96d245a83d
torch 2.9 don't ❤️ torchcodec 💔 ( #41610 )
...
pin
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-10-15 14:34:00 +02:00
b28902c86b
Remove DISABLE_KERNEL_MAPPING flag ( #41475 )
...
rm disable
2025-10-10 10:19:25 +02:00
3256773974
FP-Quant NVFP4 and Python 3.9 support ( #39876 )
...
* quartet
* quartet qat -> quartet
* format
* bf16 backward
* interfaces
* forward_method
* quartet -> fp_quant
* style
* List -> list
* list typing
* fixed format and annotations
* test_fp_quant
* docstrings and default dtypes
* better docstring and removed noop checks
* docs
* pseudoquantization support to test on non-blackwell
* pseudoquant
* Pseudoquant docs
* Update docs/source/en/quantization/fp_quant.md
Co-authored-by: Marc Sun <57196510+SunMarc@users.noreply.github.com >
* Update docs/source/en/quantization/fp_quant.md
* Update docs/source/en/quantization/fp_quant.md
* Update src/transformers/utils/quantization_config.py
Co-authored-by: Mohamed Mekkouri <93391238+MekkCyber@users.noreply.github.com >
* Update tests/quantization/fp_quant_integration/test_fp_quant.py
Co-authored-by: Mohamed Mekkouri <93391238+MekkCyber@users.noreply.github.com >
* Update tests/quantization/fp_quant_integration/test_fp_quant.py
Co-authored-by: Marc Sun <57196510+SunMarc@users.noreply.github.com >
* small test fixes
* dockerfile update
* spec link
* removed `_process_model_after_weight_loading`
* toctree
* nvfp4
* nvfp4 tests
* FP-Quant version bumped
* nvfp4 default and docs update
* trainable
* cpu if pseudoquant
* proper group size selection
* gsr
* qutlass requirement version bumo
* Upstream docker copy
* docs update
---------
Co-authored-by: Marc Sun <57196510+SunMarc@users.noreply.github.com >
Co-authored-by: Mohamed Mekkouri <93391238+MekkCyber@users.noreply.github.com >
2025-10-01 13:58:22 +00:00
3e975acc8b
Fix docker quantization ( #41201 )
...
* launch docker
* remove gptq for now
* run tests
* Revert "run tests"
This reverts commit f85718ce3a21d5937bf7405b8925c125c67d1a3e.
* revert
2025-09-29 16:36:30 +00:00
50d2448a1a
Enable fa in amd docker ( #41069 )
...
* Add FA to docker
* Use caching mechanism for qwen2_5
* Fix a typo in important models list
* Partial fixes for gemma3
* Added a commit ID for FA repo
* Detailled the expectation storage format
* Rebase fix
* Apply style fixes
---------
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
2025-09-26 13:57:58 +02:00
257fe5eea8
Switch to python:3.10-slim for CircleCI docker images ( #41067 )
...
fix
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-09-23 12:48:48 +02:00
aa30e0642e
Update quantization CI ( #41068 )
...
* fix
* new everything
* fix
2025-09-22 18:10:16 +02:00
4df2529d79
🚨 🚨 🚨 Fully remove Tensorflow and Jax support library-wide ( #40760 )
...
* setup
* start the purge
* continue the purge
* more and more
* more
* continue the quest: remove loading tf/jax checkpoints
* style
* fix configs
* oups forgot conflict
* continue
* still grinding
* always more
* in tje zone
* never stop
* should fix doc
* fic
* fix
* fix
* fix tests
* still tests
* fix non-deterministic
* style
* remove last rebase issues
* onnx configs
* still on the grind
* always more references
* nearly the end
* could it really be the end?
* small fix
* add converters back
* post rebase
* latest qwen
* add back all converters
* explicitly add functions in converters
* re-add
2025-09-18 18:27:39 +02:00
26b65fb516
Intel CPU dockerfile ( #40806 )
...
* upload intel cpu dockerfile
Signed-off-by: jiqing-feng <jiqing.feng@intel.com >
* update cpu dockerfile
Signed-off-by: jiqing-feng <jiqing.feng@intel.com >
* update label name
Signed-off-by: jiqing-feng <jiqing.feng@intel.com >
---------
Signed-off-by: jiqing-feng <jiqing.feng@intel.com >
2025-09-17 15:42:30 +00:00
5b0c01b5e2
Final test data cache - inside CI docker images ( #40689 )
...
