Repository files navigation

• What's New
• Introduction
• Models
• Features
• Results
• Getting Started (Documentation)
• Train, Validation, Inference Scripts
• Awesome PyTorch Resources
• Licenses
• Citing

❗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 totimm.layers:
  • from timm.models.layers import name will still work via deprecation mapping (but please transition totimm.layers).
  • import timm.models.layers.module orfrom timm.models.layers.module import name needs to be changed now.
• Builder, helper, non-model modules intimm.models have a_ prefix added, ietimm.models.helpers ->timm.models._helpers, there are temporary deprecation mapping files but those will be removed.
• All models now supportarchitecture.pretrained_tag naming (exresnet50.rsb_a1).
  • The pretrained_tag is the specific weight variant (different head) for the architecture.
  • Using onlyarchitecture defaults to the first weights in the default_cfgs for that model architecture.
  • 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.
• A number of models had their checkpoints remaped to match architecture changes needed to better supportfeatures_only=True, there arecheckpoint_filter_fn methods in any model module that was remapped. These can be passed totimm.models.load_checkpoint(..., filter_fn=timm.models.swin_transformer_v2.checkpoint_filter_fn) to remap your existing checkpoint.
• The Hugging Face Hub (https://huggingface.co/timm) is now the primary source fortimm weights. Model cards include link to papers, original source, license.
• Previous 0.6.x can be cloned from0.6.x branch or installed via pip with version.

• SigLIP image tower weights supported invision_transformer.py.
  • Great potential for fine-tune and downstream feature use.
• Experimental 'register' support in vit models as perVision Transformers Need Registers
• Updated RepViT with new weight release. Thankswangao
• Add patch resizing support (on pretrained weight load) to Swin models
• 0.9.8 release pending

• TinyViT added bySeeFun
• Fix EfficientViT (MIT) to use torch.autocast so it works back to PT 1.10
• 0.9.7 release

• Add dynamic img size support to models invision_transformer.py,vision_transformer_hybrid.py,deit.py, andeva.py w/o breaking backward compat.
  • Adddynamic_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).
  • Adddynamic_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 differentimg_size (interpolate pretrained embed weights once) on creation still works.
  • Existing method of changingpatch_size (resize pretrained patch_embed weights once) on creation still works.
  • Example validation cmdpython 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

• 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 (thankshttps://github.com/ChengpengChen)
  • GhostNetV2 -https://arxiv.org/abs/2211.12905 (thankshttps://github.com/yehuitang)
  • EfficientViT (MSRA) -https://arxiv.org/abs/2305.07027 (thankshttps://github.com/seefun)
  • EfficientViT (MIT) -https://arxiv.org/abs/2205.14756 (thankshttps://github.com/seefun)
• Add--reparam arg tobenchmark.py,onnx_export.py, andvalidate.py to trigger layer reparameterization / fusion for models with any one ofreparameterize(),switch_to_deploy() orfuse()
  • Including FastViT, MobileOne, RepGhostNet, EfficientViT (MSRA), RepViT, RepVGG, and LeViT
• Preparing 0.9.6 'back to school' release

• 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 resizepython 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

• Add GluonCV weights for HRNet w18_small and w18_small_v2. Converted bySeeFun
• Fixselecsls* 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

• Added timm trainedseresnextaa201d_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 bywangao
• I-JEPA ViT feature weights (no classifier) added bySeeFun
• SAM-ViT (segment anything) feature weights (no classifier) added bySeeFun
• Add support for alternative feat extraction methods and -ve indices to EfficientNet
• Add NAdamW optimizer
• Misc fixes

• timm 0.9 released, transition from 0.8.xdev releases

• Hugging Face Hub downloading is now default, 1132 models onhttps://huggingface.co/timm, 1163 weights intimm
• DINOv2 vit feature backbone weights added thanks toLeng 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 byFredo Guan
• Experimentalget_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 ifpretrained=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
• bitsandbytes (https://github.com/TimDettmers/bitsandbytes) optimizers added to factory, usebnb prefix, iebnbadam8bit
• Misc cleanup and fixes
• Final testing before switching to a 0.9 and bringingtimm out of pre-release state

• 97% oftimm 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.

• Gradient accumulation support added to train script and tested (--grad-accum-steps), thanksTaeksang Kim
• 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).

• 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 intodrop_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.

• ALL ResNet models pushed to Hugging Face Hub with multi-weight support
  • All pasttimm 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 inhttps://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

• Add first ConvNext-XXLarge CLIP -> IN-1k fine-tune and IN-12k intermediate fine-tunes for convnext-base/large CLIP models.

