw_224` - 82.0 @ 224 (G) -- (uses ConvNeXt conv block, no BatchNorm)
Update README.md
2022-10-11 13:13:05 +08:00
* `maxxvit_rmlp_nano_rw_256` - 83.0 @ 256, 83.7 @ 320 (G) (uses ConvNeXt conv block, no BN)
Add more maxxvit weights includ ConvNeXt conv block based experiments.
2022-10-11 12:49:18 +08:00
* `maxvit_rmlp_small_rw_224` - 84.5 @ 224, 85.1 @ 320 (G)
Update README.md
2022-10-11 13:13:05 +08:00
* `maxxvit_rmlp_small_rw_256` - 84.6 @ 256, 84.9 @ 288 (G) -- could be trained better, hparams need tuning (uses ConvNeXt block, no BN)
Add more maxxvit weights includ ConvNeXt conv block based experiments.
2022-10-11 12:49:18 +08:00
* `coatnet_rmlp_2_rw_224` - 84.6 @ 224, 85 @ 320 (T)
Update README.md
2022-10-12 04:13:06 +08:00
* NOTE: official MaxVit weights (in1k) have been released at https://github.com/google-research/maxvit -- some extra work is needed to port and adapt since my impl was created independently of theirs and has a few small differences + the whole TF same padding fun.
Fix typo
2023-03-09 10:42:54 +08:00
Add more maxxvit weights includ ConvNeXt conv block based experiments.
2022-10-11 12:49:18 +08:00
### Sept 23, 2022
* LAION-2B CLIP image towers supported as pretrained backbones for fine-tune or features (no classifier)
* vit_base_patch32_224_clip_laion2b
* vit_large_patch14_224_clip_laion2b
* vit_huge_patch14_224_clip_laion2b
* vit_giant_patch14_224_clip_laion2b
Update README.md
2022-09-08 01:40:58 +08:00
### Sept 7, 2022
* Hugging Face [`timm` docs](https://huggingface.co/docs/hub/timm) home now exists, look for more here in the future
* Add BEiT-v2 weights for base and large 224x224 models from https://github.com/microsoft/unilm/tree/master/beit2
* Add more weights in `maxxvit` series incl a `pico` (7.5M params, 1.9 GMACs), two `tiny` variants:
* `maxvit_rmlp_pico_rw_256` - 80.5 @ 256, 81.3 @ 320 (T)
* `maxvit_tiny_rw_224` - 83.5 @ 224 (G)
* `maxvit_rmlp_tiny_rw_256` - 84.2 @ 256, 84.8 @ 320 (T)
Add lcnet_100 and mnasnet_small weights
2022-01-07 14:21:05 +08:00
Start updating README and docs
2020-08-05 07:49:03 +08:00
## Introduction
Update README.md, check in hubconf experiment (fully dynamic)
2020-02-19 06:41:14 +08:00
Continuing README and documentation updates
2020-08-06 01:38:35 +08:00
Py**T**orch **Im**age **M**odels (`timm`) is a collection of image models, layers, utilities, optimizers, schedulers, data-loaders / augmentations, and reference training / validation scripts that aim to pull together a wide variety of SOTA models with ability to reproduce ImageNet training results.
Add ported EfficientNet-L2, B0-B7 NoisyStudent weights from TF TPU
2020-02-13 03:26:38 +08:00
Continuing README and documentation updates
2020-08-06 01:38:35 +08:00
The work of many others is present here. I've tried to make sure all source material is acknowledged via links to github, arxiv papers, etc in the README, documentation, and code docstrings. Please let me know if I missed anything.
Add update RandAugment MixNet-XL weights
2019-12-25 02:08:24 +08:00
Create README.md
2019-04-15 06:10:52 +08:00
## Models
Update README.md Remove references to github.io docs that died with repo move
2023-02-25 06:43:31 +08:00
All model architecture families include variants with pretrained weights. There are specific model variants without any weights, it is NOT a bug. Help training new or better weights is always appreciated.
