GitLab

* Support for Bart and LayoutLM, and partial support for XLNet

* Support for mbart

* A lot of new models supported

* Support for other models

* LayoutLM fix

* Use strings instead of classes

* move test model folders (TODO: fix imports and others)

* fix (potentially partially) imports (in model test modules)

* fix (potentially partially) imports (in tokenization test modules)

* fix (potentially partially) imports (in feature extraction test modules)

* fix import utils.test_modeling_tf_core

* fix path ../fixtures/

* fix imports about generation.test_generation_flax_utils

* fix more imports

* fix fixture path

* fix get_test_dir

* update module_to_test_file

* fix get_tests_dir from wrong transformers.utils

* update config.yml (CircleCI)

* fix style

* remove missing imports

* update new model script

* update check_repo

* update SPECIAL_MODULE_TO_TEST_MAP

* fix style

* add __init__

* update self-scheduled

* fix add_new_model scripts

* check one way to get location back

* python setup.py build install

* fix import in test auto

* update self-scheduled.yml

* update slack notification script

* Add comments about artifact names

* fix for yolos
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>

* run sampling test on CPU to be deterministic

* input_ids on CPU

* Per-folder tests reorganization
Co-authored-by: sgugger <sylvain.gugger@gmail.com>
Co-authored-by: Stas Bekman <stas@stason.org>

* add xglm

* update vocab size

* fix model name

* style and tokenizer

* typo

* no mask token

* fix pos embed compute

* fix args

* fix tokenizer

* fix positions

* fix tokenization

* style and dic fixes

* fix imports

* add fast tokenizer

* update names

* add pt tests

* fix tokenizer

* fix typo

* fix tokenizer import

* fix fast tokenizer

* fix tokenizer

* fix converter

* add tokenizer test

* update checkpoint names

* fix tokenizer tests

* fix slow tests

* add copied from comments

* rst -> mdx

* flax model

* update flax tests

* quality

* style

* doc

* update index and readme

* fix copies

* fix doc

* update toctrr

* fix indent

* minor fixes

* fix config doc

* don't save embed_pos weights

* Apply suggestions from code review
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* address Sylvains commnets, few doc fixes

* fix check_repo

* align order of arguments

* fix copies

* fix labels

* remove unnecessary mapping

* fix saving tokenizer
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* Add GPTJForQuestionAnswering

* Reformat for GPTJForQuestionAnswering

* Fix isort error

* make style for GPTJForQA

* Add _keys_to_ignore_on_load_missing

* Change the sequence of qa and classification
Co-authored-by: Suraj Patil <surajp815@gmail.com>

* add flax gptj

* no bias in attention dense

* no wpe

* fix rotary embeddings

* fix rotary embeds

* fix rotray embeds

* quality

* doc and quality

* fix equivalence tests

* enable common tests, small fixes

* don't tie word embeds

* don't ignore lm_head

* Remove n_ctx from configs

* Fix GPTJ and OpenAIGPT, both are acceptable breaking changes as there are no configs such that it breaks

* Remove unecessary n_positions from TFOpenAIGPT

* Use fp16 checkpoints

* Style

* Fix outputs and disable OOM tests

* Correct another output

* Use a random smaller model for generation tests

* repo quickfix

* fix gradient checkpointing

* Make gradient_checkpointing a training argument

* Update src/transformers/modeling_utils.py
Co-authored-by: Stas Bekman <stas00@users.noreply.github.com>

* Update src/transformers/configuration_utils.py
Co-authored-by: Stas Bekman <stas00@users.noreply.github.com>

* Fix tests

* Style

* document Gradient Checkpointing as a performance feature

* Small rename

* PoC for not using the config

* Adapt BC to new PoC

* Forgot to save

* Rollout changes to all other models

* Fix typo
Co-authored-by: Stas Bekman <stas00@users.noreply.github.com>
Co-authored-by: Stas Bekman <stas@stason.org>

* Test GPTJ implementation

* Fixed conflicts

* Update __init__.py

* Update __init__.py

* change GPT_J to GPTJ

* fix missing imports and typos

* use einops for now
(need to change to torch ops later)

