c-my/transformers
1
0 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
6,154 Commits 171 Branches 49 Tags
7db700e4d3ecea3e8a20cb2d3d9d791b352ecc5f
Commit Graph
6154 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
TevenLeScao 7db700e4d3 initial commit 2020-12-15 19:20:33 +01:00
patrickvonplaten 6d12bbefa5 :wqMerge branch 'new_flax_design' of https://github.com/patrickvonplaten/transformers into new_flax_design 2020-12-15 16:24:26 +00:00
patrickvonplaten fddd073626 fix copies 2020-12-15 16:24:21 +00:00
Patrick von Platen 56fc622b3d Update src/transformers/modeling_flax_utils.py 2020-12-15 17:11:14 +01:00
patrickvonplaten 22870e3866 add pretrained to init 2020-12-15 16:07:34 +00:00
patrickvonplaten 5d11af74c1 add docs 2020-12-15 16:03:53 +00:00
patrickvonplaten 0a63bd61eb fix bug 2020-12-15 15:38:50 +00:00
patrickvonplaten 2ef4c11e27 finalize revert 2020-12-15 15:38:16 +00:00
patrickvonplaten 3fb41e14c9 revert tevens changes 2 2020-12-15 15:33:29 +00:00
patrickvonplaten d16be14cb6 revert tevens changes 2020-12-15 15:30:55 +00:00
patrickvonplaten 0f34499e8e make 0.3.0 compatible 2020-12-15 14:54:41 +00:00
patrickvonplaten 0f729ba269 update init 2020-12-15 12:36:38 +00:00
Patrick von Platen 8cd2a30cd9 Merge remote-tracking branch 'main/master' into new_flax_design 2020-12-15 00:21:03 +01:00
Patrick von Platen 7f87402089 apply sylvains tips 2020-12-15 00:19:39 +01:00
Patrick von Platen 2a1ac551e9 improve test in flax 2020-12-15 00:12:17 +01:00
patrickvonplaten f0675a2c38 fix flax from pretrained 2020-12-14 22:51:01 +00:00
Sylvain Gugger 251eb70c97 Also pin TF CPU 2020-12-14 16:17:04 -05:00
Sylvain Gugger e4ef57a9bb Pin TF to < 2.4 2020-12-14 16:06:30 -05:00
TevenLeScao adf4bc6c00 reverted training_args.py changes 2020-12-14 21:46:28 +01:00
TevenLeScao d8b9fd7931 removed tokenizers warning hack, fixed model re-initialization 2020-12-14 21:42:34 +01:00
TevenLeScao fcaad14a3a run_mlm_flax improvements: improper model inputs bugfix + automatic dataset splitting + tokenizers parallelism warning + avoiding warmup_steps=0 bug 2020-12-14 21:01:57 +01:00
TevenLeScao 3bb62e4333 Merge branch 'performer' into new_flax_design 2020-12-14 20:03:33 +01:00
TevenLeScao 1de19cbe23 proper initialization 2020-12-14 19:07:49 +01:00
Julien Plu df3f4d2aef Fix T5 and BART for TF (#9063) 
* Fix T5 for graphe compilation+execution

* Fix BART

* Fix import

* Fix naming

* fix attribute name

* Oops

* fix import

* fix tests

* fix tests

* Update test

* Add mising import

* Address Patrick's comments

* Style

* Address Patrick's comment
 2020-12-14 18:47:00 +01:00
Ahmed ElnaggarandAhmed Elnaggar a9c8bff724 Add parallelization support for T5EncoderModel (#9082) 
* add model parallelism to T5EncoderModel

add model parallelism to T5EncoderModel

* remove decoder from T5EncoderModel parallelize

* uodate T5EncoderModel docs

* Extend T5ModelTest for T5EncoderModel

* fix T5Stask using range for get_device_map

* fix style

Co-authored-by: Ahmed Elnaggar <elnaggar@rostlab.informatik.tu-muenchen.de>
 2020-12-14 12:00:45 -05:00
Stas Bekman b00eb4fb02 Testing Experimental CI Features (#9070) 2020-12-14 10:34:59 -05:00
Simon Brandeis 74daf1f954 Fixed a broken link in documentation (#9101) 2020-12-14 09:12:27 -05:00
Navjot d6af344c9e correct var name in TrainingArguments docstring (#9096) 2020-12-14 09:02:54 -05:00
TevenLeScao 754307e57a proper model initialization/loading 2020-12-14 14:59:31 +01:00
Patrick von Platen fa1ddced9e [RAG, Bart] Align RAG, Bart cache with T5 and other models of transformers (#9098) 
* fix rag