Continuing README and documentation updates
2020-08-06 01:38:35 +08:00
Update README.md
2021-07-13 04:33:02 +08:00
* Aggregating Nested Transformers - https://arxiv.org/abs/2105.12723
Update README.md
2022-01-15 12:11:45 +08:00
* BEiT - https://arxiv.org/abs/2106.08254
Add BiT references and knowledge distill links to readme/docs
2021-01-17 04:30:02 +08:00
* Big Transfer ResNetV2 (BiT) - https://arxiv.org/abs/1912.11370
Update README again. Add 101x3 BiT-M model to CI ignore since it's starting to fail in GitHub runners.
2021-04-14 04:36:24 +08:00
* Bottleneck Transformers - https://arxiv.org/abs/2101.11605
Update README.md
2021-05-06 14:17:35 +08:00
* CaiT (Class-Attention in Image Transformers) - https://arxiv.org/abs/2103.17239
* CoaT (Co-Scale Conv-Attentional Image Transformers) - https://arxiv.org/abs/2104.06399
Update README.md for merge
2022-08-26 06:56:56 +08:00
* CoAtNet (Convolution and Attention) - https://arxiv.org/abs/2106.04803
Update README.md
2022-01-15 12:11:45 +08:00
* ConvNeXt - https://arxiv.org/abs/2201.03545
Add ConvNeXt-V2 support (model additions and weights) (#1614) * Add ConvNeXt-V2 support (model additions and weights) * ConvNeXt-V2 weights on HF Hub, tweaking some tests * Update README, fixing convnextv2 tests
2023-01-05 23:53:32 +08:00
* ConvNeXt-V2 - http://arxiv.org/abs/2301.00808
Update README.md before merge. Bump version to 0.4.10
2021-05-26 03:25:53 +08:00
* ConViT (Soft Convolutional Inductive Biases Vision Transformers)- https://arxiv.org/abs/2103.10697
Continuing README and documentation updates
2020-08-06 01:38:35 +08:00
* CspNet (Cross-Stage Partial Networks) - https://arxiv.org/abs/1911.11929
Update README and changelogs
2022-07-28 05:04:29 +08:00
* DeiT - https://arxiv.org/abs/2012.12877
* DeiT-III - https://arxiv.org/pdf/2204.07118.pdf
Continuing README and documentation updates
2020-08-06 01:38:35 +08:00
* DenseNet - https://arxiv.org/abs/1608.06993
* DLA - https://arxiv.org/abs/1707.06484
* DPN (Dual-Path Network) - https://arxiv.org/abs/1707.01629
Update README.md
2022-07-12 09:08:34 +08:00
* EdgeNeXt - https://arxiv.org/abs/2206.10589
Update README.md for merge
2022-08-26 06:56:56 +08:00
* EfficientFormer - https://arxiv.org/abs/2206.01191
Continuing README and documentation updates
2020-08-06 01:38:35 +08:00
* EfficientNet (MBConvNet Family)
* EfficientNet NoisyStudent (B0-B7, L2) - https://arxiv.org/abs/1911.04252
* EfficientNet AdvProp (B0-B8) - https://arxiv.org/abs/1911.09665
* EfficientNet (B0-B7) - https://arxiv.org/abs/1905.11946
* EfficientNet-EdgeTPU (S, M, L) - https://ai.googleblog.com/2019/08/efficientnet-edgetpu-creating.html
Update README.md
2021-05-06 14:17:35 +08:00
* EfficientNet V2 - https://arxiv.org/abs/2104.00298
Continuing README and documentation updates
2020-08-06 01:38:35 +08:00
* FBNet-C - https://arxiv.org/abs/1812.03443
* MixNet - https://arxiv.org/abs/1907.09595
* MNASNet B1, A1 (Squeeze-Excite), and Small - https://arxiv.org/abs/1807.11626
* MobileNet-V2 - https://arxiv.org/abs/1801.04381
* Single-Path NAS - https://arxiv.org/abs/1904.02877
Update README.md
2022-01-15 12:11:45 +08:00
* TinyNet - https://arxiv.org/abs/2010.14819
Update README, add --reparm to onnx_export
2023-08-26 02:04:34 +08:00
* EfficientViT (MIT) - https://arxiv.org/abs/2205.14756
* EfficientViT (MSRA) - https://arxiv.org/abs/2305.07027
Update README.md
2022-12-08 01:34:21 +08:00
* EVA - https://arxiv.org/abs/2211.07636
Update README.md
2023-04-01 03:24:10 +08:00
* EVA-02 - https://arxiv.org/abs/2303.11331
Update README, add --reparm to onnx_export
2023-08-26 02:04:34 +08:00
* FastViT - https://arxiv.org/abs/2303.14189
Add ConvNeXt-V2 support (model additions and weights) (#1614) * Add ConvNeXt-V2 support (model additions and weights) * ConvNeXt-V2 weights on HF Hub, tweaking some tests * Update README, fixing convnextv2 tests
