GitLab

* add forced logits processors

* delete adjust_logits method

* add forced_eos_token_id argument in config

* add tests for forced logits processors

* update gen utils tests

* add forced option to tf generate

* remove adjust_logits method from tf models

* update adjust_logits for marian

* delete _force_token_id_to_be_generated method

* style

* import warnings

* pass max_length to _get_logits_processor

* set forced_eos_token_id to None

* set forced attributes in conf utils

* typo

* fix rag generate

* add forced_eos_token_id in rag config

* remove force_bos_token_to_be_generated from BartConfig

* remove _force_token_ids_generation from FSMT

* nit

* fix negative constant

* apply suggestions from code review

* Fix test

* Remove commented test

* Fix name

* Apply style

* Fix check copies

* Remove prints

* Restore boolean

* Fix reshape

I assume the CI machine should have at least 4 cores, so let's build docs faster

* how to specify a specific gpu

* new paper

* expand on buffer sizes

* style

* where to find config examples

* specific example

* small updates

* doc: update W&B related doc

* doc(wandb): mention report_to

* doc(wandb): commit suggestion
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* doc(wandb): fix typo

* doc(wandb): remove WANDB_DISABLED
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* fix rag generate and tests

* put back adjust_logits_during_generation

* tests are okay
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* Replace tf.newaxis -> tf.expand_dims

* Fix tests

* Fix tests

* Use reshape when a tensors needs a double expand

* Fix GPT2

* Fix GPT2

* Add head masking to TF LED

* Add head_mask to Longformer + one doc piece to LED

* Fix integration tests

* Enable propagation by default

* Document enable/disable default handler

* add do_predict, pass eval_beams durig eval

* update help

* apply suggestions from code review

* Fix example in Wav2Vec2 documentation

* fix style

* add wav2vec2CTC and deprecate for maskedlm

* remove from docs

Removing run_pl_glue.py from text classification docs, include run_xnli.py & run_tf_text_classification.py (#10066)

* Removing run_pl_glue.py from seq classification docs

* Adding run_tf_text_classification.py

* Using :prefix_link: to refer local files

* Applying "make style" to the branch

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

* Removing last underscores
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* deepspeed bug fixes and tests

* manual wrap?

* Unify logging with f-strings

* Get limits from MLflow rather than hardcode

* Add a check for parameter length overflow

Also constants are marked as internal

* Don't stop run in on_train_end

This causes bad behaviour when there is a seprarte validation step:
validation gets recorded as separate run.

* Fix style

* replace -100 token ids with the tokenizer pad_id for compute_metrics

* fixed typo for label_ids

* claiming this issue

* Integration test for BertGeneration(Encoder and Decoder)

* fix code quality

* Fix template

* Fix template

* Refacto BERT

* Restore all the concerned models

* Remove print

* Update template

* Apply Sylvain's and Morgan's comments

* Fix cast

* Put the cast inside call

* Remove cond in ebds

* Fix funnel

* Restore previous dot product (attention_scores) computation

* Add ConvBERT and BART

* Make all the S2S models ONNX compliant

* Fix test

* Fix check copies

* Disable temporarily too slow tests

* Fix style

* Fix template

* Cleaning up `ConversationalPipeline` to support more than DialoGPT.

Currently ConversationalPipeline was heavily biased towards DialoGPT
,which is the default model for this pipeline.

This PR proposes changes to put back the modifications specific to
DialoGPT into tokenizer-specific behavior wherever possible, by
creating `_build_conversation_input_ids` function that takes
conversation as input, and returns a list of ints corresponding
to the tokens. It feels natural to put here because all models
have probably different strategies to build input_ids from the
full conversation and it's the tokenizer's job to transform strings
into tokens (and vice-versa)

If `_build_conversation_input_ids` is missing, previous behavior is
used so we don't break anything so far (except for blenderbot where it's a fix).

This PR also contains a fix for too long inputs. There used
to be dead code for trying to limit the size of incoming input.
The introduced fixed is that we limit
within `_build_conversation_input_ids` to `tokenizer.model_max_length`.
It corresponds to the intent of the removed dead code and is actually
better because it corresponds to `model_max_length` which is different
from `max_length` (which is a default parameter for `generate`).

- Removed `history` logic from the Conversation as it's not relevant
anymore because tokenization logic has been moved to tokenizer.
And tokenizer cannot save any cache, and conversation cannot know
what is relevant or not.
Also it's not usable from `blenderbot` because the input_ids are
not append only (EOS tokens is always at the end).

- Added `iter_texts` method on `Conversation` because all
the code was literred with some form of this iteration of
past/generated_responses.

* Removing torch mention in types.

* Adding type checking to `_build_conversation_input_ids`.

* Fixing import in strings.