- 26 Oct, 2020 10 commits
-
-
Sylvain Gugger authored
* Important files * Styling them all * Revert "Styling them all" This reverts commit 7d029395fdae8513b8281cbc2a6c239f8093503e. * Syling them for realsies * Fix syntax error * Fix benchmark_utils * More fixes * Fix modeling auto and script * Remove new line * Fixes * More fixes * Fix more files * Style * Add FSMT * More fixes * More fixes * More fixes * More fixes * Fixes * More fixes * More fixes * Last fixes * Make sphinx happy
-
Sylvain Gugger authored
* Fixes in preparation for doc styling * More fixes * Better syntax * Fixes * Style * More fixes * More fixes
-
Lysandre Debut authored
-
Sam Shleifer authored
-
Stas Bekman authored
-
Lysandre Debut authored
-
noise-field authored
* Add MLflow integration class Add integration code for MLflow in integrations.py along with the code that checks that MLflow is installed. * Add MLflowCallback import Add import of MLflowCallback in trainer.py * Handle model argument Allow the callback to handle model argument and store model config items as hyperparameters. * Log parameters to MLflow in batches MLflow cannot log more than a hundred parameters at once. Code added to split the parameters into batches of 100 items and log the batches one by one. * Fix style * Add docs on MLflow callback * Fix issue with unfinished runs The "fluent" api used in MLflow integration allows only one run to be active at any given moment. If the Trainer is disposed off and a new one is created, but the training is not finished, it will refuse to log the results when the next trainer is created. * Add MLflow integration class Add integration code for MLflow in integrations.py along with the code that checks that MLflow is installed. * Add MLflowCallback import Add import of MLflowCallback in trainer.py * Handle model argument Allow the callback to handle model argument and store model config items as hyperparameters. * Log parameters to MLflow in batches MLflow cannot log more than a hundred parameters at once. Code added to split the parameters into batches of 100 items and log the batches one by one. * Fix style * Add docs on MLflow callback * Fix issue with unfinished runs The "fluent" api used in MLflow integration allows only one run to be active at any given moment. If the Trainer is disposed off and a new one is created, but the training is not finished, it will refuse to log the results when the next trainer is created.
-
Lysandre Debut authored
-
luyug authored
* Add mixed precision evaluation * use original flag
-
Thomas Wolf authored
* fixing #8001 * make T5 tokenizer serialization more robust - style
-
- 24 Oct, 2020 1 commit
-
-
Suraj Patil authored
-
- 23 Oct, 2020 3 commits
-
-
Anthony MOI authored
-
Patrick von Platen authored
* remove reformer pad_token_id * fix pegasus
-
Thomas Wolf authored
[tests|tokenizers] Refactoring pipelines test backbone - Small tokenizers improvements - General tests speedups (#7970) * WIP refactoring pipeline tests - switching to fast tokenizers * fix dialog pipeline and fill-mask * refactoring pipeline tests backbone * make large tests slow * fix tests (tf Bart inactive for now) * fix doc... * clean up for merge * fixing tests - remove bart from summarization until there is TF * fix quality and RAG * Add new translation pipeline tests - fix JAX tests * only slow for dialog * Fixing the missing TF-BART imports in modeling_tf_auto * spin out pipeline tests in separate CI job * adding pipeline test to CI YAML * add slow pipeline tests * speed up tf and pt join test to avoid redoing all the standalone pt and tf tests * Update src/transformers/tokenization_utils_base.py Co-authored-by:
Sam Shleifer <sshleifer@gmail.com> * Update src/transformers/pipelines.py Co-authored-by:
Sylvain Gugger <35901082+sgugger@users.noreply.github.com> * Update src/transformers/pipelines.py Co-authored-by:
Lysandre Debut <lysandre@huggingface.co> * Update src/transformers/testing_utils.py Co-authored-by:
-
- 22 Oct, 2020 11 commits
-
-
Sam Shleifer authored
* Move NoLayerEmbedTokens * TFWrappedEmbeddings * Add comment
-
Sylvain Gugger authored
* Fix checkpoint loading in Trainer * Fix typo
-
Sylvain Gugger authored
* Only log total_flos at the end of training * Fix test
-
Julien Chaumond authored
* FillMaskPipeline: support passing top_k on __call__ Also move from topk to top_k * migrate to new param name in tests * Review from @sgugger
-
Sylvain Gugger authored
* Start simplification * More progress * Finished script * Address comments and update tests instructions * Wrong test * Accept files as inputs and fix test * Update src/transformers/trainer_utils.py Co-authored-by:
Julien Chaumond <chaumond@gmail.com> * Fix labels and add combined score * Add special labels * Update TPU command * Revert to old label strategy * Use model labels * Fix for STT-B * Styling * Apply suggestions from code review Co-authored-by:
Thomas Wolf <thomwolf@users.noreply.github.com> * Code styling * Fix review comments Co-authored-by:
-
Nicolas Patry authored
* Actually make the "translation", "translation_XX_to_YY" task behave correctly. Background: - Currently "translation_cn_to_ar" does not work. (only 3 pairs are supported) - Some models, contain in their config the correct values for the (src, tgt) pair they can translate. It's usually just one pair, and we can infer it automatically from the `model.config.task_specific_params`. If it's not defined we can still probably load the TranslationPipeline nevertheless. Proposed fix: - A simplified version of what could become more general which is a `parametrized` task. "translation" + (src, tgt) in this instance it what we need in the general case. The way we go about it for now is simply parsing "translation_XX_to_YY". If cases of parametrized task arise we should preferably go in something closer to what `datasets` propose which is having a secondary argument `task_options`? that will be close to what that task requires. - Should be backward compatible in all cases for instance `pipeline(task="translation_en_to_de") should work out of the box. - Should provide a warning when a specific translation pair has been selected on behalf of the user using `model.config.task_specific_params`. * Update src/transformers/pipelines.py Co-authored-by:
-
Funtowicz Morgan authored
Signed-off-by:Morgan Funtowicz <funtowiczmo@gmail.com>
-
Patrick von Platen authored
* fix config save * add test * add config class variable and another test * line break * fix fsmt and typo * god am I making many errors today :-/ * Update src/transformers/configuration_utils.py Co-authored-by:
-
wlhgtc authored
* ADD: add whole word mask proxy for both eng and chinese * MOD: adjust format * MOD: reformat code * MOD: update import * MOD: fix bug * MOD: add import * MOD: fix bug * MOD: decouple code and update readme * MOD: reformat code * Update examples/language-modeling/README.md Co-authored-by:
Sylvain Gugger <sylvain.gugger@gmail.com>
-
Haebin Shin authored
-
rmroczkowski authored
-
- 21 Oct, 2020 5 commits
-
-
Evan Pete Walsh authored
* fix docstring for 'special_tokens_mask' * revert auto formatter changes * revert another auto format * revert another auto format
-
Patrick von Platen authored
* correct xlm prophetnet auto model and examples * fix line-break docs
-
Fran莽ois Lagunas authored
Improved TensorBoard and Wandb integration, as well as optuna and ray/tune support, with minor modifications to trainer core code.