2023-01-05 23:53:32 +08:00
* FlexiViT - https://arxiv.org/abs/2212.08013
Update README
2023-03-23 06:40:10 +08:00
* FocalNet (Focal Modulation Networks) - https://arxiv.org/abs/2203.11926
Missed GCVit in README paper links
2022-08-29 06:22:28 +08:00
* GCViT (Global Context Vision Transformer) - https://arxiv.org/abs/2206.09959
Update README.md
2021-05-06 14:17:35 +08:00
* GhostNet - https://arxiv.org/abs/1911.11907
Update README, add --reparm to onnx_export
2023-08-26 02:04:34 +08:00
* GhostNet-V2 - https://arxiv.org/abs/2211.12905
Update README.md before merge. Bump version to 0.4.10
2021-05-26 03:25:53 +08:00
* gMLP - https://arxiv.org/abs/2105.08050
Update README.md and docs. Version bumped to 0.4.3
2021-02-11 06:46:07 +08:00
* GPU-Efficient Networks - https://arxiv.org/abs/2006.14090
Update README again. Add 101x3 BiT-M model to CI ignore since it's starting to fail in GitHub runners.
2021-04-14 04:36:24 +08:00
* Halo Nets - https://arxiv.org/abs/2103.12731
Continuing README and documentation updates
2020-08-06 01:38:35 +08:00
* HRNet - https://arxiv.org/abs/1908.07919
Update README, add --reparm to onnx_export
2023-08-26 02:04:34 +08:00
* InceptionNeXt - https://arxiv.org/abs/2303.16900
Continuing README and documentation updates
2020-08-06 01:38:35 +08:00
* Inception-V3 - https://arxiv.org/abs/1512.00567
* Inception-ResNet-V2 and Inception-V4 - https://arxiv.org/abs/1602.07261
Update README again. Add 101x3 BiT-M model to CI ignore since it's starting to fail in GitHub runners.
2021-04-14 04:36:24 +08:00
* Lambda Networks - https://arxiv.org/abs/2102.08602
Update README.md before merge. Bump version to 0.4.10
2021-05-26 03:25:53 +08:00
* LeViT (Vision Transformer in ConvNet's Clothing) - https://arxiv.org/abs/2104.01136
Update README.md for merge
2022-08-26 06:56:56 +08:00
* MaxViT (Multi-Axis Vision Transformer) - https://arxiv.org/abs/2204.01697
Update README, add --reparm to onnx_export
2023-08-26 02:04:34 +08:00
* MetaFormer (PoolFormer-v2, ConvFormer, CAFormer) - https://arxiv.org/abs/2210.13452
Update README.md
2021-05-06 14:17:35 +08:00
* MLP-Mixer - https://arxiv.org/abs/2105.01601
Continuing README and documentation updates
2020-08-06 01:38:35 +08:00
* MobileNet-V3 (MBConvNet w/ Efficient Head) - https://arxiv.org/abs/1905.02244
Update README.md
2022-01-15 12:11:45 +08:00
* FBNet-V3 - https://arxiv.org/abs/2006.02049
* HardCoRe-NAS - https://arxiv.org/abs/2102.11646
* LCNet - https://arxiv.org/abs/2109.15099
Update README, add --reparm to onnx_export
2023-08-26 02:04:34 +08:00
* MobileOne - https://arxiv.org/abs/2206.04040
Update README.md
2022-03-24 07:16:26 +08:00
* MobileViT - https://arxiv.org/abs/2110.02178
Update README and changelogs
2022-07-28 05:04:29 +08:00
* MobileViT-V2 - https://arxiv.org/abs/2206.02680
Update README.md for merge
2022-08-26 06:56:56 +08:00
* MViT-V2 (Improved Multiscale Vision Transformer) - https://arxiv.org/abs/2112.01526
Continuing README and documentation updates
2020-08-06 01:38:35 +08:00
* NASNet-A - https://arxiv.org/abs/1707.07012
Update README.md
2022-01-15 12:11:45 +08:00
* NesT - https://arxiv.org/abs/2105.12723
Update README.md and few more comments
2021-02-13 14:07:18 +08:00
* NFNet-F - https://arxiv.org/abs/2102.06171
Update README.md for NF-nets, bump version to 0.4.1 for merge
2021-01-31 15:19:45 +08:00
* NF-RegNet / NF-ResNet - https://arxiv.org/abs/2101.08692
Continuing README and documentation updates
2020-08-06 01:38:35 +08:00
* PNasNet - https://arxiv.org/abs/1712.00559
Update README.md
2022-03-24 07:16:26 +08:00
* PoolFormer (MetaFormer) - https://arxiv.org/abs/2111.11418
Update README.md with updates. Small tweak to head_dist handling.
