Skip to content

GitLab

Switch branch/tag
  1. 14 May, 2024 2 commits
  3. 13 May, 2024 8 commits
    • fxmarty's avatar
      CI: update to ROCm 6.0.2 and test MI300 (#30266) · 37bba2a3
      fxmarty authored 
      

* update to ROCm 6.0.2 and test MI300

* add callers for mi300

* update dockerfile

* fix trainer tests

* remove apex

* style

* Update tests/trainer/test_trainer_seq2seq.py

* Update tests/trainer/test_trainer_seq2seq.py

* Update tests/trainer/test_trainer_seq2seq.py

* Update tests/trainer/test_trainer_seq2seq.py

* update to torch 2.3

* add workflow dispatch target

* we may need branches: mi300-ci after all

* nit

* fix docker build

* nit

* add check runner

* remove docker-gpu

* fix issues

* fix

---------
Co-authored-by: default avatarYih-Dar <2521628+ydshieh@users.noreply.github.com>
Co-authored-by: default avatarydshieh <ydshieh@users.noreply.github.com>
      37bba2a3
    • Marc Sun's avatar
      skip low_cpu_mem_usage tests (#30782) · 539ed75d
      Marc Sun authored
      539ed75d
    • Alazar's avatar
      Port IDEFICS to tensorflow (#26870) · 94306352
      Alazar authored 
      

* Initial commit

* Just a copy of modeling_idefics.py that will be ported to TF

* - Prepend TF to the name of all classes
- Convert pytorch ops to TF (not all operations are converted yet)

* Add TF imports

* Add autotranslated files

* Add TF classes to model_tf_auto.py

* Add the TF classes in model_doc

* include auto-translated code

* Adopted from auto-translated version

* Add a forgotten super().build

* Add test code for TF version.

* Fix indentation and load pytorch weights for now

* Some fixes. Many tests are still failing but some are passing now.

- I have added TODO's for some of the hacks I made to unblock me
  and I will address them soon
- I have the processing_idefics.py hacked in my view to support TF temporarily

* Add ALL_LAYERNORM_LAYERS to match pytorch

* Revert "Add ALL_LAYERNORM_LAYERS to match pytorch"

This reverts commit 7e0a35119b4d7a6284d04d8c543fba1b29e573c9 as it
is not needed in the tf implementation.

* Fix freeze_relevant_params()

* Some more fixes

* Fix test_attention_outputs

* Add tf stuff to processing_idefics.py

processing_idefics.py supports both pytorch and tf now.

test_processor_idefics.py for pytorch is passing, so i didn't break anything
but still some issues with tf. I also need to add tf tests in
test_processor_idefics.py.

* Pass return_tensors to image processing code and fix test

* Pass return_tensors to the image processor __init__

* Fix several test cases

- Make input to some of the forward pass of type `TFModelInputType`
- Decorate main layer forward pass with `@unpack_inputs`
- Decorate main layer with `@keras_serializable`
- Pass `inputs` to TFIdeficsModel

* Some more fixes forgotten in last commit

* Fix processing code and vision_tf.py

* Fix perceiver bug

* Import from

* Auto-add build() methods + style pass

* Fix build() errors due to `None` being passed as shape to some layers

* Change name in TFIdeficsForVisionText2Text to attribute in IdeficsForVisionText2Text

* Fix pytorch weights load for tf2

There were a lot of `name=` missing in weight initialization code.

* Attempt to fix CI

* Add back accidently removed line

* Remove torch-specific stuff from the TF test file

* make fix-copies, make style, remove autotranslated files

* Fixes to imports/docstrings

* Let's try the from future import in desperation

* Fix the core random_attention_mask fn to match the torch/flax behaviour

* Clean random_attention_mask up correctly

* Remove torch-only test

* Fix loss shape, couple of nits

* make style

* Don't test for OOB embeddings because IDEFICS uses those deliberately

* Fix loss computation to handle masking

* Fix test failures when flattening

* Fix some test failures

- Add cross attention gate which was missing and wasn't being passed arround
- Fix overwriting of image_attention_mask due to hack I had for dummy inputs

* Add a proper stateless scaled_dot_product_attention

* make style

* Adding missing attribute from the PyTorch version

* Small cleanups to decoupledlinearlayer in case that helps

* Pass epsilon to LayerNormalization

* Attemp to fix pytorch weight cross-loading for TFIdeficsEmbedding

* Fix a bug in TFIdeficsGatedCrossAttentionLayer

* Patching up build() methods

* Constant self.inv_freq

* Constant self.inv_freq

* First working version

The TF implementation works now, there was a bug in the TFIdeficsDecoupledLinear
where the weights were mis-intialized (in_features,out_features)
when it should be: (out_features, in_features)

I have tested this so far with tiny-random and idefics-9b-instruct
and gives correct output.

