GitLab

error show like: “Currently the auto_kernel_selection does not support the grad mode! Please add torch.no_grad() before the inference runtime..”
since jit mode only work in inference mode, it's safe to add such logic.

Neuron supports extraction of XLA graphs for compilation.
However, when both do_train and do_eval options are enabled,
sizes returned by tensor operator can be 0. To avoid
INVALID_ARGUMENT error, we use inequality in the check whether
a tensor needs padding or not.

* fix `opt` bias

* revert unneeded assignment

* make add_special_tokens more clear
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>

* remove truncation

* For TFWhisper
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>

* Add transforms for object detection

* DETR models + Yolos

* Scrappy additions

* Maskformer image processor

* Fix up; MaskFormer tests

* Update owlvit processor

* Add to docs

* OwlViT tests

* Update pad logic

* Remove changes to transforms

* Import fn directly

* Update to include pad transformation

* Remove uninstended changes

* Add new owlvit post processing function

* Tidy up

* Fix copies

* Fix some copies

* Include device fix

* Fix scipy imports

* Update _pad_image

* Update padding functionality

* Fix bug

* Properly handle ignore index

* Fix up

* Remove defaults to None in docstrings

* Fix docstrings & docs

* Fix sizes bug

* Resolve conflicts in init

* Cast to float after resizing

* Tidy & add size if missing

* Allow kwards when processing for owlvit

* Update test values

* [modelcard] Set model name if empty

* no magic
Co-authored-by: Sylvain Gugger <sylvain@huggingface.co>
Co-authored-by: Sylvain Gugger <sylvain@huggingface.co>

Co-authored-by: Sandeep Kumar <sandeep.kumar@woven-planet.global>

* replace assert statements with exceptions

* made conditions more readable

* add minimal working gpt2 tokenizer

* graph mode and output equivalence tests working

* not today tensorflow. serialization test passing!

* fix style, documentation, docstrings and all that jazz

* passing consistency checks

* move keras nlp to tf dependencies

* fix tf modeling utils and gpt2 attention to enable compiling

* fix (I hope) keras nlp dependencies

* rever changes on generation

* remove debug prints

* remove redundant tf dummy objects

* add from config, get config and max length settings to address review

* let flake ignore the error on distillation you are welcome

* test from config

* add padding test

* address sgugger review

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>

* Fix init import_structure sorting

* Fix rebase

* fix torch meshgrid warnings

* support lower torch versions

* don't edit examples

* dont edit examples

* fix ci

* fix style

* rebase cleanup

* fix ci again

* Add Donut image processor

* Update src/transformers/image_transforms.py
Co-authored-by: Alara Dirik <8944735+alaradirik@users.noreply.github.com>

* Fix docstrings

* Full var names in docstring
Co-authored-by: Alara Dirik <8944735+alaradirik@users.noreply.github.com>

* extract warning from CI artifacts

* fix path

* fix logic

* fix comment

* update default values
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>

* First draft

* Fix backwards compatibility

* More fixes

* More fixes

* Make backbone more general

* Improve backbone

* Improve test

* Fix config checkpoint

* Address comments

* Use model_type

* Address more comments

* Fix special model names

* Remove MaskFormerSwinModel and MaskFormerSwinPreTrainedModel from main init

* Fix typo

* Update backbone

* Apply suggestion
Co-authored-by: Niels Rogge <nielsrogge@Nielss-MacBook-Pro.local>

* chore: add link to the video cls notebook.

* chore: segregate as resources.

* Add hidden states and attentions to backbone outputs

* Update ResNet

* Fix more tests

* Debug test

* Fix test_determinism

* Fix test_save_load

* Remove file

* Disable fx tests

* Test

* Add fx support for backbones
Co-authored-by: Niels Rogge <nielsrogge@Nielss-MacBook-Pro.local>

* INtegrate safetensos in weight offloading

* Use safetensors checkpoint for offload when available

* Make naming consistent

* Make load faster

* Quality

* Add default

* [FLAX] Add dtype to embedding for bert/bart/opt/t5

* Fix all copies

* Add a test case

* replace 4 asserts with ValueError exception for control flow

* Update examples/flax/image-captioning/run_image_captioning_flax.py
Co-authored-by: Sanchit Gandhi <93869735+sanchit-gandhi@users.noreply.github.com>

* Update examples/flax/image-captioning/run_image_captioning_flax.py
Co-authored-by: Sanchit Gandhi <93869735+sanchit-gandhi@users.noreply.github.com>

* reformatted file

* uninstalled trasformers and applied make style
Co-authored-by: Bibi <Bibi@katies-mac.local>
Co-authored-by: Sanchit Gandhi <93869735+sanchit-gandhi@users.noreply.github.com>

* fsmt doctest

* Update
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>

* Changed assert into 7-8 exceptions

* updated syntax error

* updated error

* updated file (Co-autho: Batese2001)

* Successful test on test_modeling_distilbert.py 

Successful raising errors and exceptions on the revised code in test_modeling_distilbert.py .

Co-credit: @batese2001

* Delete test_modeling_distilbert.ipynb

* Update modeling_distilbert.py

* Successful raising of exceptions with the conditions that are contrary to defined condition that asserts statements (Co-author: Batese2001)

* Successful raising of exceptions with the conditions that are contrary to defined condition that asserts statements (Co-author: Batese2001)

* committing the reformatted distilbert model

* reformatted distilbert model

* reformatted distilbert model

* reformatted distilbert model

* reformatted distilbert model with black

* Changed comments that explain better about raising exceptions for not having the even number of multi heads

* Changed comments that explain better about raising exceptions for not having the even number of multi heads

* changed based on the feedback

* Changed line 833 based on the suggestion made from @younesbelkada

* Changed line 833 based on the suggestion made from @younesbelkada draft2

* reformatted file

* Update src/transformers/models/distilbert/modeling_distilbert.py

* Update src/transformers/models/distilbert/modeling_distilbert.py
Co-authored-by: Younes Belkada <49240599+younesbelkada@users.noreply.github.com>

with pytorch cpu only version. without --no_cuda, using --bf16 will trigger error like "Your setup doesn't support bf16/gpu. You need torch>=1.10, using Ampere GPU with cuda>=11.0" (#20445)

make tensors in function build_relative_position created on proper device instead of always on cpu (#20434)
Co-authored-by: wenhanli <wenhanli@tencent.com>

* Add a test to ensure int dummy inputs are int64

* Move the test into the existing int64 test and update a lot of existing dummies

* Fix remaining dummies

* Fix remaining dummies

* Test for int64 serving sigs as well

* Update core tests to use tf.int64

* Add better messages to the assertions

* Update all serving sigs to int64

* More sneaky hiding tf.int32s

* Add an optional int32 signature in save_pretrained

* make fixup

* Add Amy's suggestions

* Switch all serving sigs back to tf.int32

* Switch all dummies to tf.int32

* Adjust tests to check for tf.int32 instead of tf.int64

* Fix base dummy_inputs dtype

* Start casting to tf.int32 in input_processing

* Change dtype for unpack_inputs test

* Add proper tf.int32 test

* Make the alternate serving signature int64

* Re-enable TF ESM tests, make sure we use facebook checkpoints

* make fixup

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>

* Fix typo

* Update tokenization_fsmt.py

* fix docstring

* fix 2

* add details