2021-04-02 08:49:05 +08:00
* Pooling-based Vision Transformer (PiT) - https://arxiv.org/abs/2103.16302
Update README.md for merge
2022-08-26 06:56:56 +08:00
* PVT-V2 (Improved Pyramid Vision Transformer) - https://arxiv.org/abs/2106.13797
Continuing README and documentation updates
2020-08-06 01:38:35 +08:00
* RegNet - https://arxiv.org/abs/2003.13678
Update README.md
2022-03-24 07:16:26 +08:00
* RegNetZ - https://arxiv.org/abs/2103.06877
Update README.md and docs. Version bumped to 0.4.3
2021-02-11 06:46:07 +08:00
* RepVGG - https://arxiv.org/abs/2101.03697
Update README, add --reparm to onnx_export
2023-08-26 02:04:34 +08:00
* RepGhostNet - https://arxiv.org/abs/2211.06088
* RepViT - https://arxiv.org/abs/2307.09283
Update README.md before merge. Bump version to 0.4.10
2021-05-26 03:25:53 +08:00
* ResMLP - https://arxiv.org/abs/2105.03404
Continuing README and documentation updates
2020-08-06 01:38:35 +08:00
* ResNet/ResNeXt
* ResNet (v1b/v1.5) - https://arxiv.org/abs/1512.03385
* ResNeXt - https://arxiv.org/abs/1611.05431
* 'Bag of Tricks' / Gluon C, D, E, S variations - https://arxiv.org/abs/1812.01187
* Weakly-supervised (WSL) Instagram pretrained / ImageNet tuned ResNeXt101 - https://arxiv.org/abs/1805.00932
* Semi-supervised (SSL) / Semi-weakly Supervised (SWSL) ResNet/ResNeXts - https://arxiv.org/abs/1905.00546
* ECA-Net (ECAResNet) - https://arxiv.org/abs/1910.03151v4
* Squeeze-and-Excitation Networks (SEResNet) - https://arxiv.org/abs/1709.01507
Update README.md
2021-05-06 14:17:35 +08:00
* ResNet-RS - https://arxiv.org/abs/2103.07579
Continuing README and documentation updates
2020-08-06 01:38:35 +08:00
* Res2Net - https://arxiv.org/abs/1904.01169
* ResNeSt - https://arxiv.org/abs/2004.08955
* ReXNet - https://arxiv.org/abs/2007.00992
* SelecSLS - https://arxiv.org/abs/1907.00837
* Selective Kernel Networks - https://arxiv.org/abs/1903.06586
Update README.md, version to 0.6.2
2022-05-14 04:54:41 +08:00
* Sequencer2D - https://arxiv.org/abs/2205.01972
Update README.md
2022-03-24 07:16:26 +08:00
* Swin S3 (AutoFormerV2) - https://arxiv.org/abs/2111.14725
Update README again. Add 101x3 BiT-M model to CI ignore since it's starting to fail in GitHub runners.