I also dumped the final outputs for both pytorch and TF
and they are identical.

* Fix some test failures

* remove print statement

* Fix return_tensors

* Fix CI test failure check_code_quality

* Attempt to fix CI failures by running `make fixup`

The hardcoded IDs in test_modeling_tf_idefics.py are for the integration
test and makes that file unreadable and should probably be moved to a seperate file.

* Attempt to fix tests_pr_documentation_tests

* Fix a test failure in test_image_processing_idefics.py

* Fix test test_pt_tf_model_equivalence

* Fix a few failures

* Tiny fix

* Some minor fixes

* Remove a duplicate test

* Override a few test failures for IDEFICS

- `test_keras_save_load` is passing now
- `test_compile_tf_model` is still failing

* Fix processing_idefics.py after rebase

* Guard import keras with is_tf_available

* fix check code quality

* fix check code quality

* Minor fixes

* Skip test_save_load temporarily

This test passed on my local box but fails on the CI, skipping
for now to see if there are other remaining failures on the CI.

* Run `ruff format tests src utils`

* Fix last failing test, `test_compile_tf_model`

* Add fixes for vision_tf.py

I forgot to add this file in last commit.

* Minor fixes

* Replace "<<<" with "<<" for doc tests

IDEFICS-9B is too big for doctest runner, so don't run it there

* Make code more readable

* Fix bug after code review

I added a layer_norm_eps to IdeficsConfig but I don't even need it
since the vision config has a layer_norm_eps.

* Fix after code review

Use original code tokenizer.convert_tokens_to_ids

* Keep PyTorch as the default return_tensors

* Fixes to modeling_tf after code review

* Fixes from code review

- Remove all references of `TF_IDEFICS_PRETRAINED_MODEL_ARCHIVE_LIST`
- Pass 1e-5 to LayerNormalization in perceiver

* Run ruff

* Undo a change

* Refactor processing code after Matt's suggestion

* Remove TODO's that aren't needed anymore

* For pytorch, Use original pytorch processing code from main

Since this PR is a TF port it shouldn't make any modifications
to pytorch IDEFICS code. This changes undo's the pytorch processing
modifications I made and uses original code from main.

* Update tests/models/idefics/test_modeling_idefics.py

* Update tests/models/idefics/test_modeling_tf_idefics.py

* Add missing imports for is_pt_tf_cross_test

* [DO NOT MERGE]: This is a commit for debugging and will be reverted

The cross test `test_pt_tf_model_equivalence` passes locally but
fails when running on the CI. This commit is to help debug that
and will be reverted.

* Revert "[DO NOT MERGE]: This is a commit for debugging and will be reverted"

This reverts commit 8f0d709ec5bd46685fb0b4259d914ffee794875b.

* [DO NOT MERGE]: This commit is for debugging a CI failure and will be reverted

* [DO NOT MERGE]: This commit is for debugging a CI failure and will be reverted

* Revert "[DO NOT MERGE]: This commit is for debugging a CI failure and will be reverted"

This reverts commit 998cc38b8c3d313bf5e5eb55a7f5b7b881897b89.

* Revert "[DO NOT MERGE]: This commit is for debugging a CI failure and will be reverted"

This reverts commit 1c695ac4219c4ae4d39b330b01744dc27deb7dd4.

* Don't skip test_save_load

IIRC test_save_load was also failing on the CI but not on my local
box, it might be easier to debug that on the CI first than the cross tests

* Debugging commit, will be reverted

* Revert "Debugging commit, will be reverted"

This reverts commit 8eafc8e41e20c4e95a3a90834f06a6e9f445e2d5.

* Override `test_save_load` and push model to save

Maybe this will help me repro this weird bug

* pass my repo_id

* add endpoint

* Pass a temp (write) token just for this CI

* Undo last few commits, still pushing to hub for model debugging

The issue seems to be with save_pretrained(),  when I looked at the model saved
from the CI test failure it is basically empty and has no weights.
`self.save_weights(..)` seems to be failing in save_pretrained but needs
more debugging

* Add logging to modeling tf utils, will be reverted just for debugging

* Debugging, will revert

* Revert "Debugging, will revert"

This reverts commit 9d0d3075fb7c82d8cde3a5c76bc8f3876c5c55d3.

* Revert "Add logging to modeling tf utils, will be reverted just for debugging"

This reverts commit 774b6b7b1c17b3ce5d7634ade768f2f686cee617.

