Hugo Touvron
d0ea707976
Update cait_models.py
2021-12-07 12:40:47 +01:00
Hugo Touvron
5a1eda0744
Update README_resmlp.md
2021-12-07 09:38:15 +01:00
Hugo Touvron
6e33ce7d91
Update README_resmlp.md
2021-12-07 09:37:48 +01:00
Hugo Touvron
b086ebc174
Update README_cait.md
2021-12-07 09:37:07 +01:00
Hugo Touvron
d8d8837124
Update README.md
2021-12-07 09:36:57 +01:00
Hugo Touvron
e8a2d2d5ea
Update README_cait.md
2021-12-07 09:30:27 +01:00
Hugo Touvron
06e80acefe
add ResMLP and CaiT models
2021-12-07 09:15:17 +01:00
Hugo Touvron
f512394620
Merge pull request #118 from developer0hye/patch-1
...
save best checkpoint
2021-11-21 15:24:12 +01:00
Hugo Touvron
8469882667
Merge pull request #91 from maxwell-aladago/main
...
BUG Fix: ensure the right loss function is selected when mixtup or cu…
2021-11-04 09:59:04 +01:00
Hugo Touvron
0c85644702
Merge pull request #123 from lmk123568/patch-1
...
fix bug - add droppath in cait
2021-11-04 09:55:09 +01:00
Mike
b61ce9afc5
fix bug - add droppath in cait
2021-11-04 16:03:11 +08:00
Hugo Touvron
ae4dba9b45
Add comments in the code
2021-10-08 09:41:34 +02:00
Yonghye Kwon
fdafaf0b8c
save best checkpoint
2021-10-01 17:12:25 +09:00
rv
806fb71d37
Update README.md
2021-08-30 12:26:40 +02:00
rv
f1fc1ff5b9
Update README.md
...
Update bibtex entry: provide official ICML reference instead of arxiv
2021-08-30 12:26:02 +02:00
Hugo Touvron
e6b10b554d
Update resmlp_models.py
2021-06-18 22:36:35 +02:00
Hugo Touvron
31b3d676b3
Update resmlp_models.py
2021-06-18 20:59:28 +02:00
Hugo Touvron
6fa7ef60b4
Update resmlp_models.py
2021-06-10 17:59:43 +02:00
Hugo Touvron
37cf44c581
Update README.md
2021-06-10 17:58:27 +02:00
Hugo Touvron
bfe21d2671
Add files via upload
2021-06-10 17:57:55 +02:00
Hugo Touvron
5b9e31db23
add resmlp
2021-06-10 17:57:44 +02:00
Hugo Touvron
094e79047a
add resmlp
2021-06-10 17:57:03 +02:00
Maxwell Aladago
2f76ec64fd
BUG Fix: ensure the right loss function is selected when mixtup or cutmix is enabeled
2021-05-02 10:27:25 -04:00
Maxwell Aladago
686507a6a5
BUG Fix: ensure the right loss function is selected when mixtup or cutmix is enabeled
2021-05-02 10:09:58 -04:00
Hugo Touvron
cb29b5efd5
Add XXS models
2021-04-14 09:07:53 +02:00
Hugo Touvron
542e05e021
Update README_cait.md
2021-04-12 08:03:12 +02:00
Hugo Touvron
0176669156
remove unused args
2021-04-12 07:52:27 +02:00
Hugo Touvron
bd59da1f5f
Update README.md
2021-04-09 16:46:40 +02:00
Hugo Touvron
4fcfe17ec3
add cait models and README
2021-04-09 16:45:17 +02:00
Hugo Touvron
7e0acbca7b
add cait fig
2021-04-09 16:39:57 +02:00
Francisco Massa
ab5715372d
Add deit_base_distilled_patch16_384 ( #55 )
2021-01-18 12:19:47 +01:00
Francisco Massa
84f7cf316f
Add Colab for inference ( #54 )
2021-01-18 11:37:44 +01:00
rv
23c7d85da9
Update README.md
2021-01-18 10:52:28 +01:00
Francisco Massa
277659c937
Add distilled and finetuned models to MODEL_ZOO ( #50 )
...
* Add results to README
* Add new model URLs
* Fill table with new models
* Use details in README
* Add URLs in the command-lines
* Add Training commands
* Bugfix
* Add models to hubconf
* Fix models
2021-01-18 10:43:54 +01:00
Matthijs Douze
0e32bfec3f
Merge pull request #53 from fmassa/update-figure
...
Update figure with newer models
2021-01-18 10:37:53 +01:00
Francisco Massa
7771d70fd8
Update figure with new models
2021-01-18 10:19:35 +01:00
Francisco Massa
1f961b060a
Fix no-weight-decay in optimizer ( #52 )
2021-01-16 11:53:29 +01:00
Francisco Massa
ac2b171ba0
Add copyright headers to losses.py ( #51 )
2021-01-15 12:06:11 +01:00
Francisco Massa
adc7f1e849
Add loss_scaler to checkpoints ( #49 )
...
Ensure bit-wise reproducibility when rescheduling jobs
2021-01-15 10:52:41 +01:00
Francisco Massa
a8e90967a3
Add option to finetune on larger resolution ( #43 )
...
* Add option for finetuning a model
* Fixes
* Keep model in eval mode during finetuning
* Only skip head weights if size mismatch
* Remove finetune-epochs
Might not be needed
* Raise error if distillation + finetune are enabled
2021-01-15 10:13:52 +01:00
Francisco Massa
d9932c08b5
Update .gitignore
2021-01-13 15:58:31 -08:00
Francisco Massa
726f061615
Fix lint
2021-01-13 05:45:38 -08:00
Francisco Massa
8eae3269da
Add Knowledge-Distillation ( #42 )
...
* Add knowledge distillation
* Bugfix
* Bugfix
* Make names more readable and use single torch.cat call
* Remove criterion.train() in engine
The teacher should stay in eval mode
* Change default argument for teacher-model
* Return the average of classifiers during inference
* Cleanup unused code
* Add docstring for DistillationLoss
* Remove warnings from newer PyTorch
Also uses more stable variant, instead of using softmax + log, use directly log_softmax
2021-01-13 14:19:23 +01:00
Francisco Massa
30eb3186da
Add --output_dir to README ( #36 )
...
This will make it clearer to users that they need to specify it if running without run_with_submitit, so that the results can be saved
2021-01-11 14:23:46 +01:00
Changlin Li
b06fcba6bf
Support parallelized evaluation ( #24 )
...
* support parallelized evaluation
* remove shuffle arg of loader val, add sampler val in non-dist branch
* replace timm eval sampler with torch sampler
* add logger synchronizing to support parallelized evaluation
* add command line argument dist-eval and warning
2021-01-08 11:05:39 +01:00
Matthijs Douze
38fcfbd863
Merge pull request #27 from fmassa/license-change
...
Change LICENSE to Apache 2.0
2021-01-08 11:02:40 +01:00
Francisco Massa
0c4b8f60bd
Change LICENSE to Apache 2.0
2021-01-08 10:51:58 +01:00
Francisco Massa
dcd888df20
Update default data-path ( #25 )
...
Points to the new valid location
2021-01-07 22:18:54 +01:00
Zhiyuan Chen
0282d2a175
Remove drop block args ( #23 )
...
timm's Vision Transformer does not support drop block
2021-01-07 17:28:41 +01:00
sanjaydatasciencedojo
4e91d2588f
Remove unused libraries ( #9 )
2020-12-27 21:43:54 +01:00