* run
* build
* build
* fix
---------
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-09-04 13:12:49 +00:00
30a4b8707d
CircleCI docker images cleanup / update / fix ( #40681 )
...
* fix
* fix
---------
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-09-04 10:42:18 +02:00
b3655507bb
Pin torchcodec to 0.5 in AMD docker ( #40598 )
2025-09-01 20:39:55 +02:00
1363fceeec
remove the redundant non maintained jieba and use rjieba instead ( #40383 )
...
* porting not maintained jieba to rjieba
* Fix format
* replaced the line with rjieba instead of removing it
* cut_all is not included as a parameter. cut_all is a seperate function rjieba
* rev
* jieba remove installation
* Trigger tests
* Update tokenization_cpm.py
* Update tokenization_cpm_fast.py
---------
Co-authored-by: Yih-Dar <2521628+ydshieh@users.noreply.github.com >
2025-08-30 13:28:52 +02:00
851b8f281d
[kernels] If flash attention2 is not installed / fails to import (cc on our cluster) default to kernels ( #40178 )
...
* first step if flash not installed but you set to use it
* try importing
* now default to using it
* update our tests as well
* wow yesterday I was not awake
* fixup
* style
* lol the fix was very very simple
* `RUN python3 -m pip install --no-cache-dir git+https://github.com/huggingface/kernels@main#egg=kernels
` for updated dockers
* push review comments
* fix
---------
Co-authored-by: Cyril Vallez <cyril.vallez@huggingface.co >
Co-authored-by: Cyril Vallez <cyril.vallez@gmail.com >
2025-08-28 16:20:25 +02:00
ff8b88a948
Fix nightly torch CI ( #40469 )
...
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-08-26 22:02:15 +02:00
4f9b4e62bc
Run FA2 tests in CI ( #40397 )
...
up
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-08-23 12:30:18 +02:00
3b7230124b
No more natten ( #40287 )
...
get rid off natten
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-08-20 16:10:15 +02:00
2914ceca20
Revert "Pin torch to 2.7.1 on CircleCI for now" + Final fix for too long with no output ( #40201 )
...
* Revert "Pin torch to 2.7.1 on CircleCI for now (#40174 )"
This reverts commit 31b6e6e1dac0d32f74ec5cd6b3c1868534ccd7b5.
* fix
---------
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-08-18 08:40:53 +02:00
31b6e6e1da
Pin torch to 2.7.1 on CircleCI for now ( #40174 )
...
* fix
* fix
---------
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-08-14 20:19:35 +02:00
be1ab5103f
Update Dockerfiles to install packages inside a virtual environment ( #39098 )
...
* Removed un-necessary virtual environment creation in Dockerfiles.
* Updated Dockerfiles to install packages in a virtual environment.
* use venv's python
* update
* build and trigger
* trigger
* build and trigger
* build and trigger
* build and trigger
* build and trigger
* build and trigger
* build and trigger
* update
* update
* update
* update
---------
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-08-13 23:51:52 +02:00
99c747539e
unpin torchcodec==0.5.0 and use torch 2.8 on daily CI ( #40072 )
...
fix
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-08-10 22:27:39 +02:00
4d57c39007
pin torchcodec==0.5.0 for now with torch 2.7.1 on daily CI ( #40013 )
...
* update
* update
---------
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-08-07 23:05:39 +02:00
ca4cbb1e3f
unpin torch<2.8 on circleci ( #40012 )
...
update
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-08-07 21:31:17 +02:00
9e76a6bb54
circleci: pin torch 2.7.1 until torchcodec is updated ( #39951 )
...
circleci torch 2.7.1
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-08-06 11:18:00 +02:00
5fb5b6cfaf
Fix quant docker for fp-quant ( #39641 )
...
* fix quant docker
* Apply style fixes
---------
Co-authored-by: Mohamed Mekkouri <93391238+MekkCyber@users.noreply.github.com >
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
2025-08-04 11:57:08 +00:00
65df73aa88
AMD disable torchcodec ( #39757 )
...
Temporarily disable torchcodec installation because of bizarre segfault
2025-07-29 13:07:25 +00:00
14adcbd937
Fix AMD dockerfile for audio models ( #39669 )
2025-07-28 19:05:41 +02:00
623ab01039
FP-Quant support ( #38696 )
...
