Ross Wightman
5dce710101
Add vit_little in12k + in12k-ft-in1k weights
2024-05-27 14:56:03 -07:00
Ross Wightman
3c0283f9ef
Fix reparameterize for NextViT. Fix #2187
2024-05-27 14:48:58 -07:00
Ross Wightman
4ff7c25766
Pass layer_scale_init_value to Mnv3Features module
2024-05-24 16:44:50 -07:00
Ross Wightman
a12b72b5c4
Fix missing head_norm arg pop for feature model
2024-05-24 15:50:34 -07:00
Ross Wightman
7fe96e7a92
More MobileNet-v4 fixes
...
* missed final norm after post pooling 1x1 PW head conv
* improve repr of model by flipping a few modules to None when not used, nn.Sequential for MultiQueryAttention query/key/value/output
* allow layer scaling to be enabled/disabled at model variant level, conv variants don't use it
2024-05-24 15:09:29 -07:00
Ross Wightman
28d76a97db
Mixed up kernel size for last blocks in mnv4-conv-small
2024-05-24 11:50:42 -07:00
Ross Wightman
0c6a69e7ef
Add comments to MNV4 model defs with block variants
2024-05-23 15:54:05 -07:00
Ross Wightman
cb33956b20
Fix some mistakes in mnv4 model defs
2024-05-23 14:24:32 -07:00
Ross Wightman
cee79dada0
Merge remote-tracking branch 'origin/main' into efficientnet_x
2024-05-23 11:01:39 -07:00
Ross Wightman
6a8bb03330
Initial MobileNetV4 pass
2024-05-23 10:49:18 -07:00
Ross Wightman
e748805be3
Add regex matching support to AttentionExtract. Add return_dict support to graph extractors and use returned output in AttentionExtractor
2024-05-22 14:33:39 -07:00
Ross Wightman
84cb225ecb
Add in12k + 12k_ft_in1k vit_medium weights
2024-05-20 15:52:46 -07:00
Beckschen
7a2ad6bce1
Add link to model weights on Hugging Face
2024-05-17 06:51:35 -04:00
Beckschen
530fb49e7e
Add link to model weights on Hugging Face
2024-05-17 06:48:59 -04:00
Ross Wightman
cd0e7b11ff
Merge pull request #2180 from yvonwin/main
...
Remove a duplicate function in mobilenetv3.py
2024-05-15 07:54:17 -07:00
Ross Wightman
83aee5c28c
Add explicit GAP (avg pool) variants of other SigLIP models.
2024-05-15 07:53:19 -07:00
yvonwin
58f2f79b04
Remove a duplicate function in mobilenetv3.py: _gen_lcnet
is repeated in mobilenetv3.py.Remove the duplicate code.
2024-05-15 17:59:34 +08:00
Ross Wightman
7b3b11b63f
Support loading of paligemma weights into GAP variants of SigLIP ViT. Minor tweak to npz loading for packed transformer weights.
2024-05-14 15:44:37 -07:00
Beckschen
df304ffbf2
the dataclass init needs to use the default factory pattern, according to Ross
2024-05-14 15:10:05 -04:00
Ross Wightman
a69863ad61
Merge pull request #2156 from huggingface/hiera
...
WIP Hiera implementation.
2024-05-13 14:58:12 -07:00
Ross Wightman
f7aa0a1a71
Add missing vit_wee weight
2024-05-13 12:05:47 -07:00
Ross Wightman
7a4e987b9f
Hiera weights on hub
2024-05-13 11:43:22 -07:00
Ross Wightman
23f09af08e
Merge branch 'main' into efficientnet_x
2024-05-12 21:31:08 -07:00
Ross Wightman
c838c4233f
Add typing to reset_classifier() on other models
2024-05-12 11:12:00 -07:00
Ross Wightman
3e03b2bf3f
Fix a few more hiera API issues
2024-05-12 11:11:45 -07:00
Ross Wightman
211d18d8ac
Move norm & pool into Hiera ClassifierHead. Misc fixes, update features_intermediate() naming
2024-05-11 23:37:35 -07:00
Ross Wightman
2ca45a4ff5
Merge remote-tracking branch 'upstream/main' into hiera
2024-05-11 15:43:05 -07:00
Ross Wightman
1d3ab176bc
Remove debug / staging code
2024-05-10 22:16:34 -07:00
Ross Wightman
aa4d06a11c
sbb vit weights on hub, testing
2024-05-10 17:15:01 -07:00
Ross Wightman
3582ca499e
Prepping weight push, benchmarking.
2024-05-10 14:14:06 -07:00
Ross Wightman
2bfa5e5d74
Remove JIT activations, take jit out of ME activations. Remove other instances of torch.jit.script. Breaks torch.compile and is much less performant. Remove SpaceToDepthModule
2024-05-06 16:32:49 -07:00
Beckschen
99d4c7d202
add ViTamin models
2024-05-05 02:50:14 -04:00
Ross Wightman
07535f408a
Add AttentionExtract helper module
2024-05-04 14:10:00 -07:00
Ross Wightman
45b7ae8029
forward_intermediates() support for byob/byoanet models
2024-05-04 14:06:52 -07:00
Ross Wightman
c4b8897e9e
attention -> attn in davit for model consistency
2024-05-04 14:06:11 -07:00
Ross Wightman
cb57a96862
Fix early stop for efficientnet/mobilenetv3 fwd inter. Fix indices typing for all fwd inter.
2024-05-04 10:21:58 -07:00
Ross Wightman
01dd01b70e
forward_intermediates() for MlpMixer models and RegNet.
2024-05-04 10:21:03 -07:00
Ross Wightman
f8979d4f50
Comment out time local files while testing new vit weights
2024-05-03 20:26:56 -07:00
Ross Wightman
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
Ross Wightman
d6da4fb01e
Add forward_intermediates() to efficientnet / mobilenetv3 based models as an exercise.
2024-05-02 14:19:16 -07:00
Ross Wightman
c22efb9765
Add wee & little vits for some experiments
2024-05-02 10:51:35 -07:00
Ross Wightman
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
user-miner1
740f4983b3
Assert messages added
2024-04-30 10:10:02 +03:00
Ross Wightman
c6db4043cd
Update forward_intermediates for hiera to have its own fwd impl w/ early stopping. Remove return_intermediates bool from forward(). Still an fx issue with None mask arg :(
2024-04-29 17:23:37 -07:00
Ross Wightman
9b9a356a04
Add forward_intermediates support for xcit, cait, and volo.
2024-04-29 16:30:45 -07:00
Ross Wightman
ef147fd2fb
Add forward_intermediates API to Hiera for features_only=True support
2024-04-21 11:30:41 -07:00
Ross Wightman
d88bed6535
Bit more Hiera fiddling
2024-04-21 09:36:57 -07:00
Ross Wightman
8a54d2a930
WIP Hiera implementation. Fix #2083 . Trying to get image size adaptation to work.
2024-04-20 09:47:17 -07:00
Ross Wightman
d6b95520f1
Merge pull request #2136 from huggingface/vit_features_only
...
Exploring vit features_only via new forward_intermediates() API, inspired by #2131
2024-04-11 08:38:20 -07:00
Ross Wightman
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