64667bfa0e
Add 'gigantic' vit clip variant for feature extraction and future fine-tuning
2 years ago
60ebb6cefa
Re-order vit pretrained entries for more sensible default weights (no .tag specified)
2 years ago
e861b74cf8
Pass through --model-kwargs (and --opt-kwargs for train) from command line through to model __init__. Update some models to improve arg overlay. Cleanup along the way.
2 years ago
8ece53e194
Switch BEiT to HF hub weights
2 years ago
9a51e4ea2e
Add FlexiViT models and weights, refactoring, push more weights
...
* push all vision_transformer*.py weights to HF hub
* finalize more pretrained tags for pushed weights
* refactor pos_embed files and module locations, move some pos embed modules to layers
* tweak hf hub helpers to aid bulk uploading and updating
2 years ago
6a01101905
Update efficientnet.py and convnext.py to multi-weight, add ImageNet-12k pretrained EfficientNet-B5 and ConvNeXt-Nano.
2 years ago
d5e7d6b27e
Merge remote-tracking branch 'origin/main' into refactor-imports
2 years ago
7c4ed4d5a4
Add EVA-large models
2 years ago
927f031293
Major module / path restructure, timm.models.layers -> timm.layers, add _ prefix to all non model modules in timm.models
2 years ago
3785c234d7
Remove clip vit models that won't be ft and comment two that aren't uploaded yet
2 years ago
755570e2d6
Rename _pretrained.py -> pretrained.py, not feasible to change the other files to same scheme without breaking uses
2 years ago
72cfa57761
Add ported Tensorflow MaxVit weights. Add a few more CLIP ViT fine-tunes. Tweak some model tag names. Improve model tag name sorting. Update HF hub push config layout.
2 years ago
4d5c395160
MaxVit, ViT, ConvNeXt, and EfficientNet-v2 updates
...
* Add support for TF weights and modelling specifics to MaxVit (testing ported weights)
* More fine-tuned CLIP ViT configs
* ConvNeXt and MaxVit updated to new pretrained cfgs use
* EfficientNetV2, MaxVit and ConvNeXt high res models use squash crop/resize
2 years ago
b2b6285af7
Add two more FT clip weights
2 years ago
5895056dc4
Add openai b32 ft
2 years ago
9dea5143d5
Adding more clip ft variants
2 years ago
444dcba4ad
CLIP B16 12k weights added
2 years ago
dff4717cbf
Add clip b16 384x384 finetunes
2 years ago
883fa2eeaa
Add fine-tuned B/16 224x224 in1k clip models
2 years ago
9a3d2ac2d5
Add latest CLIP ViT fine-tune pretrained configs / model entrypt updates
2 years ago
def68befa7
Updating vit model defs for mult-weight support trial (vit first). Prepping for CLIP (laion2b and openai) fine-tuned weights.
2 years ago
0dadb4a6e9
Initial multi-weight support, handled so old pretraing config handling co-exists with new tags.
2 years ago
8fda68aff6
Fix repo id bug
...
This to fix this issue #1482
2 years ago
1199c5a1a4
clip_laion2b models need 1e-5 eps for LayerNorm
2 years ago
e069249a2d
Add hf hub entries for laion2b clip models, add huggingface_hub dependency, update some setup/reqs, torch >= 1.7
2 years ago
9d65557be3
Fix errant import
2 years ago
9709dbaaa9
Adding support for fine-tune CLIP LAION-2B image tower weights for B/32, L/14, H/14 and g/14. Still WIP
2 years ago
e11efa872d
Update a bunch of weights with external links to timm release assets. Fixes issue with *aliyuncs.com returning forbidden. Did pickle scan / verify and re-hash. Add TresNet-V2-L weights.
2 years ago
0b64117592
Take `no_emb_class` into account when calling `resize_pos_embed`
2 years ago
1b278136c3
Change models with mean 0,0,0 std 1,1,1 from int to float for consistency as mentioned in #1355
2 years ago
a8e34051c1
Unbreak gamma remap impacting beit checkpoint load, version bump to 0.6.4
2 years ago
7d4b3807d5
Support DeiT-3 (Revenge of the ViT) checkpoints. Add non-overlapping (w/ class token) pos-embed support to vit.
2 years ago
7d657d2ef4
Improve resolve_pretrained_cfg behaviour when no cfg exists, warn instead of crash. Improve usability ex #1311
2 years ago
f5ca4141f7
Adjust arg order for recent vit model args, add a few comments
3 years ago
41dc49a337
Vision Transformer refactoring and Rel Pos impl
3 years ago
1618527098
Add layer scale and parallel blocks to vision_transformer
3 years ago
0862e6ebae
Fix correctness of some group matching regex (no impact on result), some formatting, missed forward_head for resnet
3 years ago
372ad5fa0d
Significant model refactor and additions:
...
* All models updated with revised foward_features / forward_head interface
* Vision transformer and MLP based models consistently output sequence from forward_features (pooling or token selection considered part of 'head')
* WIP param grouping interface to allow consistent grouping of parameters for layer-wise decay across all model types
* Add gradient checkpointing support to a significant % of models, especially popular architectures
* Formatting and interface consistency improvements across models
* layer-wise LR decay impl part of optimizer factory w/ scale support in scheduler
* Poolformer and Volo architectures added
3 years ago
5f81d4de23
Move DeiT to own file, vit getting crowded. Working towards fixing #1029 , make pooling interface for transformers and mlp closer to convnets. Still working through some details...
3 years ago
95cfc9b3e8
Merge remote-tracking branch 'origin/master' into norm_norm_norm
3 years ago
abc9ba2544
Transitioning default_cfg -> pretrained_cfg. Improving handling of pretrained_cfg source (HF-Hub, files, timm config, etc). Checkpoint handling tweaks.
3 years ago
07379c6d5d
Add vit_base2_patch32_256 for a model between base_patch16 and patch32 with a slightly larger img size and width
3 years ago
010b486590
Add Dino pretrained weights (no head) for vit models. Add support to tests and helpers for models w/ no classifier (num_classes=0 in pretrained cfg)
3 years ago
e967c72875
Update REAMDE.md. Sneak in g/G (giant / gigantic?) ViT defs from scaling paper
3 years ago
656757d26b
Fix MobileNetV2 head conv size for multiplier < 1.0. Add some missing modification copyrights, fix starting date of some old ones.
3 years ago
5220711d87
Added B/8 models to ViT.
3 years ago
f805ba86d9
use .unbind instead of explicitly listing the indices
3 years ago
78933122c9
Fix silly typo
3 years ago
708d87a813
Fix ViT SAM weight compat as weights at URL changed to not use repr layer. Fix #825 . Tweak optim test.
3 years ago
20b2d4b69d
Use bicubic interpolation in resize_pos_embed()
3 years ago
Ross Wightman