2021-04-14 04:36:24 +08:00
* Swin Transformer - https://arxiv.org/abs/2103.14030
Update README.md
2022-03-24 07:16:26 +08:00
* Swin Transformer V2 - https://arxiv.org/abs/2111.09883
Update README.md with updates. Small tweak to head_dist handling.
2021-04-02 08:49:05 +08:00
* Transformer-iN-Transformer (TNT) - https://arxiv.org/abs/2103.00112
Continuing README and documentation updates
2020-08-06 01:38:35 +08:00
* TResNet - https://arxiv.org/abs/2003.13630
Update README.md before merge. Bump version to 0.4.10
2021-05-26 03:25:53 +08:00
* Twins (Spatial Attention in Vision Transformers) - https://arxiv.org/pdf/2104.13840.pdf
Update README.md
2022-01-15 12:11:45 +08:00
* Visformer - https://arxiv.org/abs/2104.12533
Update README.md and docs in prep for 0.3.0 PyPi release.
2020-10-31 04:27:19 +08:00
* Vision Transformer - https://arxiv.org/abs/2010.11929
Update README.md
2022-03-24 07:16:26 +08:00
* VOLO (Vision Outlooker) - https://arxiv.org/abs/2106.13112
More documentation updates, fix a typo
2020-08-06 06:59:31 +08:00
* VovNet V2 and V1 - https://arxiv.org/abs/1911.06667
Continuing README and documentation updates
2020-08-06 01:38:35 +08:00
* Xception - https://arxiv.org/abs/1610.02357
* Xception (Modified Aligned, Gluon) - https://arxiv.org/abs/1802.02611
* Xception (Modified Aligned, TF) - https://arxiv.org/abs/1802.02611
Update README.md
2021-07-13 04:33:02 +08:00
* XCiT (Cross-Covariance Image Transformers) - https://arxiv.org/abs/2106.09681
Add some info to README
2019-05-16 00:54:07 +08:00
Create README.md
2019-04-15 06:10:52 +08:00
## Features
Continuing README and documentation updates
2020-08-06 01:38:35 +08:00
Create README.md
2019-04-15 06:10:52 +08:00
Several (less common) features that I often utilize in my projects are included. Many of their additions are the reason why I maintain my own set of models, instead of using others' via PIP:
Continuing README and documentation updates
2020-08-06 01:38:35 +08:00
Create README.md
2019-04-15 06:10:52 +08:00
* All models have a common default configuration interface and API for
* accessing/changing the classifier - `get_classifier` and `reset_classifier`
Update README.md Remove references to github.io docs that died with repo move
2023-02-25 06:43:31 +08:00
* doing a forward pass on just the features - `forward_features` (see [documentation](https://huggingface.co/docs/timm/feature_extraction))
Create README.md
2019-04-15 06:10:52 +08:00
* these makes it easy to write consistent network wrappers that work with any of the models
Update README.md Remove references to github.io docs that died with repo move
2023-02-25 06:43:31 +08:00
* All models support multi-scale feature map extraction (feature pyramids) via create_model (see [documentation](https://huggingface.co/docs/timm/feature_extraction))
Start updating README and docs
2020-08-05 07:49:03 +08:00
* `create_model(name, features_only=True, out_indices=..., output_stride=...)`
More documentation updates, fix a typo
2020-08-06 06:59:31 +08:00
* `out_indices` creation arg specifies which feature maps to return, these indices are 0 based and generally correspond to the `C(i + 1)` feature level.
* `output_stride` creation arg controls output stride of the network by using dilated convolutions. Most networks are stride 32 by default. Not all networks support this.
Start updating README and docs
2020-08-05 07:49:03 +08:00
* feature map channel counts, reduction level (stride) can be queried AFTER model creation via the `.feature_info` member
Create README.md
2019-04-15 06:10:52 +08:00
* All models have a consistent pretrained weight loader that adapts last linear if necessary, and from 3 to 1 channel input if desired
Update README.md Remove references to github.io docs that died with repo move
2023-02-25 06:43:31 +08:00
* High performance [reference training, validation, and inference scripts](https://huggingface.co/docs/timm/training_script) that work in several process/GPU modes:
Create README.md
2019-04-15 06:10:52 +08:00
* NVIDIA DDP w/ a single GPU per process, multiple processes with APEX present (AMP mixed-precision optional)
* PyTorch DistributedDataParallel w/ multi-gpu, single process (AMP disabled as it crashes when enabled)
* PyTorch w/ single GPU single process (AMP optional)
* A dynamic global pool implementation that allows selecting from average pooling, max pooling, average + max, or concat([average, max]) at model creation. All global pooling is adaptive average by default and compatible with pretrained weights.
