Ross Wightman
8ece53e194
Switch BEiT to HF hub weights
2022-12-22 21:43:04 -08:00
Ross Wightman
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
2022-12-22 17:23:09 -08:00
Ross Wightman
6a01101905
Update efficientnet.py and convnext.py to multi-weight, add ImageNet-12k pretrained EfficientNet-B5 and ConvNeXt-Nano.
2022-12-14 20:33:23 -08:00
Ross Wightman
d5e7d6b27e
Merge remote-tracking branch 'origin/main' into refactor-imports
2022-12-09 14:49:44 -08:00
Ross Wightman
7c4ed4d5a4
Add EVA-large models
2022-12-08 16:21:30 -08:00
Ross Wightman
927f031293
Major module / path restructure, timm.models.layers -> timm.layers, add _ prefix to all non model modules in timm.models
2022-12-06 15:00:06 -08:00
Ross Wightman
3785c234d7
Remove clip vit models that won't be ft and comment two that aren't uploaded yet
2022-12-05 10:21:34 -08:00
Ross Wightman
755570e2d6
Rename _pretrained.py -> pretrained.py, not feasible to change the other files to same scheme without breaking uses
2022-12-05 10:21:34 -08:00
Ross Wightman
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.
2022-12-05 10:21:34 -08:00
Ross Wightman
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
2022-12-05 10:21:34 -08:00
Ross Wightman
b2b6285af7
Add two more FT clip weights
2022-12-05 10:21:34 -08:00
Ross Wightman
5895056dc4
Add openai b32 ft
2022-12-05 10:21:34 -08:00
Ross Wightman
9dea5143d5
Adding more clip ft variants
2022-12-05 10:21:34 -08:00
Ross Wightman
444dcba4ad
CLIP B16 12k weights added
2022-12-05 10:21:34 -08:00
Ross Wightman
dff4717cbf
Add clip b16 384x384 finetunes
2022-12-05 10:21:34 -08:00
Ross Wightman
883fa2eeaa
Add fine-tuned B/16 224x224 in1k clip models
2022-12-05 10:21:34 -08:00
Ross Wightman
9a3d2ac2d5
Add latest CLIP ViT fine-tune pretrained configs / model entrypt updates
2022-12-05 10:21:34 -08:00
Ross Wightman
def68befa7
Updating vit model defs for mult-weight support trial (vit first). Prepping for CLIP (laion2b and openai) fine-tuned weights.
2022-12-05 10:21:34 -08:00
Ross Wightman
0dadb4a6e9
Initial multi-weight support, handled so old pretraing config handling co-exists with new tags.
2022-12-05 10:21:34 -08:00
Mohamed Rashad
8fda68aff6
Fix repo id bug
...
This to fix this issue #1482
2022-10-05 16:26:06 +02:00
Ross Wightman
1199c5a1a4
clip_laion2b models need 1e-5 eps for LayerNorm
2022-09-25 10:36:54 -07:00
Ross Wightman
e069249a2d
Add hf hub entries for laion2b clip models, add huggingface_hub dependency, update some setup/reqs, torch >= 1.7
2022-09-16 21:39:05 -07:00
Ross Wightman
9d65557be3
Fix errant import
2022-09-15 17:47:23 -07:00
Ross Wightman
9709dbaaa9
Adding support for fine-tune CLIP LAION-2B image tower weights for B/32, L/14, H/14 and g/14. Still WIP
2022-09-15 17:25:59 -07:00
Ross Wightman
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.
2022-09-13 16:35:26 -07:00
Ceshine Lee
0b64117592
Take no_emb_class into account when calling resize_pos_embed
2022-07-24 19:11:45 +08:00
Ross Wightman
1b278136c3
Change models with mean 0,0,0 std 1,1,1 from int to float for consistency as mentioned in #1355
2022-07-21 17:36:15 -07:00
Ross Wightman
a8e34051c1
Unbreak gamma remap impacting beit checkpoint load, version bump to 0.6.4
2022-07-07 23:07:43 -07:00
Ross Wightman
7d4b3807d5
Support DeiT-3 (Revenge of the ViT) checkpoints. Add non-overlapping (w/ class token) pos-embed support to vit.
2022-07-04 22:25:22 -07:00
Ross Wightman
7d657d2ef4
Improve resolve_pretrained_cfg behaviour when no cfg exists, warn instead of crash. Improve usability ex #1311
2022-06-24 14:55:25 -07:00
Ross Wightman
f5ca4141f7
Adjust arg order for recent vit model args, add a few comments
2022-05-02 22:41:38 -07:00
Ross Wightman
41dc49a337
Vision Transformer refactoring and Rel Pos impl
2022-05-02 15:37:39 -07:00
Ross Wightman
1618527098
Add layer scale and parallel blocks to vision_transformer
2022-03-23 16:09:07 -07:00
Ross Wightman
0862e6ebae
Fix correctness of some group matching regex (no impact on result), some formatting, missed forward_head for resnet
2022-03-19 14:58:54 -07:00
Ross Wightman
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
2022-02-28 13:56:23 -08:00
Ross Wightman
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...
2022-01-26 22:53:57 -08:00
Ross Wightman
95cfc9b3e8
Merge remote-tracking branch 'origin/master' into norm_norm_norm
2022-01-25 22:20:45 -08:00
Ross Wightman
abc9ba2544
Transitioning default_cfg -> pretrained_cfg. Improving handling of pretrained_cfg source (HF-Hub, files, timm config, etc). Checkpoint handling tweaks.
2022-01-25 21:54:13 -08:00
Ross Wightman
07379c6d5d
Add vit_base2_patch32_256 for a model between base_patch16 and patch32 with a slightly larger img size and width
2022-01-24 14:46:47 -08:00
Ross Wightman
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)
2022-01-17 12:20:02 -08:00
Ross Wightman
e967c72875
Update REAMDE.md. Sneak in g/G (giant / gigantic?) ViT defs from scaling paper
2022-01-14 16:28:27 -08:00
Ross Wightman
656757d26b
Fix MobileNetV2 head conv size for multiplier < 1.0. Add some missing modification copyrights, fix starting date of some old ones.
2022-01-14 16:28:27 -08:00
Martins Bruveris
5220711d87
Added B/8 models to ViT.
2021-11-14 11:01:48 +00:00
Thomas Viehmann
f805ba86d9
use .unbind instead of explicitly listing the indices
2021-10-24 21:08:47 +02:00
Ross Wightman
78933122c9
Fix silly typo
2021-08-27 09:22:20 -07:00
Ross Wightman
708d87a813
Fix ViT SAM weight compat as weights at URL changed to not use repr layer. Fix #825 . Tweak optim test.
2021-08-27 09:20:13 -07:00
Ying Jin
20b2d4b69d
Use bicubic interpolation in resize_pos_embed()
2021-07-12 10:38:31 -07:00
Ross Wightman
6d8272e92c
Add SAM pretrained model defs/weights for ViT B16 and B32 models.
2021-07-08 11:51:12 -07:00
Ross Wightman
85f894e03d
Fix ViT in21k representation (pre_logits) layer handling across old and new npz checkpoints
2021-06-23 10:38:34 -07:00
Ross Wightman
b41cffaa93
Fix a few issues loading pretrained vit/bit npz weights w/ num_classes=0 __init__ arg. Missed a few other small classifier handling detail on Mlp, GhostNet, Levit. Should fix #713
2021-06-22 23:16:05 -07:00
