2b251fb291
Wrap torch checkpoint() fn to default use_reentrant flag to False and allow env var override
2025-01-06 11:28:39 -08:00
cec70b6779
Merge pull request #2225 from huggingface/small_things
...
Small things
2024-07-25 20:29:13 -07:00
d2240745d3
Fix issue where feature out_indices out of order after wrapping with FeatureGetterNet due to use of set()
2024-07-22 13:33:30 -07:00
f81b094aaa
Add 'qkv_bias_separate' flag for EVA/beit/swinv2 attn modules to allow an override for easy quantization wrappers. Fix #2098
2024-07-08 13:48:38 -07:00
a5a2ad2e48
Fix consistency, testing for forward_head w/ pre_logits, reset_classifier, models with pre_logits size != unpooled feature size
...
* add test that model supports forward_head(x, pre_logits=True)
* add head_hidden_size attr to all models and set differently from num_features attr when head has hidden layers
* test forward_features() feat dim == model.num_features and pre_logits feat dim == self.head_hidden_size
* more consistency in reset_classifier signature, add typing
* asserts in some heads where pooling cannot be disabled
Fix #2194
2024-06-07 13:53:00 -07:00
88a1006e02
checkpoint filter fns with consistent name, add mobileclip-b pretrained cfgs
2024-06-06 12:38:52 -07:00
c838c4233f
Add typing to reset_classifier() on other models
2024-05-12 11:12:00 -07:00
c719f7eb86
More forward_intermediates() updates
...
* add convnext, resnet, efficientformer, levit support
* remove kwargs only for fn so that torchscript isn't broken for all :(
* use reset_classifier() consistently in prune
2024-05-03 16:22:32 -07:00
67332fce24
Add features_intermediate() support to coatnet, maxvit, swin* models. Refine feature interface. Start prep of new vit weights.
2024-04-30 16:56:33 -07:00
4b2565e4cb
More forward_intermediates() / FeatureGetterNet work
...
* include relpos vit
* refactor reduction / size calcs so hybrid vits work and dynamic_img_size works
* fix -ve feature indices when pruning
* fix mvitv2 w/ class token
* refine naming
* add tests
2024-04-10 15:11:34 -07:00
679daef76a
More forward_intermediates() & features_only work
...
* forward_intermediates() added to beit, deit, eva, mvitv2, twins, vit, vit_sam
* add features_only to forward intermediates to allow just intermediate features
* fix #2060
* fix #1374
* fix #657
2024-04-09 21:29:16 -07:00
88889de923
Fix meshgrid deprecation warnings and backward compat with explicit 'ndgrid' and 'meshgrid' fn w/o indexing arg
2024-01-27 13:48:33 -08:00
884ef88818
fix all SDPA dropouts
2023-10-05 08:58:41 -07:00
c153cd4a3e
Add more advanced interpolation method from BEiT and support non-square window & image size adaptation for
...
* beit/beit-v2
* maxxvit/coatnet
* swin transformer
And non-square windows for swin-v2
2023-08-08 16:41:16 -07:00
6e8c53d0d3
Comment out beit url, no longer valid as now require long query string, leave for reference, must use HF hub now.
2023-08-03 10:00:46 -07:00
f744bda994
use torch.jit.Final instead of Final for beit, eva
2023-05-12 09:12:14 -07:00
21e57c0b9e
Add missing beitv2 in1k -> in1k models
2023-05-08 17:03:51 -07:00
e4e43190ce
Add typing to all model entrypoint fns, add old cache check env var to builder
2023-05-08 08:52:38 -07:00
965d0a2d36
fast_attn -> fused_attn, implement global config for enable/disable fused_attn, add to more models. vit clip openai 336 weights.
2023-04-10 12:04:33 -07:00
4d135421a3
Implement patch dropout for eva / vision_transformer, refactor / improve consistency of dropout args across all vit based models
2023-04-07 20:27:23 -07:00
1bb3989b61
Improve kwarg passthrough for swin, vit, deit, beit, eva
2023-04-05 21:37:16 -07:00
ac67098147
Add final attr for fast_attn on beit / eva
2023-03-28 08:40:40 -07:00
3863d63516
Adding EVA02 weights and model defs, move beit based eva_giant to same eva.py file. Cleanup rotary pos, add lang oriented freq bands to be compat with eva design choice. Fix #1738
2023-03-27 17:16:07 -07:00
572f05096a
Swin and FocalNet weights on HF hub. Add model deprecation functionality w/ some registry tweaks.
2023-03-18 14:55:09 -07:00
8ece53e194
Switch BEiT to HF hub weights
2022-12-22 21:43:04 -08:00
d5e7d6b27e
Merge remote-tracking branch 'origin/main' into refactor-imports
2022-12-09 14:49:44 -08:00
98047ef5e3
Add EVA FT results, hopefully fix BEiT test failures
2022-12-07 08:54:06 -08:00
3cc4d7a894
Fix missing register for 224 eva model
2022-12-07 08:54:06 -08:00
eba07b0de7
Add eva models to beit.py
2022-12-07 08:54:06 -08:00
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
c8ab747bf4
BEiT-V2 checkpoints didn't remove 'module' from weights, adapt checkpoint filter
2022-09-13 17:56:49 -07:00
914544fc81
Add beitv2 224x224 checkpoints from https://github.com/microsoft/unilm/tree/master/beit2
2022-09-06 20:25:18 -07:00
909705e7ff
Remove some redundant requires_grad=True from nn.Parameter in third party code
2022-07-20 12:37:41 -07:00
09e9f3defb
migrate azure blob for beit checkpoints
...
## Motivation
We are going to use a new blob account to store the checkpoints.
## Modification
Modify the azure blob storage URLs for BEiT checkpoints.
2022-04-23 13:02:29 +08:00
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
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
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
f2006b2437
Cleanup qkv_bias cat in beit model so it can be traced
2021-11-18 21:25:00 -08:00
f805ba86d9
use .unbind instead of explicitly listing the indices
2021-10-24 21:08:47 +02:00
1c9284c640
Add BeiT 'finetuned' 1k weights and pretrained 22k weights, pretraining specific (masked) model excluded for now
2021-09-13 16:38:23 -07:00
Ross Wightman
Ross Wightman
Yassine
Li Dong
Thomas Viehmann