README: fix simple typos
2021-05-26 04:42:44 +08:00
* A 'Test Time Pool' wrapper that can wrap any of the included models and usually provides improved performance doing inference with input images larger than the training size. Idea adapted from original DPN implementation when I ported (https://github.com/cypw/DPNs)
Start updating README and docs
2020-08-05 07:49:03 +08:00
* Learning rate schedulers
* Ideas adopted from
* [AllenNLP schedulers](https://github.com/allenai/allennlp/tree/master/allennlp/training/learning_rate_schedulers)
* [FAIRseq lr_scheduler](https://github.com/pytorch/fairseq/tree/master/fairseq/optim/lr_scheduler)
* SGDR: Stochastic Gradient Descent with Warm Restarts (https://arxiv.org/abs/1608.03983)
* Schedulers include `step`, `cosine` w/ restarts, `tanh` w/ restarts, `plateau`
* Optimizers:
* `rmsprop_tf` adapted from PyTorch RMSProp by myself. Reproduces much improved Tensorflow RMSProp behaviour.
* `radam` by [Liyuan Liu](https://github.com/LiyuanLucasLiu/RAdam) (https://arxiv.org/abs/1908.03265)
* `novograd` by [Masashi Kimura](https://github.com/convergence-lab/novograd) (https://arxiv.org/abs/1905.11286)
* `lookahead` adapted from impl by [Liam](https://github.com/alphadl/lookahead.pytorch) (https://arxiv.org/abs/1907.08610)
* `fused<name>` optimizers by name with [NVIDIA Apex](https://github.com/NVIDIA/apex/tree/master/apex/optimizers) installed
* `adamp` and `sgdp` by [Naver ClovAI](https://github.com/clovaai) (https://arxiv.org/abs/2006.08217)
Update README, ensure vit excluded from all tests (not ready)
2020-10-14 05:51:08 +08:00
* `adafactor` adapted from [FAIRSeq impl](https://github.com/pytorch/fairseq/blob/master/fairseq/optim/adafactor.py) (https://arxiv.org/abs/1804.04235)
* `adahessian` by [David Samuel](https://github.com/davda54/ada-hessian) (https://arxiv.org/abs/2006.00719)
Start updating README and docs
2020-08-05 07:49:03 +08:00
* Random Erasing from [Zhun Zhong](https://github.com/zhunzhong07/Random-Erasing/blob/master/transforms.py) (https://arxiv.org/abs/1708.04896)
Update README.md
2020-08-13 14:26:59 +08:00
* Mixup (https://arxiv.org/abs/1710.09412)
* CutMix (https://arxiv.org/abs/1905.04899)
Update Auto/RandAugment comments, README, more. * Add a weighted choice option for RandAugment * Adjust magnitude noise/std naming, config
2019-11-23 05:00:24 +08:00
* AutoAugment (https://arxiv.org/abs/1805.09501) and RandAugment (https://arxiv.org/abs/1909.13719) ImageNet configurations modeled after impl for EfficientNet training (https://github.com/tensorflow/tpu/blob/master/models/official/efficientnet/autoaugment.py)
Update README.md
2020-01-13 10:16:36 +08:00
* AugMix w/ JSD loss (https://arxiv.org/abs/1912.02781), JSD w/ clean + augmented mixing support works with AutoAugment and RandAugment as well
* SplitBachNorm - allows splitting batch norm layers between clean and augmented (auxiliary batch norm) data
Fix typo
2023-03-09 10:42:54 +08:00
* DropPath aka "Stochastic Depth" (https://arxiv.org/abs/1603.09382)
Update README.md, check in hubconf experiment (fully dynamic)
2020-02-19 06:41:14 +08:00
* DropBlock (https://arxiv.org/abs/1810.12890)
Start updating README and docs
2020-08-05 07:49:03 +08:00
* Blur Pooling (https://arxiv.org/abs/1904.11486)
Continuing README and documentation updates
2020-08-06 01:38:35 +08:00
* Space-to-Depth by [mrT23](https://github.com/mrT23/TResNet/blob/master/src/models/tresnet/layers/space_to_depth.py) (https://arxiv.org/abs/1801.04590) -- original paper?