* quartet
* quartet qat -> quartet
* format
* bf16 backward
* interfaces
* forward_method
* quartet -> fp_quant
* style
* List -> list
* list typing
* fixed format and annotations
* test_fp_quant
* docstrings and default dtypes
* better docstring and removed noop checks
* docs
* pseudoquantization support to test on non-blackwell
* pseudoquant
* Pseudoquant docs
* Update docs/source/en/quantization/fp_quant.md
Co-authored-by: Marc Sun <57196510+SunMarc@users.noreply.github.com >
* Update docs/source/en/quantization/fp_quant.md
* Update docs/source/en/quantization/fp_quant.md
* Update src/transformers/utils/quantization_config.py
Co-authored-by: Mohamed Mekkouri <93391238+MekkCyber@users.noreply.github.com >
* Update tests/quantization/fp_quant_integration/test_fp_quant.py
Co-authored-by: Mohamed Mekkouri <93391238+MekkCyber@users.noreply.github.com >
* Update tests/quantization/fp_quant_integration/test_fp_quant.py
Co-authored-by: Marc Sun <57196510+SunMarc@users.noreply.github.com >
* small test fixes
* dockerfile update
* spec link
* removed `_process_model_after_weight_loading`
* toctree
---------
Co-authored-by: Marc Sun <57196510+SunMarc@users.noreply.github.com >
Co-authored-by: Mohamed Mekkouri <93391238+MekkCyber@users.noreply.github.com >
2025-07-23 11:41:10 +02:00
0cae633ce1
Bump AMD container for 2.7.1 PyTorch ( #39458 )
...
* Bump AMD container for 2.7.1 PyTorch
* Forgot to update pinned packages
2025-07-22 12:11:38 +02:00
dc98fb3e5e
update docker file to use latest timm (for perception_lm) ( #39380 )
...
update docker file for timm
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-07-12 23:19:37 +02:00
1ecd52e50a
Add torchcodec in docstrings/tests for datasets 4.0 ( #39156 )
...
* fix dataset run_object_detection
* bump version
* keep same dataset actually
* torchcodec in docstrings and testing utils
* torchcodec in dockerfiles and requirements
* remove duplicate
* add torchocodec to all the remaining docker files
* fix tests
* support torchcodec in audio classification and ASR
* [commit to revert] build ci-dev images
* [commit to revert] trigger circleci
* [commit to revert] build ci-dev images
* fix
* fix modeling_hubert
* backward compatible run_object_detection
* revert ci trigger commits
* fix mono conversion and support torch tensor as input
* revert map_to_array docs + fix it
* revert mono
* nit in docstring
* style
* fix modular
---------
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-07-08 17:06:12 +02:00
cb17103bd5
Uninstallling Flash attention from quantization docker ( #39078 )
...
* update
* revert
2025-06-27 13:51:46 +02:00
8ac2d75353
Pin PyTorch extras for AMD containers ( #38941 )
...
* Pin additional Torch packages
* Remove unused def
---------
Co-authored-by: ivarflakstad <69173633+ivarflakstad@users.noreply.github.com >
2025-06-20 12:17:21 +00:00
38a9b70786
add pytorch-xpu Dockerfile ( #38875 )
...
* first commit
Signed-off-by: YAO Matrix <matrix.yao@intel.com >
* use rls pytorch
Signed-off-by: YAO Matrix <matrix.yao@intel.com >
---------
Signed-off-by: YAO Matrix <matrix.yao@intel.com >
2025-06-20 11:42:44 +02:00
10627c1a0f
Use torch 2.7.1 on daily CI ( #38620 )
...
* fix
* fix
---------
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-06-08 14:37:45 +02:00
dc76eff12b
remove ipex_optimize_model usage ( #38632 )
...
* remove ipex_optimize_model usage
Signed-off-by: YAO Matrix <matrix.yao@intel.com >
* update Dockerfile
Signed-off-by: root <root@a4bf01945cfe.jf.intel.com >
---------
Signed-off-by: YAO Matrix <matrix.yao@intel.com >
Signed-off-by: root <root@a4bf01945cfe.jf.intel.com >
Co-authored-by: root <root@a4bf01945cfe.jf.intel.com >
2025-06-06 20:04:44 +02:00
5009252a05
Better CI ( #38552 )
...
better CI
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-06-06 17:59:14 +02:00
90c4b90a10
Use torch 2.7.1 on CircleCI jobs ( #37856 )
...
2.7.1
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-06-06 10:16:57 +02:00
ca0a682796
Update docker image to use av ( #38548 )
...
* Update
* Update
---------
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-06-03 11:04:41 +02:00
6e3063422c
Uninstall kernels for AMD docker images ( #38354 )
...