• 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 athttps://github.com/baaivision/EVA for more details on their work blending MIM, CLIP w/ many model, dataset, and train recipe tweaks.

• Multi-weight and HF hub for DeiT and MLP-Mixer based models

• 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 forswinv2_cr_*, and NHWC for all others) and spatial embedding outputs.
• FocalNet (fromhttps://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-tunedtimm 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 pretrained image tower weights for fine-tune & features (fine-tuning TBD) -- seemodel card
• Updateconvnext_xxlarge default LayerNorm eps to 1e-5 (for CLIP weights, improved stability)
• 0.8.15dev0

• Add 320x320convnext_large_mlp.clip_laion2b_ft_320 andconvnext_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

• 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) tovit_*,vit_relpos*,coatnet /maxxvit (to start)
• Lion optimizer (w/ multi-tensor option) added (https://arxiv.org/abs/2302.06675)
• gradient checkpointing works withfeatures_only=True

• New inference benchmark numbers added inresults 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
  • convnext_large_mlp.clip_laion2b_augreg_ft_in1k_384 - 87.9% @ 384x384
• Add DaViT models. Supportsfeatures_only=True. Adapted fromhttps://github.com/dingmyu/davit byFredo.
• 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). Supportsfeatures_only=True.
  • Minor updates to EfficientFormer.
  • Refactor LeViT models to stages, addfeatures_only=True support to newconv variants, weight remap required.
• Move ImageNet meta-data (synsets, indices) from/results totimm/data/_info.
• Add ImageNetInfo / DatasetInfo classes to provide labelling for various ImageNet classifier layouts intimm
  • Updateinference.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).

• 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 forrw base MaxViT and CoAtNet 1/2 models

• 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

• 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
  • train.py /imagenet --model vit_base_patch16_clip_224 --img-size 240 --amp --model-kwargs img_size=240 patch_size=12
• Cleanup some popular models to better support arg passthrough / merge with model configs, more to go.

• ConvNeXt-V2 models and weights added to existingconvnext.py
  • Paper:ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders
  • Reference impl:https://github.com/facebookresearch/ConvNeXt-V2 (NOTE: weights currently CC-BY-NC)

• Add FlexiViT models and weights fromhttps://github.com/google-research/big_vision (check out paper athttps://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
  • vit_medium_patch16_gap_384.in12k_ft_in1k - 85.5 @ 384x384
  • vit_medium_patch16_gap_256.in12k_ft_in1k - 84.5 @ 256x256
  • convnext_nano.in12k_ft_in1k - 82.9 @ 288x288

• Add 'EVA l' tovision_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

• Add 'EVA g', BEiT style ViT-g/14 model weights w/ both MIM pretrain and CLIP pretrain tobeit.py.
  • original source:https://github.com/baaivision/EVA
  • paper:https://arxiv.org/abs/2211.07636

• Pre-release (0.8.0dev0) of multi-weight support (model_arch.pretrained_tag). Install withpip install --pre timm
  • 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 from0.6.x branch
• 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

• Port of MaxViT Tensorflow Weights from official impl athttps://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

• 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

• More weights inmaxxvit series, incl first ConvNeXt block basedcoatnext andmaxxvit experiments:
  • coatnext_nano_rw_224 - 82.0 @ 224 (G) -- (uses ConvNeXt conv block, no BatchNorm)
  • maxxvit_rmlp_nano_rw_256 - 83.0 @ 256, 83.7 @ 320 (G) (uses ConvNeXt conv block, no BN)
  • maxvit_rmlp_small_rw_224 - 84.5 @ 224, 85.1 @ 320 (G)
  • maxxvit_rmlp_small_rw_256 - 84.6 @ 256, 84.9 @ 288 (G) -- could be trained better, hparams need tuning (uses ConvNeXt block, no BN)
  • coatnet_rmlp_2_rw_224 - 84.6 @ 224, 85 @ 320 (T)
  • NOTE: official MaxVit weights (in1k) have been released athttps://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.

• 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

• Hugging Facetimm docs home now exists, look for more here in the future
• Add BEiT-v2 weights for base and large 224x224 models fromhttps://github.com/microsoft/unilm/tree/master/beit2
• Add more weights inmaxxvit series incl apico (7.5M params, 1.9 GMACs), twotiny 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)

PyTorchImageModels (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.

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.

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.