Update README for AGC and bump version to 0.4.4
2021-02-17 01:12:23 +08:00
* Adaptive Gradient Clipping (https://arxiv.org/abs/2102.06171, https://github.com/deepmind/deepmind-research/tree/master/nfnets)
Add non-local and BAT attention. Merge attn and self-attn factories into one. Add attention references to README. Add mlp 'mode' to ECA.
2021-06-01 04:18:11 +08:00
* An extensive selection of channel and/or spatial attention modules:
* Bottleneck Transformer - https://arxiv.org/abs/2101.11605
* CBAM - https://arxiv.org/abs/1807.06521
* Effective Squeeze-Excitation (ESE) - https://arxiv.org/abs/1911.06667
* Efficient Channel Attention (ECA) - https://arxiv.org/abs/1910.03151
* Gather-Excite (GE) - https://arxiv.org/abs/1810.12348
* Global Context (GC) - https://arxiv.org/abs/1904.11492
* Halo - https://arxiv.org/abs/2103.12731
* Involution - https://arxiv.org/abs/2103.06255
* Lambda Layer - https://arxiv.org/abs/2102.08602
* Non-Local (NL) - https://arxiv.org/abs/1711.07971
* Squeeze-and-Excitation (SE) - https://arxiv.org/abs/1709.01507
* Selective Kernel (SK) - (https://arxiv.org/abs/1903.06586
* Split (SPLAT) - https://arxiv.org/abs/2004.08955
* Shifted Window (SWIN) - https://arxiv.org/abs/2103.14030
Create README.md
2019-04-15 06:10:52 +08:00
Update README.md Add blurb about results csv
2019-07-05 01:52:03 +08:00
## Results
Update README and version
2023-02-07 15:43:14 +08:00
Model validation results can be found in the [results tables](results/README.md)
Add some info to README
2019-05-16 00:54:07 +08:00
Update README.md, documentation, add TOC, sponsors, and links to timmdocs and paperswithcode
2021-03-11 02:23:51 +08:00
## Getting Started (Documentation)
Update README.md Remove references to github.io docs that died with repo move
2023-02-25 06:43:31 +08:00
The official documentation can be found at https://huggingface.co/docs/hub/timm. Documentation contributions are welcome.
Update README.md
2022-09-08 01:40:58 +08:00
Update README.md
2022-02-03 01:15:20 +08:00
[Getting Started with PyTorch Image Models (timm): A Practitioner’s Guide](https://towardsdatascience.com/getting-started-with-pytorch-image-models-timm-a-practitioners-guide-4e77b4bf9055) by [Chris Hughes](https://github.com/Chris-hughes10) is an extensive blog post covering many aspects of `timm` in detail.
Update README.md Remove references to github.io docs that died with repo move
2023-02-25 06:43:31 +08:00
[timmdocs](http://timm.fast.ai/) is an alternate set of documentation for `timm`. A big thanks to [Aman Arora](https://github.com/amaarora) for his efforts creating timmdocs.
Add some info to README
2019-05-16 00:54:07 +08:00
Update README.md, documentation, add TOC, sponsors, and links to timmdocs and paperswithcode
2021-03-11 02:23:51 +08:00
[paperswithcode](https://paperswithcode.com/lib/timm) is a good resource for browsing the models within `timm`.
Update README.md
2020-08-13 14:26:59 +08:00
## Train, Validation, Inference Scripts
Fix typo
2023-03-09 10:42:54 +08:00
The root folder of the repository contains reference train, validation, and inference scripts that work with the included models and other features of this repository. They are adaptable for other datasets and use cases with a little hacking. See [documentation](https://huggingface.co/docs/timm/training_script).
