f6e74a63ca
Add m2m100 ( #10236 )
...
* m2m_100
* no layernorm_embedding
* sinusoidal positional embeddings
* update pos embeddings
* add default config values
* tokenizer
* add conversion script
* fix config
* fix pos embed
* remove _float_tensor
* update tokenizer
* update lang codes
* handle lang codes
* fix pos embeds
* fix spm key
* put embedding weights on device
* remove qa and seq classification heads
* fix convert script
* lang codes pn one line
* fix embeds
* fix tokenizer
* fix tokenizer
* add fast tokenizer
* style
* M2M100MT => M2M100
* fix copyright, style
* tokenizer converter
* vocab file
* remove fast tokenizer
* fix embeds
* fix tokenizer
* fix tests
* add tokenizer tests
* add integration test
* quality
* fix model name
* fix test
* doc
* doc
* fix doc
* add copied from statements
* fix tokenizer tests
* apply review suggestions
* fix urls
* fix shift_tokens_right
* apply review suggestions
* fix
* fix doc
* add lang code to id
* remove unused function
* update checkpoint names
* fix copy
* fix tokenizer
* fix checkpoint names
* fix merge issue
* style
2021-03-06 22:14:16 +05:30
fd01104435
Temporarily disable stale bot
2021-03-06 00:21:50 -05:00
88a951e3cc
offline mode for firewalled envs ( #10407 )
...
* offline mode start
* add specific values
* fix fallback
* add test
* better values check and range
* test that actually works
* document the offline mode
* Apply suggestions from code review
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
* more strict check
* cleaner test
* pt-only test
* style
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
2021-03-05 17:27:48 -08:00
90ecc29656
Refactoring checkpoint names for multiple models ( #10527 )
...
* Refactor checkpoint name in ALBERT and ALBERT_tf
* Refactor checkpoint name in BART and BART_tf
* Refactor checkpoint name in BERT generation
* Refactor checkpoint name in Blenderbot_tf
* Refactor checkpoint name in Blenderbot_small_tf
* Refactor checkpoint name in ConvBERT AND CONVBERT_TF
* Refactor checkpoint name in CTRL AND CTRL_TF
* Refactor checkpoint name in DistilBERT AND DistilBERT_TF
* Refactor checkpoint name in DistilBERT redo
* Refactor checkpoint name in Electra and Electra_tf
* Refactor checkpoint name in FlauBERT and FlauBERT_tf
* Refactor checkpoint name in FSMT
* Refactor checkpoint name in GPT2 and GPT2_tf
* Refactor checkpoint name in IBERT
* Refactor checkpoint name in LED and LED_tf
* Refactor checkpoint name in Longformer and Longformer_tf
* Refactor checkpoint name in Lxmert and Lxmert_tf
* Refactor checkpoint name in Marian_tf
* Refactor checkpoint name in MBART and MBART_tf
* Refactor checkpoint name in MobileBERT and MobileBERT_tf
* Refactor checkpoint name in mpnet and mpnet_tf
* Refactor checkpoint name in openai and openai_tf
* Refactor checkpoint name in pegasus_tf
* Refactor checkpoint name in reformer
* Refactor checkpoint name in Roberta and Roberta_tf
* Refactor checkpoint name in SqueezeBert
* Refactor checkpoint name in Transformer_xl and Transformer_xl_tf
* Refactor checkpoint name in XLM and XLM_tf
* Refactor checkpoint name in XLNET and XLNET_tf
* Refactor checkpoint name in BERT_tf
* run make tests, style, quality, fixup
2021-03-05 18:06:55 -05:00
defe9e20fe
Stale Bot ( #10509 )
...
* Add stale bot to Github Actions
* Update message
* Message for assignee
* Update scripts/stale.py
* Uncomment & stop testing
2021-03-05 16:41:50 -05:00
7da995c00c
Fix embeddings for PyTorch 1.8 ( #10549 )
...
* Fix embeddings for PyTorch 1.8
* Try with PyTorch 1.8.0
* Fix embeddings init
* Fix copies
* Typo
* More typos
2021-03-05 16:18:48 -05:00
3e056c1003
Typo correction. ( #10531 )
...
DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST => DEBERTA_V2_PRETRAINED_MODEL_ARCHIVE_LIST in line 31.
2021-03-05 15:27:09 -05:00
9f8bc87cbe
fixed dead link in trainer doc ( #10554 )
2021-03-05 14:56:37 -05:00
6b58e15507
Fix torch 1.8.0 segmentation fault ( #10546 )
...