• Aggregating Nested Transformers -https://arxiv.org/abs/2105.12723
• BEiT -https://arxiv.org/abs/2106.08254
• Big Transfer ResNetV2 (BiT) -https://arxiv.org/abs/1912.11370
• Bottleneck Transformers -https://arxiv.org/abs/2101.11605
• 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
• CoAtNet (Convolution and Attention) -https://arxiv.org/abs/2106.04803
• ConvNeXt -https://arxiv.org/abs/2201.03545
• ConvNeXt-V2 -http://arxiv.org/abs/2301.00808
• ConViT (Soft Convolutional Inductive Biases Vision Transformers)-https://arxiv.org/abs/2103.10697
• CspNet (Cross-Stage Partial Networks) -https://arxiv.org/abs/1911.11929
• DeiT -https://arxiv.org/abs/2012.12877
• DeiT-III -https://arxiv.org/pdf/2204.07118.pdf
• DenseNet -https://arxiv.org/abs/1608.06993
• DLA -https://arxiv.org/abs/1707.06484
• DPN (Dual-Path Network) -https://arxiv.org/abs/1707.01629
• EdgeNeXt -https://arxiv.org/abs/2206.10589
• EfficientFormer -https://arxiv.org/abs/2206.01191
• 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
  • EfficientNet V2 -https://arxiv.org/abs/2104.00298
  • 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
  • TinyNet -https://arxiv.org/abs/2010.14819
• EfficientViT (MIT) -https://arxiv.org/abs/2205.14756
• EfficientViT (MSRA) -https://arxiv.org/abs/2305.07027
• EVA -https://arxiv.org/abs/2211.07636
• EVA-02 -https://arxiv.org/abs/2303.11331
• FastViT -https://arxiv.org/abs/2303.14189
• FlexiViT -https://arxiv.org/abs/2212.08013
• FocalNet (Focal Modulation Networks) -https://arxiv.org/abs/2203.11926
• GCViT (Global Context Vision Transformer) -https://arxiv.org/abs/2206.09959
• GhostNet -https://arxiv.org/abs/1911.11907
• GhostNet-V2 -https://arxiv.org/abs/2211.12905
• gMLP -https://arxiv.org/abs/2105.08050
• GPU-Efficient Networks -https://arxiv.org/abs/2006.14090
• Halo Nets -https://arxiv.org/abs/2103.12731
• HRNet -https://arxiv.org/abs/1908.07919
• InceptionNeXt -https://arxiv.org/abs/2303.16900
• Inception-V3 -https://arxiv.org/abs/1512.00567
• Inception-ResNet-V2 and Inception-V4 -https://arxiv.org/abs/1602.07261
• Lambda Networks -https://arxiv.org/abs/2102.08602
• LeViT (Vision Transformer in ConvNet's Clothing) -https://arxiv.org/abs/2104.01136
• MaxViT (Multi-Axis Vision Transformer) -https://arxiv.org/abs/2204.01697
• MetaFormer (PoolFormer-v2, ConvFormer, CAFormer) -https://arxiv.org/abs/2210.13452
• MLP-Mixer -https://arxiv.org/abs/2105.01601
• MobileNet-V3 (MBConvNet w/ Efficient Head) -https://arxiv.org/abs/1905.02244
  • FBNet-V3 -https://arxiv.org/abs/2006.02049
  • HardCoRe-NAS -https://arxiv.org/abs/2102.11646
  • LCNet -https://arxiv.org/abs/2109.15099
• MobileOne -https://arxiv.org/abs/2206.04040
• MobileViT -https://arxiv.org/abs/2110.02178
• MobileViT-V2 -https://arxiv.org/abs/2206.02680
• MViT-V2 (Improved Multiscale Vision Transformer) -https://arxiv.org/abs/2112.01526
• NASNet-A -https://arxiv.org/abs/1707.07012
• NesT -https://arxiv.org/abs/2105.12723
• NFNet-F -https://arxiv.org/abs/2102.06171
• NF-RegNet / NF-ResNet -https://arxiv.org/abs/2101.08692
• PNasNet -https://arxiv.org/abs/1712.00559
• PoolFormer (MetaFormer) -https://arxiv.org/abs/2111.11418
• Pooling-based Vision Transformer (PiT) -https://arxiv.org/abs/2103.16302
• PVT-V2 (Improved Pyramid Vision Transformer) -https://arxiv.org/abs/2106.13797
• RegNet -https://arxiv.org/abs/2003.13678
• RegNetZ -https://arxiv.org/abs/2103.06877
• RepVGG -https://arxiv.org/abs/2101.03697
• RepGhostNet -https://arxiv.org/abs/2211.06088
• RepViT -https://arxiv.org/abs/2307.09283
• ResMLP -https://arxiv.org/abs/2105.03404
• 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
  • ResNet-RS -https://arxiv.org/abs/2103.07579
• 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
• Sequencer2D -https://arxiv.org/abs/2205.01972
• Swin S3 (AutoFormerV2) -https://arxiv.org/abs/2111.14725
• Swin Transformer -https://arxiv.org/abs/2103.14030
• Swin Transformer V2 -https://arxiv.org/abs/2111.09883
• Transformer-iN-Transformer (TNT) -https://arxiv.org/abs/2103.00112
• TResNet -https://arxiv.org/abs/2003.13630
• Twins (Spatial Attention in Vision Transformers) -https://arxiv.org/pdf/2104.13840.pdf
• Visformer -https://arxiv.org/abs/2104.12533
• Vision Transformer -https://arxiv.org/abs/2010.11929
• VOLO (Vision Outlooker) -https://arxiv.org/abs/2106.13112
• VovNet V2 and V1 -https://arxiv.org/abs/1911.06667
• 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
• XCiT (Cross-Covariance Image Transformers) -https://arxiv.org/abs/2106.09681