* Use torch ops instead of einsum

* remove einops deps

* Update configuration_auto.py

* Added GPT J

* Update gptj.rst

* Update __init__.py

* Update test_modeling_gptj.py

* Added GPT J

* Changed configs to match GPT2 instead of GPT Neo

* Removed non-existent sequence model

* Update configuration_auto.py

* Update configuration_auto.py

* Update configuration_auto.py

* Update modeling_gptj.py

* Update modeling_gptj.py

* Progress on updating configs to agree with GPT2

* Update modeling_gptj.py

* num_layers -> n_layer

* layer_norm_eps -> layer_norm_epsilon

* attention_layers -> num_hidden_layers

* Update modeling_gptj.py

* attention_pdrop -> attn_pdrop

* hidden_act -> activation_function

* Update configuration_gptj.py

* Update configuration_gptj.py

* Update configuration_gptj.py

* Update configuration_gptj.py

* Update configuration_gptj.py

* Update modeling_gptj.py

* Update modeling_gptj.py

* Update modeling_gptj.py

* Update modeling_gptj.py

* Update modeling_gptj.py

* Update modeling_gptj.py

* fix layernorm and lm_head size
delete attn_type

* Update docs/source/model_doc/gptj.rst
Co-authored-by: Suraj Patil <surajp815@gmail.com>

* removed claim that GPT J uses local attention

* Removed GPTJForSequenceClassification

* Update src/transformers/models/gptj/configuration_gptj.py
Co-authored-by: Lysandre Debut <lysandre@huggingface.co>

* Removed unsupported boilerplate

* Update tests/test_modeling_gptj.py
Co-authored-by: Suraj Patil <surajp815@gmail.com>

* Update src/transformers/models/gptj/modeling_gptj.py
Co-authored-by: Suraj Patil <surajp815@gmail.com>

* Update src/transformers/models/gptj/modeling_gptj.py
Co-authored-by: Suraj Patil <surajp815@gmail.com>

* Update src/transformers/models/gptj/modeling_gptj.py
Co-authored-by: Lysandre Debut <lysandre@huggingface.co>

* Update tests/test_modeling_gptj.py
Co-authored-by: Eric Hallahan <eric@hallahans.name>

* Update tests/test_modeling_gptj.py
Co-authored-by: Eric Hallahan <eric@hallahans.name>

* Update tests/test_modeling_gptj.py
Co-authored-by: Eric Hallahan <eric@hallahans.name>

* Update src/transformers/models/gptj/modeling_gptj.py
Co-authored-by: Suraj Patil <surajp815@gmail.com>

* Update __init__.py

* Update configuration_gptj.py

* Update modeling_gptj.py

* Corrected indentation

* Remove stray backslash

* Delete .DS_Store

* Delete .DS_Store

* Delete .DS_Store

* Delete .DS_Store

* Delete .DS_Store

* Update docs to match

* Remove tf loading

* Remove config.jax

* Remove stray `else:` statement

* Remove references to `load_tf_weights_in_gptj`

* Adapt tests to match output from GPT-J 6B

* Apply suggestions from code review
Co-authored-by: Suraj Patil <surajp815@gmail.com>

* Default `activation_function` to `gelu_new`

- Specify the approximate formulation of GELU to ensure parity with the default setting of `jax.nn.gelu()`

* Fix part of the config documentation

* Revert "Update configuration_auto.py"

This reverts commit e9860e9c043b6ebf57a0e705044e9ec9ba2263bb.

* Revert "Update configuration_auto.py"

This reverts commit cfaaae4c4dc70f1fbe9abd60fc8bd0b863b8c011.

* Revert "Update configuration_auto.py"

This reverts commit 687788954fd0cfbc567fa1202d56a4ff9271944f.

* Revert "Update configuration_auto.py"

This reverts commit 194d024ea87d4fcef0dcb08e57f52c47511a9fc6.

* Hyphenate GPT-J

* Undid sorting of the models alphabetically

* Reverting previous commit

* fix style and quality issues

* Update docs/source/model_doc/gptj.rst
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/__init__.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update tests/test_modeling_gptj.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/gptj/modeling_gptj.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/__init__.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/gptj/modeling_gptj.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/gptj/modeling_gptj.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/gptj/configuration_gptj.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/gptj/configuration_gptj.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/gptj/configuration_gptj.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/gptj/modeling_gptj.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/gptj/modeling_gptj.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/gptj/modeling_gptj.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/gptj/modeling_gptj.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/gptj/modeling_gptj.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Replaced GPTJ-specific code with generic code

* Update src/transformers/models/gptj/modeling_gptj.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Made the code always use rotary positional encodings

* Update index.rst

* Fix documentation

* Combine attention classes

- Condense all attention operations into `GPTJAttention`
- Replicate GPT-2 and improve code clarity by renaming `GPTJAttention.attn_pdrop` and `GPTJAttention.resid_pdrop` to `GPTJAttention.attn_dropout` and `GPTJAttention.resid_dropout`

* Removed `config.rotary_dim` from tests

* Update test_modeling_gptj.py

* Update test_modeling_gptj.py

* Fix formatting

* Removed depreciated argument `layer_id` to `GPTJAttention`

* Update modeling_gptj.py

* Update modeling_gptj.py

* Fix code quality

* Restore model functionality

* Save `lm_head.weight` in checkpoints

* Fix crashes when loading with reduced precision

* refactor self._attn(...)` and rename layer weights"