* Remove `test_save_load`

The CI failures are gone after my latest rebase, no idea why
but I was still saving the model to my hub on HF and the tf_model.h5
file now has everything.

* Run make fix-copies

* Run ruff format tests src utils

* Debugging commit, will be reverted

* Run ruff, also trigger CI run

* Run ruff again

* Undo debugging commit

---------
Co-authored-by: default avatarMatt <rocketknight1@gmail.com>
Co-authored-by: default avatarMatt <Rocketknight1@users.noreply.github.com>
      94306352
    • Fanli Lin's avatar
      enable Pipeline to get device from model (#30534) · 69d9bca5
      Fanli Lin authored 
      

* check model.device

* fix

* style fix

* move model device

* remove print

* add comment

* fix

* add unit test

* optimize

* change test names and add more cases

* Update tests/pipelines/test_pipelines_common.py
Co-authored-by: default avataramyeroberts <22614925+amyeroberts@users.noreply.github.com>

---------
Co-authored-by: default avataramyeroberts <22614925+amyeroberts@users.noreply.github.com>
      69d9bca5
    • Poedator's avatar
      Llama: fix custom 4D masks, v2 (#30348) · a0779b9e
      Poedator authored 
      

* 4d mask fixes

* Update custom 4D mask logic

* test moved to mixin

* extra tests 4d mask

* upd 4d mask and StaticCache handling

* added Mask4DTestHard to mistral tests

* post-rebase fixes

* test fixes for StaticCache

* make fix-copies

* upd 1 after #30476

* fix common tests

* rm elif attention_mask.dim() == 4:

* tests combined, fixed, mixtral supported

* bigbird style chg reverted

* rm if attention_mask.dim() == 2

* modeling_llama formatting chg

---------
Co-authored-by: default avatarJoao Gante <joao@huggingface.co>
      a0779b9e
    • Nilabhra Roy Chowdhury's avatar
      Support for Falcon2-11B (#30771) · e52741f6
      Nilabhra Roy Chowdhury authored 
      

* remove unrelated changes

* remove unrelated changes on phi and stable LM

* add: Test for Falcon 10B

* fix: formatting

* fix: loading the falcon 10B in 8 bit precision using bitsanbytes.

* fix: device placement

* fix: broken tests.

* fix: backwards compatibility for falcon 1B architecture.

* chore: updated test.

* chore: test_modeling_falcon.py to use the 11B model.

* chore: minor edit

* chore: formating.

---------
Co-authored-by: default avatarPablo Montalvo <39954772+molbap@users.noreply.github.com>
Co-authored-by: default avatarArthurZucker <arthur.zucker@gmail.com>
      e52741f6
    • Zafir Stojanovski's avatar
      Blip dynamic input resolution (#30722) · f63d8222
      Zafir Stojanovski authored 
      * blip with interpolated pos encoding

* feat: Add interpolate_pos_encoding option to other models from `BLIP` family.

* include check for textual generated content in tests
      f63d8222
    • Marc Sun's avatar
      [awq] replace scale when we have GELU (#30074) · de6e0db1
      Marc Sun authored 
      * fix awq test

* style

* add log

* new fix

* style

* only modifying impacted model in the end

* rename function
      de6e0db1
  5. 09 May, 2024 6 commits
  7. 08 May, 2024 2 commits
  9. 07 May, 2024 4 commits
  11. 06 May, 2024 4 commits
    • Nate Cibik's avatar
      Trainer - add cache clearing and the option for batched eval metrics computation (#28769) · df475bf8
      Nate Cibik authored 
      * Added cache clearing for GPU efficiency.

* Added cache clearing for GPU efficiency.

* Added batch_eval_metrics capability

* Ran make fixup

* Fixed bug

* Fixed whitespace issue

* Fixed outdated condition

* Updated docstrings with instructions for batch_eval_metrics. Updated end of dataloader logic

* Added first version of batch_eval_metrics Trainer test

* Fixed batch_eval_metrics Trainer tests for both eval and predict

* Fixed batch_eval_metrics behavior for new Trainer variables

* Fixed batch_eval_metrics Trainer tests

* Ran fixup
      df475bf8
    • Clara Pohland's avatar
      Trainer._load_from_checkpoint - support loading multiple Peft adapters (#30505) · e0769530
      Clara Pohland authored 
      

* Trainer: load checkpoint model with multiple adapters

* Trainer._load_from_checkpoint support multiple active adapters

* PeftModel.set_adapter does not support multiple adapters yet

* Trainer._load_from_checkpoint test multiple adapters

---------
Co-authored-by: default avatarClara Luise Pohland <clara-luise.pohland@telekom.de>
      e0769530
    • Younes Belkada's avatar
      Quantization / HQQ: Fix HQQ tests on our runner (#30668) · 9c772ac8
      Younes Belkada authored 
      Update test_hqq.py
      9c772ac8
    • Arthur's avatar
      [`CI update`] Try to use dockers and no cache (#29202) · 307f632b
      Arthur authored 
      

* change cis

* nits

* update

* minor updates

* [push-ci-image]

* nit [push-ci-image]

* nitsssss

* [build-ci-image]

* [push-ci-image]

* [push-ci-image]

* both

* [push-ci-image]

* this?