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:

• All models have a common default configuration interface and API for
  • accessing/changing the classifier -get_classifier andreset_classifier
  • doing a forward pass on just the features -forward_features (seedocumentation)
  • these makes it easy to write consistent network wrappers that work with any of the models
• All models support multi-scale feature map extraction (feature pyramids) via create_model (seedocumentation)
  • create_model(name, features_only=True, out_indices=..., output_stride=...)
  • out_indices creation arg specifies which feature maps to return, these indices are 0 based and generally correspond to theC(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.
  • feature map channel counts, reduction level (stride) can be queried AFTER model creation via the.feature_info member
• All models have a consistent pretrained weight loader that adapts last linear if necessary, and from 3 to 1 channel input if desired
• High performancereference training, validation, and inference scripts that work in several process/GPU modes:
  • 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.
• 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)
• Learning rate schedulers
  • Ideas adopted from
    • AllenNLP schedulers
    • FAIRseq lr_scheduler
    • SGDR: Stochastic Gradient Descent with Warm Restarts (https://arxiv.org/abs/1608.03983)
  • Schedulers includestep,cosine w/ restarts,tanh w/ restarts,plateau
• Optimizers:
  • rmsprop_tf adapted from PyTorch RMSProp by myself. Reproduces much improved Tensorflow RMSProp behaviour.
  • radam byLiyuan Liu (https://arxiv.org/abs/1908.03265)
  • novograd byMasashi Kimura (https://arxiv.org/abs/1905.11286)
  • lookahead adapted from impl byLiam (https://arxiv.org/abs/1907.08610)
  • fused<name> optimizers by name withNVIDIA Apex installed
  • adamp andsgdp byNaver ClovAI (https://arxiv.org/abs/2006.08217)
  • adafactor adapted fromFAIRSeq impl (https://arxiv.org/abs/1804.04235)
  • adahessian byDavid Samuel (https://arxiv.org/abs/2006.00719)
• Random Erasing fromZhun Zhong (https://arxiv.org/abs/1708.04896)
• Mixup (https://arxiv.org/abs/1710.09412)
• CutMix (https://arxiv.org/abs/1905.04899)
• 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)
• 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
• DropPath aka "Stochastic Depth" (https://arxiv.org/abs/1603.09382)
• DropBlock (https://arxiv.org/abs/1810.12890)
• Blur Pooling (https://arxiv.org/abs/1904.11486)
• Space-to-Depth bymrT23 (https://arxiv.org/abs/1801.04590) -- original paper?
• Adaptive Gradient Clipping (https://arxiv.org/abs/2102.06171,https://github.com/deepmind/deepmind-research/tree/master/nfnets)
• 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

Model validation results can be found in theresults tables

The official documentation can be found athttps://huggingface.co/docs/hub/timm. Documentation contributions are welcome.

Getting Started with PyTorch Image Models (timm): A Practitioner’s Guide byChris Hughes is an extensive blog post covering many aspects oftimm in detail.

timmdocs is an alternate set of documentation fortimm. A big thanks toAman Arora for his efforts creating timmdocs.

paperswithcode is a good resource for browsing the models withintimm.

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. Seedocumentation.

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.

• 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

• Albumentations -https://github.com/albumentations-team/albumentations
• Kornia -https://github.com/kornia/kornia

• RepDistiller -https://github.com/HobbitLong/RepDistiller
• torchdistill -https://github.com/yoshitomo-matsubara/torchdistill

• PyTorch Metric Learning -https://github.com/KevinMusgrave/pytorch-metric-learning

• fastai -https://github.com/fastai/fastai

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.

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.

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.

@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}}}