Update README.md
2020-08-27 06:15:51 +08:00
## Awesome PyTorch Resources
README: fix simple typos
2021-05-26 04:42:44 +08:00
One of the greatest assets of PyTorch is the community and their contributions. A few of my favourite resources that pair well with the models and components here are listed below.
Update README.md
2020-08-27 06:15:51 +08:00
Update README.md and docs. Version bumped to 0.4.3
2021-02-11 06:46:07 +08:00
### Object Detection, Instance and Semantic Segmentation
* Detectron2 - https://github.com/facebookresearch/detectron2
* Segmentation Models (Semantic) - https://github.com/qubvel/segmentation_models.pytorch
* EfficientDet (Obj Det, Semantic soon) - https://github.com/rwightman/efficientdet-pytorch
Update README.md
2020-08-27 06:15:51 +08:00
### Computer Vision / Image Augmentation
* Albumentations - https://github.com/albumentations-team/albumentations
* Kornia - https://github.com/kornia/kornia
Add BiT references and knowledge distill links to readme/docs
2021-01-17 04:30:02 +08:00
### Knowledge Distillation
* RepDistiller - https://github.com/HobbitLong/RepDistiller
* torchdistill - https://github.com/yoshitomo-matsubara/torchdistill
Update README.md
2020-08-27 06:15:51 +08:00
### Metric Learning
* PyTorch Metric Learning - https://github.com/KevinMusgrave/pytorch-metric-learning
Update README.md and docs. Version bumped to 0.4.3
2021-02-11 06:46:07 +08:00
### Training / Frameworks
* fastai - https://github.com/fastai/fastai
Update README.md
2020-08-27 06:15:51 +08:00
## Licenses
### Code
README: fix simple typos
2021-05-26 04:42:44 +08:00
The code here is licensed Apache 2.0. I've taken care to make sure any third party code included or adapted has compatible (permissive) licenses such as MIT, BSD, etc. I've made an effort to avoid any GPL / LGPL conflicts. That said, it is your responsibility to ensure you comply with licenses here and conditions of any dependent licenses. Where applicable, I've linked the sources/references for various components in docstrings. If you think I've missed anything please create an issue.
Update README.md
2020-08-27 06:15:51 +08:00
### Pretrained Weights
Update README.md
2021-07-29 00:00:48 +08:00
So far all of the pretrained weights available here are pretrained on ImageNet with a select few that have some additional pretraining (see extra note below). ImageNet was released for non-commercial research purposes only (https://image-net.org/download). It's not clear what the implications of that are for the use of pretrained weights from that dataset. Any models I have trained with ImageNet are done for research purposes and one should assume that the original dataset license applies to the weights. It's best to seek legal advice if you intend to use the pretrained weights in a commercial product.
Update README.md
2020-08-27 06:15:51 +08:00
#### Pretrained on more than ImageNet
Several weights included or references here were pretrained with proprietary datasets that I do not have access to. These include the Facebook WSL, SSL, SWSL ResNe(Xt) and the Google Noisy Student EfficientNet models. The Facebook models have an explicit non-commercial license (CC-BY-NC 4.0, https://github.com/facebookresearch/semi-supervised-ImageNet1K-models, https://github.com/facebookresearch/WSL-Images). The Google models do not appear to have any restriction beyond the Apache 2.0 license (and ImageNet concerns). In either case, you should contact Facebook or Google with any questions.
Add citation info
2021-01-04 07:46:12 +08:00
## Citing
### BibTeX
Add color highlighting to BibTeX entry in README
2021-07-01 05:07:16 +08:00
```bibtex
Add citation info
2021-01-04 07:46:12 +08:00
@misc{rw2019timm,
author = {Ross Wightman},
title = {PyTorch Image Models},
year = {2019},
publisher = {GitHub},
journal = {GitHub repository},
doi = {10.5281/zenodo.4414861},
howpublished = {\url{https://github.com/rwightman/pytorch-image-models}}
}
```
### Latest DOI
[![DOI](https://zenodo.org/badge/168799526.svg)](https://zenodo.org/badge/latestdoi/168799526)