GitLab

* [Trainer] Report both steps and num samples per second

* Fix batch number

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

* Address review comments
Co-authored-by: Stas Bekman <stas00@users.noreply.github.com>

* typo2

* fix typo

* fixing flos bug/typo in non-distributed setting

* storing flos every logging_interval

* Switch mem metrics flag

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

* fix_torch_device_generate_test

* remove @

* change pytorch import to flax import

get_length_grouped_indices() in LengthGroupedSampler and DistributedLengthGroupedSampler
is prohibitively slow for large number of megabatches (in test case takes hours for ~270k
megabatches with 100 items each) due to slow list concatenation with sum(megabatches, []).

Resolves: #11795
Co-authored-by: ctheodoris <cvtheodo@ds.dfci.harvard.edu>

* fix_torch_device_generate_test

* remove @

* add flax glue link

* support zero.Init in from_config

* no need for eval test

* fix_torch_device_generate_test

* remove @

* correct best seed for flax fine-tuning
Co-authored-by: Patrick von Platen <patrick@huggingface.co>

* speed up flax glue

* remove unnecessary line

* remove folder

* remove run in loop
Co-authored-by: Patrick von Platen <patrick@huggingface.co>

* add separator for windows

* fixes test_is_copy_consistent on Windows

* fixing writing encoding issue on extended test (for Windows)

* resolving comments

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>

* Fix regression in regression

* Add test

* add roformer

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

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

* update

* add TFRoFormerSinusoidalPositionalEmbedding and fix TFMarianSinusoidalPositionalEmbedding

* update docs

* make style and make quality

* roback

* unchanged

* rm copies from , this is a error in TFMarianSinusoidalPositionalEmbedding

* update Copyright year

* move # Add modeling imports here to the correct position

* max_position_embeddings can be set to 1536

* # Copied from transformers.models.bert.modeling_bert.BertOutput with Bert->RoFormer

* # Copied from transformers.models.bert.modeling_bert.BertLayer.__init__ with Bert->RoFormer

* update tokenization_roformer

* make style

* add staticmethod apply_rotary_position_embeddings

* add TF staticmethod apply_rotary_position_embeddings

* update torch apply_rotary_position_embeddings

* fix tf apply_rotary_position_embeddings error

* make style

* add pytorch RoFormerSelfAttentionRotaryPositionEmbeddingTest

* add TF rotary_position_embeddings test

* update test_modeling_rofomer

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

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

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

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

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

* refact roformer tokenizer

* add RoFormerTokenizerFast

* add RoFormerTokenizationTest

* add require_jieba

* update Copyright

* update tokenizer & add copy from

* add option rotary_value

* use rust jieba

* use rjieba

* use rust jieba

* fix test_alignement_methods

* slice normalized_string is too slow

* add config.embedding_size when embedding_size!=hidden_size

* fix pickle tokenizer

* Update docs/source/model_doc/roformer.rst
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* make style and make quality
Co-authored-by: Suraj Patil <surajp815@gmail.com>
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* refactor

* update

* update

* update

* refactor run mlm

* finalize

* refactor more

* fix typo

* update

* finish refactor

* modify run mlm

* Apply suggestions from code review

* Apply suggestions from code review

* Apply suggestions from code review

* small fixes

* upload

* upload

* finish run mlm script
Co-authored-by: Patrick von Platen <patrick@huggingface.co>

* fix_torch_device_generate_test

* remove @

* Add missing head masking for generate() function

* Add head_mask, decoder_head_mask and cross_attn_head_mask
into prepare_inputs_for_generation for generate() function
for multiple encoder-decoder models.

* Add test_genereate_with_head_masking

* [WIP] Update the new test and handle special cases

* make style

* Omit ProphetNet test so far

* make fix-copies

* flax gpt2

* combine masks

* handle shared embeds

* add causal LM sample

* style

* add tests

* style

* fix imports, docs, quality

* don't use cache

* add cache

* add cache 1st version

* make use cache work

* start adding test for generation

* finish generation loop compilation

* rewrite test

* finish

* update

* update

* apply sylvains suggestions

* update

* refactor

* fix typo
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* Add Flax Examples README

* Apply suggestions from code review

* Update examples/flax/README.md

* add nice table

* fix

* fix

* apply suggestions

* upload

* finish flax readme.md
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* add `dataset_name` to data_args and added accuracy metric

* added documentation for dataset_name

* spelling correction

* Fixed: Better names for nlp variables in pipelines' tests and docs.

* Fixed: Better variable names

* add headers to main doc

* Apply suggestions from code review

* update

* upload

* [TokenClassification] Label realignment for subword aggregation

Tentative to replace https://github.com/huggingface/transformers/pull/11622/files



- Added `AggregationStrategy`
- `ignore_subwords` and `grouped_entities` arguments are now fused
  into `aggregation_strategy`. It makes more sense anyway because
  `ignore_subwords=True` with `grouped_entities=False` did not have a
  meaning anyway.
- Added 2 new ways to aggregate which are MAX, and AVERAGE
- AVERAGE requires a bit more information than the others, for now this
case is slightly specific, we should keep that in mind for future
changes.
- Testing has been modified to reflect new argument, and to check the
correct deprecation and the new aggregation_strategy.
- Put the testing argument and testing results for aggregation_strategy,
close together, so that readers can understand what is supposed to
happen.
- `aggregate` is now only tested on a small model as it does not mean
anything to test it globally for all models.
- Previous tests are unchanged in desired output.
- Added a new test case that showcases better the difference between the
  FIRST, MAX and AVERAGE strategies.

* Wrong framework.

* Addressing three issues.

1- Tags might not follow B-, I- convention, so any tag should work now
(assumed as B-TAG)
2- Fixed an issue with average that leads to a substantial code change.
3- The testing suite was not checking for the "index" key for "none"
strategy. This is now fixed.

The issue is that "O" could not be chosen by AVERAGE strategy because
those tokens were filtered out beforehand, so their relative scores were
not counted in the average. Now filtering on
ignore_labels will happen at the very end of the pipeline fixing
that issue.
It's a bit hard to make sure this stays like that because we do
not have a end-to-end test for that behavior

* Formatting.

* Adding formatting to code + cleaner handling of B-, I- tags.
Co-authored-by: Francesco Rubbo <rubbo.francesco@gmail.com>
Co-authored-by: elk-cloner <rezakakhki.rk@gmail.com>

* Typo.
Co-authored-by: Francesco Rubbo <rubbo.francesco@gmail.com>
Co-authored-by: elk-cloner <rezakakhki.rk@gmail.com>

* improve tests

* remove bogus file

* make style
Co-authored-by: Patrick von Platen <patrick@huggingface.co>