* Only run one test
* Patch segfault
* Fix summarization pipeline
* Ready for merge
2021-03-05 12:10:19 -05:00
395ffcd757
fix run seq2seq ( #10547 )
2021-03-05 18:17:12 +03:00
54e55b52d4
Fixing conversation test for torch 1.8 ( #10545 )
2021-03-05 09:24:14 -05:00
dc9aaa3848
Pin torch to 1.7.1 in tests while we resolve issues
2021-03-05 07:57:35 -05:00
12b66215cf
Fix example of custom Trainer to reflect signature of compute_loss ( #10537 )
2021-03-05 07:44:53 -05:00
093b88f4e9
Update scatter to use torch 1.8.0
2021-03-05 07:31:51 -05:00
c503a1c15e
[ProphetNet] Bart-like Refactor ( #10501 )
...
* first step to refactor
* make all fast tests pass
* make all slow tests pass
* save intermediate
* correct cache
* finish PR
* make fp16 work
2021-03-04 23:27:12 +03:00
6290169eb3
Rework TPU checkpointing in Trainer ( #10504 )
...
* Rework TPU checkpointing in Trainer
* Wraps the barrier in a dist test
* Address review comments
* Remove line
2021-03-04 11:46:11 -05:00
805c5200dc
Removes overwrites for output_dir ( #10521 )
...
* removed overwrites
* remove default value for output_dir
* adjusted typing
2021-03-04 17:12:37 +01:00
a5bd40b75c
Not always consider a local model a checkpoint in run_glue ( #10517 )
2021-03-04 11:11:39 -05:00
745ea78dcc
Revert "Not always consider a local model a checkpoint in run_glue"
...
This reverts commit f3660613bc
2021-03-04 09:45:18 -05:00
f3660613bc
Not always consider a local model a checkpoint in run_glue
2021-03-04 09:44:02 -05:00
948b730f97
Remove unsupported methods from ModelOutput doc ( #10505 )
2021-03-03 14:55:18 -05:00
b70f441b72
Smp grad accum ( #10488 )
...
* Fix gradient accumulation for SM Model Parallelism
* Style and divide loss by grad accum steps
2021-03-03 12:13:29 -05:00
d064fb5647
Fix the bug in constructing the all_hidden_states of DeBERTa v2 ( #10466 )
...
* fix all_hidden_states
* use output_states instead of next_kv
2021-03-03 12:05:21 -05:00
188574ac50
remap MODEL_FOR_QUESTION_ANSWERING_MAPPING classes to names auto-generated file ( #10487 )
...
* remap classes to strings
* missing new util
* style
* doc
* move the autogenerated file
* Trigger CI
2021-03-03 08:54:00 -08:00
801ff969ce
Refactor checkpoint name in BERT and MobileBERT ( #10424 )
...
* Refactor checkpoint name in BERT and MobileBERT
* Add option to check copies
* Add QuestionAnswering
* Add last models
* Make black happy
2021-03-03 11:21:17 -05:00
39f70a4058
feat(docs): navigate with left/right arrow keys ( #10481 )
...
* feat(docs): navigate with left/right arrow keys
* fix: add missing comma
2021-03-03 11:17:12 -05:00
2d2ed2cc18
[T5] Fix speed degradation bug t5 ( #10496 )
...
* fix speed degradation bug t5
* fix for all models
* fix code quality
2021-03-03 12:42:41 +03:00
5dc303e281
Fixed minor spelling mistakes ( #10489 )
...
Co-authored-by: WybeKoper <WybeKoper@users.noreply.github.com>
2021-03-03 14:17:25 +05:30
1750e62900
Generate can return cross-attention weights too ( #10493 )
2021-03-03 13:57:02 +05:30
b013842244
Changed `num_beams` to `num_beams // num_beam_groups` when initialising `PrefixConstrainedLogitsProcessor` in `_get_logits_processor` to fix compatibility issue when constrained decoding is used together with grouped beam search ( #10475 )
2021-03-02 10:41:54 +03:00
0c2325198f
Add I-BERT to README ( #10462 )
2021-03-01 12:12:31 -05:00
9248e27037
Remove Anthony from the bug reports in Transformers
2021-03-01 10:23:40 -05:00
a106bde5a7
[Wav2Vec2FeatureExtractor] smal fixes ( #10455 )
...
* smal fixes
* don't check for None
2021-03-01 20:19:52 +05:30
11655fafdd
remove feature extraction config ( #10457 )
2021-03-01 12:30:12 +03:00
0234de8418
Add Fine-Tuning for Wav2Vec2 ( #10145 )
...