Uninstall kernels for AMD docker images
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-05-25 19:42:25 +02:00
9895819514
Update CI Docker base image for AMD tests ( #38261 )
...
use newer Pytorch base image for AMD CI tests
2025-05-22 16:38:40 +02:00
dcaf47dde5
add liger-kernel to docker file ( #38292 )
...
add
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-05-22 11:58:17 +02:00
4143f94d51
uninstall kernels from docker images ( #38083 )
...
uninstall kernels
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-05-12 18:03:47 +02:00
ca790303f7
Pin torch == 2.6 on PR CI docker images for now ( #37695 )
...
pin 2.6 on CircleCi images
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-04-23 11:47:23 +02:00
b3492ff9f7
Add AutoRound quantization support ( #37393 )
...
* add auto-round support
* Update src/transformers/quantizers/auto.py
Co-authored-by: Ilyas Moutawwakil <57442720+IlyasMoutawwakil@users.noreply.github.com >
* fix style issue
Signed-off-by: wenhuach <wenhuach87@gmail.com >
* tiny change
* tiny change
* refine ut and doc
* revert unnecessary change
* tiny change
* try to fix style issue
* try to fix style issue
* try to fix style issue
* try to fix style issue
* try to fix style issue
* try to fix style issue
* try to fix style issue
* fix doc issue
* Update tests/quantization/autoround/test_auto_round.py
* fix comments
* Update tests/quantization/autoround/test_auto_round.py
Co-authored-by: Marc Sun <57196510+SunMarc@users.noreply.github.com >
* Update tests/quantization/autoround/test_auto_round.py
Co-authored-by: Marc Sun <57196510+SunMarc@users.noreply.github.com >
* update doc
* Update src/transformers/quantizers/quantizer_auto_round.py
Co-authored-by: Marc Sun <57196510+SunMarc@users.noreply.github.com >
* update
* update
* fix
* try to fix style issue
* Update src/transformers/quantizers/auto.py
Co-authored-by: Mohamed Mekkouri <93391238+MekkCyber@users.noreply.github.com >
* Update docs/source/en/quantization/auto_round.md
Co-authored-by: Mohamed Mekkouri <93391238+MekkCyber@users.noreply.github.com >
* Update docs/source/en/quantization/auto_round.md
Co-authored-by: Mohamed Mekkouri <93391238+MekkCyber@users.noreply.github.com >
* Update docs/source/en/quantization/auto_round.md
Co-authored-by: Mohamed Mekkouri <93391238+MekkCyber@users.noreply.github.com >
* update
* fix style issue
* update doc
* update doc
* Refine the doc
* refine doc
* revert one change
* set sym to True by default
* Enhance the unit test's robustness.
* update
* add torch dtype
* tiny change
* add awq convert test
* fix typo
* update
* fix packing format issue
* use one gpu
---------
Signed-off-by: wenhuach <wenhuach87@gmail.com >
Co-authored-by: Ilyas Moutawwakil <57442720+IlyasMoutawwakil@users.noreply.github.com >
Co-authored-by: Marc Sun <57196510+SunMarc@users.noreply.github.com >
Co-authored-by: Mohamed Mekkouri <93391238+MekkCyber@users.noreply.github.com >
Co-authored-by: Shen, Haihao <haihao.shen@intel.com >
2025-04-22 13:56:54 +02:00
897874748b
Disable kernels for quantization ( #37446 )
...
fix
2025-04-11 16:35:38 +02:00
3c39c07939
Remove triton mlp kernel, not compiling for some models ( #37449 )
...
* remove mlp for now
* disable on docker
2025-04-11 12:47:13 +02:00
7ae0be722e
update deepspeed docker ( #37371 )
...
* update
* create docker image
* 03
* uninstall pytest as it conflits with transformers
* wrong one
* better
* see which package depends on pytest
* up
* resintall
* fix
* deepspeedddddddd
* deepspeedddddddd
* deepspeedddddddd
* deepspeedddddddd
* deepspeedddddddd
* deepspeedddddddd
* deepspeedddddddd
* deepspeedddddddd
---------
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-04-09 14:54:06 +02:00
352ec8ef22
pin specific natten version in docker file ( #37274 )
...
fix
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-04-04 13:47:16 +02:00
211e4dc9a4
[chat-template] fix video loading ( #37146 )
...
* fix
* add video
* trigger
* push new iamges
* fix tests
* revert
---------
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com >
2025-04-02 11:27:50 +02:00