* [push-ci-image]

* pypi-kenlm needs g++

* [push-ci-image]

* nit

* more nits [push-ci-image]

* nits [push-ci-image]

* [push-ci-image]

* [push-ci-image]

* [push-ci-image]

* add vision

* [push-ci-image]

* [push-ci-image]

* add new dummy file but will need to update them [push-ci-image]

* [push-ci-image]

* show package size as well

* [push-ci-image]

* potentially ignore failures

* workflow updates

* nits [push-ci-image]

* [push-ci-image]

* fix consistency

* clean nciida triton

* also show big packages [push-ci-image]

* nit

* update

* another one

* line escape?

* add accelerate [push-ci-image]

* updates [push-ci-image]

* nits to run tests, no push-ci

* try to parse skip reason to make sure nothing is skipped that should no be skippped

* nit?

* always show skipped reasons

* nits

* better parsing of the test outputs

* action="store_true",

* failure on failed

* show matched

* debug

* update short summary with skipped, failed and errors

* nits

* nits

* coolu pdates

* remove docbuilder

* fix

* always run checks

* oups

* nits

* don't error out on library printing

* non zero exi codes

* no warning

* nit

* WAT?

* format nit

* [push-ci-image]

* fail if fail is needed

* [push-ci-image]

* sound file for torch light?

* [push-ci-image]

* order is important [push-ci-image]

* [push-ci-image] reduce even further

* [push-ci-image]

* use pytest rich !

* yes [push-ci-image]

* oupsy

* bring back the full traceback, but pytest rich should help

* nit

* [push-ci-image]

* re run

* nit

* [push-ci-image]

* [push-ci-image]

* [push-ci-image]

* empty push to trigger

* [push-ci-image]

* nit? [push-ci-image]

* empty

* try to install timm with no deps

* [push-ci-image]

* oups [push-ci-image]

* [push-ci-image]

* [push-ci-image] ?

* [push-ci-image] open ssh client for git checkout fast

* empty for torch light

* updates [push-ci-image]

* nit

* @v4 for checkout

* [push-ci-image]

* [push-ci-image]

* fix fetch tests with parallelism

* [push-ci-image]

* more parallelism

* nit

* more nits

* empty to re-trigger

* empty to re-trigger

* split by timing

* did not work with previous commit

* junit.xml

* no path?

* mmm this?

* junitxml format

* split by timing

* nit

* fix junit family

* now we can test if the xunit1 is compatible!

* this?

* fully list tests

* update

* update

* oups

* finally

* use classname

* remove working directory to make sure the path does not interfere

* okay no juni should have the correct path

* name split?

* sort by classname is what make most sense

* some testing

* naem

* oups

* test something fun

* autodetect

* 18?

* nit

* file size?

* uip

* 4 is best

* update to see versions

* better print

* [push-ci-image]

* [push-ci-image]

* please install the correct keras version

* [push-ci-image]

* [push-ci-image]

* [push-ci-image]

* [push-ci-image]

* [push-ci-image]

* uv is fucking me up

* [push-ci-image]

* [push-ci-image]

* [push-ci-image]

* nits

* [push-ci-image]

* [push-ci-image]

* install issues an pins

* tapas as well

* nits

* more paralellism

* short tb

* soundfile

* soundfile

* [push-ci-image]

* [push-ci-image]

* [push-ci-image]

* oups

* [push-ci-image]

* fix some things

* [push-ci-image]

* [push-ci-image]

* [push-ci-image]

* [push-ci-image]

* use torch-light for hub

* small git lfs for hub job

* [push-ci-image]

* [push-ci-image]

* [push-ci-image]

* [push-ci-image]

* fix tf tapas

* [push-ci-image]

* nits

* [push-ci-image]

* don't update the test

* [push-ci-image]

* [push-ci-image]

* [push-ci-image]

* no use them

* [push-ci-image]

* [push-ci-image]

* [push-ci-image]

* [push-ci-image]

* update tf proba

* [push-ci-image]