* add encode labels function to tokenizer
* start adding finetuning
* init dropout
* upload
* correct convert script
* apply changes
* fix second typo
* make first dummy training run
* adapt convert script
* push confg for comparison
* remove conf
* finish training
* adapt data collator
* add research folder
* update according to fairseq feedback
* some minor corrections
* refactor masking indices a bit
* some minor changes
* clean tokenizer
* finish clean-up
* remove previous logic
* update run script
* correct training
* finish changes
* finish model
* correct bug
* fix training a bit more
* add some tests
* finish gradient checkpointing
* finish example
* correct gradient checkpointing
* improve tokenization method
* revert changes in tokenizer
* revert general change
* adapt fine-tuning
* update
* save intermediate test
* Update README.md
* finish finetuning
* delete conversion script
* Update src/transformers/models/wav2vec2/configuration_wav2vec2.py
* Update src/transformers/models/wav2vec2/processing_wav2vec2.py
Co-authored-by: Lysandre Debut <lysandre@huggingface.co>
* finish wav2vec2 script
* finish wav2vec2 fine-tuning
* finalize test
* correct test
* adapt tests
* finish
* remove test file
Co-authored-by: Lysandre Debut <lysandre@huggingface.co>
2021-03-01 12:13:17 +03:00
3c733f3208
Update ibert.rst ( #10445 )
2021-02-28 19:03:49 +03:00
aeba4f95bb
Adds terms to Glossary ( #10443 )
...
* feat: Adds three definitions to glossary from @cronoik
Needed a definition for transformer which in turn needed 2 more definitions
To do with issue https://github.com/huggingface/transformers/issues/9078
* fix: Adjusts definition of neural network to make it easier to read
2021-02-28 08:27:54 -05:00
256482ac92
Introduce save_strategy training argument ( #10286 )
...
* Introduce save_strategy training argument
* deprecate EvaluationStrategy
* collapse EvaluationStrategy and LoggingStrategy into a single
IntervalStrategy enum
* modify tests to use modified enum
2021-02-27 19:34:22 -05:00
aca6288ff4
updated logging and saving metrics ( #10436 )
...
* updated logging and saving metrics
* space removal
2021-02-27 09:53:44 -08:00
f52a15897b
[run_seq2seq.py] restore functionality: saving to test_generations.txt ( #10428 )
...
This PR restores the original functionality that for some reason was modified.
Fixes: https://github.com/huggingface/transformers/issues/10381
@sgugger
2021-02-27 08:21:50 -08:00
311b7048c5
Fix conda-build ( #10431 )
2021-02-26 20:20:30 -05:00
ee04b69822
[examples] better model example ( #10427 )
...
* refactors
* typo
2021-02-26 17:01:01 -08:00
a85eb616f7
Ray Tune Integration Bug Fixes ( #10406 )
...
* fixes
* update resources
* formatting
* remove import
* add log statement
* use fstring
* add period
* Update src/transformers/integrations.py
2021-02-26 19:06:08 -05:00
98569d4ba2
Add Ray Tune hyperparameter search integration test ( #10414 )
2021-02-26 10:18:33 -05:00
d03695f3a2
[LED] Correct Docs ( #10419 )
...
* correct docs
* correct tf model docs as well
2021-02-26 17:53:28 +03:00
7fc686efb1
Sagemaker Model Parallel tensoboard writing fix ( #10403 )
...
* Added tb fix
* Removed local rank condition
* Updated reference to args
2021-02-26 08:04:55 -05:00
83d2d55c94
[ci, flax] non-existing models are unlikely to pass tests ( #10409 )
...
😂
2021-02-26 12:35:36 +03:00
17b6e0d474
Fix run_glue evaluation when model has a label correspondence ( #10401 )
2021-02-25 15:30:38 -05:00
26f8b2cb10
Make Barthez tokenizer tests a bit faster ( #10399 )
...
* Make Barthez tokenizer tests a bit faster
* Quality
2021-02-25 11:42:25 -05:00
b040e6efc1
Fix None in add_token_positions - issue #10210 ( #10374 )
...
* Fix None in add_token_positions - issue #10210
Fix None in add_token_positions related to the issue #10210
* add_token_positions fix None values in end_positions vector
add_token_positions fix None in end_positions vector as proposed by @joeddav
2021-02-25 09:18:33 -07:00
Suraj Patil
Lysandre
Stas Bekman
Daniel Hug
Lysandre Debut
Sylvain Gugger
Chen Liang
Joakim Warholm
Patrick von Platen
Nicolas Patry
lewtun
Philipp Schmid
Sylvain Gugger
felixgwu
Jeff Yang
WybeKoper
Mehrad Moradshahi
Martin Schmitt
Darigov Research
Tanmay Garg
Bhadresh Savani
Amog Kamsetty
Kai Fricke
Mansi Mane
Julien Chaumond
Andrea Bacciu