-
Stas Bekman authored
* make the save_load special key tests common * handle mbart * cleaner solution * fix * move test_save_load_missing_keys back into fstm for now * restore * style * add marian * add pegasus * blenderbot * revert - no static embed
-
Sam Shleifer authored
* half done * doc improvement * Cp test file * brokedn * broken test * undo some mess * ckpt * borked * Halfway * 6 passing * boom boom * Much progress but still 6 * boom boom * merged master * 10 passing * boom boom * Style * no t5 changes * 13 passing * Integration test failing, but not gibberish * Frustrated * Merged master * 4 fail * 4 fail * fix return_dict * boom boom * Still only 4 * prepare method * prepare method * before delete classif * Skip tests to avoid adding boilerplate * boom boom * fast tests passing * style * boom boom * Switch to supporting many input types * remove FIXMENORM * working * Fixed past_key_values/decoder_cached_states confusion * new broken test * Fix attention mask kwarg name * undo accidental * Style and reviewers * style * Docs and common tests * Cleaner assert messages * copy docs * style issues * Sphinx fix * Simplify caching logic * test does not require torch * copy _NoLayerEmbedTokens * Update src/transformers/modeling_tf_bart.py Co-authored-by:
-
- 20 Oct, 2020 4 commits
-
-
Lysandre authored
-
Shai Erera authored
I'm using transformers 3.3.1 and run a training script with `--save_total_limit 3`. I hit the exception below, and after debugging the code found that it wrongly tries to index into the `best_model_checkpoint`'s *str* rather than the `sorted_checkpoints` array. When running without the fix I got this exception: ``` Traceback (most recent call last): File "/<HOME>/.conda/envs/transformers/lib/python3.7/site-packages/transformers/trainer.py", line 921, in _save_training self._rotate_checkpoints(use_mtime=True) File "/<HOME>/.conda/envs/transformers/lib/python3.7/site-packages/transformers/trainer.py", line 1283, in _rotate_checkpoints checkpoints_sorted = self._sorted_checkpoints(use_mtime=use_mtime) File "/<HOME>/.conda/envs/transformers/lib/python3.7/site-packages/transformers/trainer.py", line 1274, in _sorted_checkpoints checkpoints_sorted[best_model_index], TypeError: 'str' object does not support item assignment ``` -
Sylvain Gugger authored
-
Stas Bekman authored
* rename skip targets + docs * fix quotes * style * Apply suggestions from code review Co-authored-by:
-
- 19 Oct, 2020 6 commits
-
-
Patrick von Platen authored
* fix encoder decoder models * add .gitignore
-
Bram Vanroy authored
* Raise error when using AMP on non-CUDA device * make style * make style
-
Ayub Subhaniya authored
Seeing error when sending `decoder_config` as a parameter while initializing a encoder-decoder model from pretrained. fixed "UnboundLocalError: local variable 'decoder_config' referenced before assignment"
-
Quentin Lhoest authored
* add CustomHFIndex * typo in config * update tests * add custom dataset example * clean script * update test data * minor in test * docs * docs * style * fix imports * allow to pass the indexed dataset directly * update tests * use multiset DPR * address thom and patrick's comments * style * update dpr tokenizer * add output_dir flag in use_own_knowledge_dataset.py * allow custom datasets in examples/rag/finetune.py * add test for custom dataset in distributed rag retriever
-
Julien Rossi authored
* fix 5990 * accomodate iterable dataset without predefined length * set it as 1 use case: provide max_steps, and NO num_epochs * Is a merge of master and PR 5995 * fix trainer test under TF * fix only for torch * TF trainer untouched * trainer tests are skipped when no torch * address comments * fix quality checks * remove torch.dataset from test_trainer * unnecessary inheritance * RegressionDataset implements all needed methods __len__ and __getitem__ * fix quality checks * restore RegressionDataset * was wrongly under is_torch_available()
-
Weizhen authored
* add new model prophetnet prophetnet modified modify codes as suggested v1 add prophetnet test files * still bugs, because of changed output formats of encoder and decoder * move prophetnet into the latest version * clean integration tests * clean tokenizers * add xlm config to init * correct typo in init * further refactoring * continue refactor * save parallel * add decoder_attention_mask * fix use_cache vs. past_key_values * fix common tests * change decoder output logits * fix xlm tests * make common tests pass * change model architecture * add tokenizer tests * finalize model structure * no weight mapping * correct n-gram stream attention mask as discussed with qweizhen * remove unused import * fix index.rst * fix tests * delete unnecessary code * add fast integration test * rename weights * final weight remapping * save intermediate * Descriptions for Prophetnet Config File * finish all models * finish new model outputs * delete unnecessary files * refactor encoder layer * add dummy docs * code quality * fix tests * add model pages to doctree * further refactor * more refactor, more tests * finish code refactor and tests * remove unnecessary files * further clean up * add docstring template * finish tokenizer doc * finish prophetnet * fix copies * fix typos * fix tf tests * fix fp16 * fix tf test 2nd try * fix code quality * add test for each model * merge new tests to branch * Update model_cards/microsoft/prophetnet-large-uncased-cnndm/README.md Co-authored-by:
weizhen <weizhen@mail.ustc.edu.cn> Co-authored-by:
Patrick von Platen <patrick.v.platen@gmail.com> Co-authored-by:
-