* [push-ci-image]

* woops

* [push-ci-image]

* [push-ci-image]

* [push-ci-image]

* [push-ci-image]

* [push-ci-image]

* [push-ci-image]

* test with built dockers

* [push-ci-image]

* skip annoying tests

* revert fix copy

* update test values

* update

* last skip and fixup

* nit

* ALL GOOOD

* quality

* Update tests/models/layoutlmv2/test_image_processing_layoutlmv2.py

* Update docker/quality.dockerfile
Co-authored-by: default avatarLysandre Debut <hi@lysand.re>

* Update src/transformers/models/tapas/modeling_tf_tapas.py
Co-authored-by: default avatarLysandre Debut <hi@lysand.re>

* Apply suggestions from code review
Co-authored-by: default avatarLysandre Debut <hi@lysand.re>

* use torch-speed

* updates

* [push-ci-image]

* [push-ci-image]

* [push-ci-image]

* [push-ci-image]

* fuck ken-lm [push-ci-image]

* [push-ci-image]

* [push-ci-image]

---------
Co-authored-by: default avatarLysandre Debut <hi@lysand.re>
      307f632b
  13. 02 May, 2024 4 commits
    • mobicham's avatar
      Add HQQ quantization support (#29637) · 59952994
      mobicham authored 
      

* update HQQ transformers integration

* push import_utils.py

* add force_hooks check in modeling_utils.py

* fix | with Optional

* force bias as param

* check bias is Tensor

* force forward for multi-gpu

* review fixes pass

* remove torch grad()

* if any key in linear_tags fix

* add cpu/disk check

* isinstance return

* add multigpu test + refactor tests

* clean hqq_utils imports in hqq.py

* clean hqq_utils imports in quantizer_hqq.py

* delete hqq_utils.py

* Delete src/transformers/utils/hqq_utils.py

* ruff init

* remove torch.float16 from __init__ in test

* refactor test

* isinstance -> type in quantizer_hqq.py

* cpu/disk device_map check in quantizer_hqq.py

* remove type(module) nn.linear check in quantizer_hqq.py

* add BaseQuantizeConfig import inside HqqConfig init

* remove hqq import in hqq.py

* remove accelerate import from test_hqq.py

* quant config.py doc update

* add hqqconfig to main_classes doc

* make style

* __init__ fix

* ruff __init__

* skip_modules list

* hqqconfig format fix

* hqqconfig doc fix

* hqqconfig doc fix

* hqqconfig doc fix

* hqqconfig doc fix

* hqqconfig doc fix

* hqqconfig doc fix

* hqqconfig doc fix

* hqqconfig doc fix

* hqqconfig doc fix

* test_hqq.py remove mistral comment

* remove self.using_multi_gpu is False

* torch_dtype default val set and logger.info

* hqq.py isinstance fix

* remove torch=None

* torch_device test_hqq

* rename test_hqq

* MODEL_ID in test_hqq

* quantizer_hqq setattr fix

* quantizer_hqq typo fix

* imports quantizer_hqq.py

* isinstance quantizer_hqq

* hqq_layer.bias reformat quantizer_hqq

* Step 2 as comment in quantizer_hqq

* prepare_for_hqq_linear() comment

* keep_in_fp32_modules fix

* HqqHfQuantizer reformat

* quantization.md hqqconfig

* quantization.md model example reformat

* quantization.md # space

* quantization.md space   })

* quantization.md space   })

* quantization_config fix doc
Co-authored-by: default avataramyeroberts <22614925+amyeroberts@users.noreply.github.com>

* axis value check in quantization_config

* format

* dynamic config explanation

* quant config method in quantization.md

* remove shard-level progress

* .cuda fix modeling_utils

* test_hqq fixes

* make fix-copies

---------
Co-authored-by: default avataramyeroberts <22614925+amyeroberts@users.noreply.github.com>
      59952994
    • Jonghwan Hyeon's avatar
      Output `None` as attention when layer is skipped (#30597) · 4c940934
      Jonghwan Hyeon authored 
      * Output `None` as attention when layer is skipped

* Add test for output_attentions
      4c940934
    • Richard Brown's avatar
      馃毃 Update image_processing_vitmatte.py (#30566) · f9530258
      Richard Brown authored 
      * Update image_processing_vitmatte.py

* add test

* [run-slow]vitmatte
      f9530258
    • Michael Benayoun's avatar
      Fix for Neuron (#30259) · fbabd674
      Michael Benayoun authored
      fbabd674
  15. 01 May, 2024 3 commits
  17. 30 Apr, 2024 5 commits
  19. 26 Apr, 2024 2 commits