riaz.somc/transformers
1
0
Fork 0
mirror of https://github.com/huggingface/transformers.git synced 2025-07-24 23:08:57 +06:00
Code Issues Actions 35 Packages Projects Releases Wiki Activity
5,342 Commits 887 Branches 229 Tags 2.1 GiB
9e9a1fb8c7
Commit Graph

5342 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Teven
9e9a1fb8c7
Adding gradient checkpointing to GPT2 (#7446) 
* 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>
 2020-09-29 12:26:26 -04:00
Sylvain Gugger
52e8392b7e
Add automatic best model loading to Trainer (#7431) 
* Add automatic best model loading to Trainer

* Some small fixes

* Formatting
 2020-09-29 10:41:18 -04:00
Sylvain Gugger
1fc4de69ed
Document new features of make fixup (#7434) 2020-09-29 03:56:57 -04:00
GmailB
205bf0b7ea
Update README.md (#7444) 
Hi, just corrected the example code, add 2 links and fixed some typos
 2020-09-29 03:18:01 -04:00
Sam Shleifer
74d8d69bd4
[s2s] consistent output format across eval scripts (#7435) 2020-09-28 23:20:03 -04:00
Typicasoft
671b278e25
Create README.md (#7436) 
* Create README.md

MagBERT-NER : Added widget (Text)

* Rename model_cards/README.md to model_cards/TypicaAI/magbert-ner/README.md
 2020-09-28 18:25:25 -04:00
Manuel Romero
a1a8ffa512
Update README.md (#7429) 
Add links to models fine-tuned on a downstream task
 2020-09-28 13:40:09 -04:00
Stas Bekman
f62f2ffdcc
[makefile] 10x speed up checking/fixing (#7403) 
* [makefile] check/fix only modified since branching files

* fix phonies

* parametrize dirs

* have only one source for dirs to check

* look ma, no autoformatters here
 2020-09-28 10:45:42 -04:00
Lysandre
16c213820e Update docs to version v3.3.0 2020-09-28 16:32:00 +02:00
Lysandre
0613f05226 Release: v3.3.0 2020-09-28 16:24:43 +02:00
Sylvain Gugger
ca3fc36de3
Reorganize documentation navbar (#7423) 
* Reorganize documentation navbar

* Update css to have clear sections
 2020-09-28 16:22:58 +02:00
Lysandre Debut
7f4115c099
Pull request template (#7392) 
co-authored-by: sgugger <sylvain.gugger@gmail.com>

Co-authored-by: sgugger <sylvain.gugger@gmail.com>
 2020-09-28 09:51:49 -04:00
Sylvain Gugger
0611eab5e3
Document RAG again (#7377) 
Do not merge before Monday
 2020-09-28 08:31:46 -04:00
Sylvain Gugger
7563d5a3cf
Catch PyTorch warning when saving/loading scheduler (#7401) 2020-09-28 08:20:10 -04:00
Boris Dayma
1749ca317e
docs: fix model sharing file names (#5855) 
* docs: fix model sharing file names

* Update docs/source/model_sharing.rst

Co-authored-by: Julien Chaumond <chaumond@gmail.com>

* docs(model_sharing.rst): fix new line

Co-authored-by: Julien Chaumond <chaumond@gmail.com>
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
 2020-09-28 08:17:30 -04:00
Patrick von Platen
8279471506
correct RAG model cards (#7420) 2020-09-28 11:08:39 +02:00
Marcin Zabłocki
4083a55ab0
Flos fix (#7384) 2020-09-28 04:09:26 -04:00
Ola Piktus
ae3e84f3ba
[RAG] Clean Rag readme in examples (#7413) 
* Improve README + consolidation script

* Reformat README

* Reformat README

Co-authored-by: Your Name <you@example.com>
 2020-09-28 10:06:39 +02:00
Sam Shleifer
748425d47d
[T5] allow config.decoder_layers to control decoder size (#7409) 
* Working assymmetrical T5

* rename decoder_layers -> num_decoder_layers

* Fix docstring

* Allow creation of asymmetric t5 students
 2020-09-28 03:08:04 -04:00
Sam Shleifer
7296fea1d6
[s2s] rougeLSum expects \n between sentences (#7410) 
Co-authored-by: Swetha Mandava <smandava@nvidia.com>
 2020-09-27 16:27:19 -04:00
Suraj Patil
eab5f59682
[s2s] add create student script (#7290) 
Co-authored-by: Suraj Patil <surajp815@gmail.com>
Co-authored-by: Sam Shleifer <sshleifer@gmail.com>
 2020-09-27 15:10:46 -04:00
Patrick von Platen
e50a931c11
[Longformer, Bert, Roberta, ...] Fix multi gpu training (#7272) 
* fix multi-gpu