* make sure logits are in fp32 for sampling

* improve docs

* Add `GPTJForCausalLM` to `TextGenerationPipeline` whitelist

* Added GPT-J to the README

* Fix doc/readme consistency

* Add rough parallelization support

- Remove unused imports and variables
- Clean up docstrings
- Port experimental parallelization code from GPT-2 into GPT-J

* Clean up loose ends

* Fix index.rst
Co-authored-by: kurumuz <kurumuz1@gmail.com>
Co-authored-by: Suraj Patil <surajp815@gmail.com>
Co-authored-by: Lysandre Debut <lysandre@huggingface.co>
Co-authored-by: Eric Hallahan <eric@hallahans.name>
Co-authored-by: Leo Gao <54557097+leogao2@users.noreply.github.com>
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
Co-authored-by: your_github_username <your_github_email>
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* Add GPT2ForTokenClassification

* Fix dropout exception for GPT2 NER

* Remove sequence label in test

* Change TokenClassifierOutput to TokenClassifierOutputWithPast

* Fix for black formatter

* Remove dummy

* Update docs for GPT2ForTokenClassification

* Fix check_inits ci fail

* Update dummy_pt_objects after make fix-copies

* Remove TokenClassifierOutputWithPast

* Fix tuple input issue
Co-authored-by: danielsejong55@gmail.com <danielsejong55@gmail.com>

* Expose get_config() on ModelTesters

* Typo

* seq classification changes

* fix tests

Cleaner and more scalable implementation of symbolic tracing with torch.fx, and provides support for new architectures:
- ALBERT
- DistilBERT
- MobileBERT
- MegatronBERT
- GPT2
- GPT Neo
Co-authored-by: Michael Benayoun <michael@huggingface.co>

* GPT2DoubleHeadsModel made parallelizeable

* GPT2DoubleHeadsModel added as parallelizeable onto the GPT2 test suite

* [WIP] Adding new parameter to `generate`:  `max_time`.

Generation by tokens number is sometimes a bit clunky because we don't
know how many tokens are good enough or even how many tokens are in
the payload (for pipelines users for instance). This leads to hard
to understand behavior.

This PR proposes a new argument `max_time` which is a float of seconds
for the allowed time for `generate` to run on.
Ideally combinations of `max_tokens=None`, `max_time=2` could be used to
generate as many tokens as possible within time budget.

NB: Another possible approach consists of passing a callback to `generate`
  putting the caller in charge of the actual decision of when to stop
  generating tokens. It opens the door to 'which args should we pass'
  to this callback. It's hard to imagine other use-cases for this
  early stopping behavior than time (that are not already covered by
  parameters of generate)

* Revamp with StoppingCriteria

* Removing deprecated mentions.

* Forgot arguments to stopping criteria.

* Readding max_length it's not just used as a stopping criteria.

* Default value for `stopping_criteria`.

* Address @patrickvonplaten comments.

- More docstrings
- Actual doc
- Include in global namespace
- Remove TF work.

* Put back `max_length` (deprecation different PR).

* Doc quality.

* Fixing old behavior without `stopping_criteria` but with `max_length`.

Making sure we don't break that in the future.

* Adding more tests for possible inconsistencies between

`max_length` and `stopping_criteria`.

* Fixing the torch imports.

* Update past_key_values in gpt2 (#9391)

* Update generation_utils, and rename some items

* Update modeling_gpt2 to avoid an error in gradient_checkpointing

* Remove 'reorder_cache' from util and add variations to XLNet, TransfoXL, GPT-2

* Change the location of '_reorder_cache' in modeling files

* Add '_reorder_cache' in modeling_ctrl

* Fix a bug of my last commit in CTRL

* Add '_reorder_cache' to GPT2DoubleHeadsModel

* Manage 'use_cache' in config of test_modeling_gpt2

* Clean up the doc string

* Update src/transformers/models/gpt2/modeling_gpt2.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Fix the doc string (GPT-2, CTRL)

* improve gradient_checkpointing_behavior
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* add tests

* make style and fix bart bug

* fix bart past key value edge case

* correct tf bart test

* fix gpt2 tf

* fix t5 test

* Add copyright everywhere missing

* Style

* gpt2 and t5 parallel modeling

* model_parallel utils update

* adding missing model_parallel_utils

Adds missing model_parallel_utils and reverses the changes to code in modeling_gpt2 and modeling_t5

* training_args reformat

Reformatted training_args

* style formatting

Style formatting doc string length on training_args and model_parallel_utils

* style changes

make style && make quality for training_args and model_parallel_utils.