* fix slow test

* fix past in bart
 2020-12-14 12:32:26 +01:00
Lysandre Debut 6587cf9f84 Patch *ForCausalLM model (#9092) 2020-12-14 00:39:55 -05:00
Julien Plu 51d9c569fa Fix embeddings resizing in TF models (#8657) 
* Resize the biases in same time than the embeddings

* Trigger CI

* Biases are not reset anymore

* Remove get_output_embeddings + better LM model detection in generation utils

* Apply style

* First test on BERT

* Update docstring + new name

* Apply the new resizing logic to all the models

* fix tests

* Apply style

* Update the template

* Fix naming

* Fix naming

* Apply style

* Apply style

* Remove unused import

* Revert get_output_embeddings

* Trigger CI

* Update num parameters

* Restore get_output_embeddings in TFPretrainedModel and add comments

* Style

* Add decoder resizing

* Style

* Fix tests

* Separate bias and decoder resize

* Fix tests

* Fix tests

* Apply style

* Add bias resizing in MPNet

* Trigger CI

* Apply style
 2020-12-13 23:05:24 -05:00
TevenLeScao 9a0a30ba80 declaration order fix 2020-12-13 20:38:02 +01:00
TevenLeScao 025af73c6c initial_evaluation argument 2020-12-13 20:32:58 +01:00
TevenLeScao e0fcea4dd3 dirty print 2020-12-13 18:59:00 +01:00
TevenLeScao 1581b4b3fd dirty print 2020-12-13 18:46:50 +01:00
patrickvonplaten a455dd7cc1 make fix-copies 2020-12-13 17:44:23 +00:00
TevenLeScao ef167c3b03 smaller splits 2020-12-13 18:31:04 +01:00
TevenLeScao 2aedabbf09 preventing warmup_steps == 0 2020-12-13 15:26:14 +01:00
TevenLeScao ab2c38e7fd dirty print 2020-12-13 13:56:14 +01:00
TevenLeScao 87d1edcf59 splits 2020-12-12 21:45:04 +01:00
Patrick von Platen ff552aee9d remove Module from inits 2020-12-12 13:39:32 +00:00
Patrick von Platen a5bc339c4d fix gelu | gelu_new 2020-12-12 13:29:41 +00:00
Patrick von Platen 2787f5251b remove pooled from run_mlm_flax.py` 2020-12-12 12:58:38 +00:00
TevenLeScao 7ae3ed31f0 fixes in run_mlm_flax.py 2020-12-12 13:32:15 +01:00
Julien ChaumondSylvain GuggerPatrick von Platen
 		3552d0e0d8 [model_cards] Migrate cards from this repo to model repos on huggingface.co (#9013) 
* rm all model cards

* Update the .rst

@sgugger it is still not super crystal clear/streamlined so let me know if any ideas to make it simpler

* Add a rootlevel README.md with simple instructions/context

* Update docs/source/model_sharing.rst

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

* Apply suggestions from code review

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* make style

* rm all model cards

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
 2020-12-11 18:24:42 -05:00
Sylvain Gugger 29e4597950 Fix min_null_pred in the run_qa script (#9067) 2020-12-11 16:26:05 -05:00
Patrick von Platen 9cc9f4122e Make ProphetNetModel really compatible with EncoderDecoder (#9033) 
* improve

* finish

* upload model

* fix lm head

* fix test
 2020-12-11 16:59:54 +01:00
Patrick von Platen d655d510b8 last refactor 2020-12-11 15:57:48 +00:00
Patrick von Platen 2e8a3883ac delete keys file 2020-12-11 15:40:18 +00:00