* fix longformer

* force to delete unnecessary layers

* fix notifications

* fix warning

* fix roberta

* fix tests

* remove hasattr

* fix tests

* fix roberta

* merge and clean authorized keys
 2020-09-25 20:33:21 +02:00
Patrick von Platen
2c8ecdf8a8
fix rag retriever save pretrained (#7399) 2020-09-25 19:47:12 +02:00
Patrick von Platen
1a14687e6f
Update README.md 2020-09-25 19:43:48 +02:00
Patrick von Platen
3327c2b0f6
Update README.md 2020-09-25 19:43:36 +02:00
Ola Piktus
fe326bd5cf
Remove dependency on examples/seq2seq from rag (#7395) 
Co-authored-by: Your Name <you@example.com>
 2020-09-25 18:20:49 +02:00
Sylvain Gugger
ad39271ae8
Fix FP16 and attention masks in FunnelTransformer (#7374) 
* Fix #7371

* Fix training

* Fix test values

* Apply the fix to TF as well
 2020-09-25 12:20:39 -04:00
Patrick von Platen
4e5b036bdd
Update README.md 2020-09-25 18:16:46 +02:00
Patrick von Platen
55eccfbb49
Update README.md 2020-09-25 18:16:44 +02:00
Sylvain Gugger
e2e77f02c2
Fix BartModel output documentation (#7390) 2020-09-25 11:48:13 -04:00
Sylvain Gugger
bbb07830ff
Speedup check_copies script (#7394) 2020-09-25 11:47:22 -04:00
Stas Bekman
8859c4f841
[code quality] new make target that combines style and quality targets (#7310) 
* [code quality] merge style and quality targets

Any reason why we don't run `flake8` in `make style`? I find myself needing to run `make style` and `make quality` all the time, but I need the latter just for the last 2 checks. Since we have no control over the source code why bother with separating checking and fixing - let's just have one target that fixes and then performs the remaining checks, as we know the first two have been done already.

This PR suggests to merge the 2 targets into one efficient target.

I will edit the docs if this change resonates with the team.

* move checks into style, re-use target

* better name

* add fixup target

* document new target
 2020-09-25 11:37:40 -04:00
Sam Shleifer
38a1b03f4d
Remove unhelpful bart warning (#7391) 2020-09-25 11:01:07 -04:00
Patrick von Platen
5ff0d6d7d0
Update README.md 2020-09-25 16:58:29 +02:00
Quentin Lhoest
cf1c88e092
[RAG] Fix retrieval offset in RAG's HfIndex and better integration tests (#7372) 
* Fix retrieval offset in RAG's HfIndex

* update slow tests

* style

* fix new test

* style

* add better tests

Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
 2020-09-25 16:12:46 +02:00
Patrick von Platen
571c7a11c1
[Rag] Fix wrong usage of num_beams and bos_token_id in Rag Sequence generation (#7386) 
* fix_rag_sequence

* add second bug fix
 2020-09-25 14:35:49 +02:00
Suraj Patil
415071b4c2
doc changes (#7385) 2020-09-25 08:00:36 -04:00
Patrick von Platen
2dd652d757
[RAG] Add missing doc and attention_mask to rag (#7382) 
* add docs

* add missing docs and attention_mask in fine-tune
 2020-09-25 11:23:55 +02:00
Lysandre Debut
7cdd9da5bf
Check config type using type instead of isinstance (#7363) 
* Check config type instead of instance


Bad merge

* Remove for loops

* Style
 2020-09-25 05:09:09 -04:00
Sam Shleifer
3c6bf8998f
modeling_bart: 3 small cleanups that dont change outputs (#7381) 
* Mbart passing

* boom boom

* cleaner assert

* add assert

* Fix tests
 2020-09-25 04:24:14 -04:00
Suraj Patil
9e68d075a4
Seq2SeqTrainer (#6769) 
Co-authored-by: Sam Shleifer <sshleifer@gmail.com>
 2020-09-24 18:46:58 -04:00
Sam Shleifer
d9d0f1140b
[s2s] distributed eval allows num_return_sequences > 1 (#7254) 2020-09-24 17:30:09 -04:00
Patrick von Platen
0804d077c6
correct attention mask (#7373) 2020-09-24 23:22:04 +02:00
Stas Bekman
a8cbc4269c
[fsmt] build/test scripts (#7257) 
Co-authored-by: Sam Shleifer <sshleifer@gmail.com>
 2020-09-24 17:10:26 -04:00
Sylvain Gugger
a8e7982f84
Remove mentions of RAG from the docs (#7376) 
* Remove mentions of  RAG from the docs

* Deactivate check
 2020-09-24 17:07:14 -04:00
Stas Bekman
eadd870b2f
[seq2seq] make it easier to run the scripts (#7274) 2020-09-24 15:23:48 -04:00
Lysandre Debut
8d3bb781ee
Formatter (#7368) 
* Formatter

* Docs
 2020-09-24 10:59:21 -04:00
Teven
7dfdf793bb
Fixing case in which Trainer hung while saving model in distributed training (#7365) 
* remote debugging

* remote debugging

* moved _store_flos call

* moved _store_flos call

* moved _store_flos call

* removed debugging artefacts
 2020-09-24 09:56:40 -04:00
Sylvain Gugger
0ccb6f5c6d
Clean RAG docs and template docs (#7348) 
* Clean RAG docs and template docs

* Fix typo

* Better doc
 2020-09-24 09:24:41 -04:00
Sylvain Gugger
27174bd4fe
Make PyTorch model files independent from each other (#7352) 2020-09-24 08:53:54 -04:00