* adding tests

* minor change in trainer

reverts loss calculation

* Update training_args.py

* Update training_args.py

added back docstring language for adam_beta1 and adam_beta2

* Update trainer.py

* Update src/transformers/trainer.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Fix style & rebase
Co-authored-by: Lysandre Debut <lysandre@huggingface.co>
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
Co-authored-by: LysandreJik <lysandre.debut@reseau.eseo.fr>

* Remove old deprecated arguments
Co-authored-by: LysandreJik <lysandre.debut@reseau.eseo.fr>

* Remove needless imports

* Fix tests
Co-authored-by: LysandreJik <lysandre.debut@reseau.eseo.fr>

* Use the CI to identify failing tests

* Remove from all examples and tests

* More default switch

* Fixes

* More test fixes

* More fixes

* Last fixes hopefully

* Use the CI to identify failing tests

* Remove from all examples and tests

* More default switch

* Fixes

* More test fixes

* More fixes

* Last fixes hopefully

* Run on the real suite

* Fix slow tests

* Fix passing token_type_ids during GPT2DoubleHeadsModel.generate() if used

and for GPT2LMHeadModel too

* Update tests to check token_type_ids usage in GPT2 models

* add training tests

* correct longformer

* fix docs

* fix some tests

* fix some more train tests

* remove ipdb

* fix multiple edge case model training

* fix funnel and prophetnet

* clean gpt models

* undo renaming of albert

* first draft

* show design proposition for new generate method

* up

* make better readable

* make first version

* gpt2 tests pass

* make beam search for gpt2 work

* add first encoder-decoder code

* delete typo

* make t5 work

* save indermediate

* make bart work with beam search

* finish beam search bart / t5

* add default kwargs

* make more tests pass

* fix no bad words sampler

* some fixes and tests for all distribution processors

* fix test

* fix rag slow tests

* merge to master

* add nograd to generate

* make all slow tests pass

* speed up generate

* fix edge case bug

* small fix

* correct typo

* add type hints and docstrings

* fix typos in tests

* add beam search tests

* add tests for beam scorer

* fix test rag

* finish beam search tests

* move generation tests in seperate file

* fix generation tests

* more tests

* add aggressive generation tests

* fix tests

* add gpt2 sample test

* add more docstring

* add more docs

* finish doc strings

* apply some more of sylvains and sams comments

* fix some typos

* make fix copies

* apply lysandres and sylvains comments

* final corrections on examples

* small fix for reformer

* Add support for gpt2 batch inferencing

* add test

* remove typo
Co-authored-by: patrickvonplaten <patrick.v.platen@gmail.com>

* Add GPT2ForSequenceClassification based on DialogRPT

* Better documentation

* Code quality

* clean T5

* fix t5 tests

* fix index typo

* fix tf common test

* fix examples

* change positional ordering for Bart and FSTM

* add signature test

* clean docs and add tests

* add docs to encoder decoder

* clean docs

* correct two doc strings

* remove sig test for TF Elektra & Funnel

* fix tf t5 slow tests

* fix input_ids to inputs in tf

* Update src/transformers/modeling_bart.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/modeling_bart.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* implement lysandre results

* make style

* fix encoder decoder typo

* fix tf slow tests

* fix slow tests

* renaming

* remove unused input
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* GPT2 gradient checkpointing

* find_unused_parameters removed if checkpointing

* find_unused_parameters removed if checkpointing

* Update src/transformers/configuration_gpt2.py
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* Added a test for generation with checkpointing

* Update src/transformers/configuration_gpt2.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* fix generate for GPT2 Double Head

* fix gpt2 double head model

* fix  bart / t5

* also add for no beam search

* fix no beam search

* fix encoder decoder

* simplify t5

* simplify t5

* fix t5 tests

* fix BART

* fix transfo-xl

* fix conflict

* integrating sylvains and sams comments

* fix tf past_decoder_key_values

* fix enc dec test

* Run new isort

* More changes

* Update CI, CONTRIBUTING and benchmarks

* fix distilbert

* fix typo

* add cross attention layers for gpt2

* make gpt2 cross attention work

* finish bert2gpt2

* add explicit comments

* remove attention mask since not yet supported

* revert attn mask in pipeline

* Update src/transformers/modeling_gpt2.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/modeling_encoder_decoder.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* cleanup torch unittests: part 2

* remove trailing comma added by isort, and which breaks flake

* one more comma

* revert odd balls

* part 3: odd cases

* more ["key"] -> .key refactoring

* .numpy() is not needed

* more unncessary .numpy() removed

* more simplification

* improve unit tests

this is a sample of one test according to the request in https://github.com/huggingface/transformers/issues/5973
before I apply it to the rest

* batch 1

* batch 2

* batch 3

* batch 4

* batch 5

* style

* non-tf template

* last deletion of check_loss_output