Skip to content

GitLab

Switch branch/tag
  1. 13 Apr, 2022 1 commit
  3. 12 Apr, 2022 4 commits
  5. 11 Apr, 2022 6 commits
  7. 08 Apr, 2022 1 commit
    • NielsRogge's avatar
      Add TAPEX (#16473) · 4ef0abb7
      NielsRogge authored 
      

* Add TapexTokenizer

* Improve docstrings and provide option to provide answer

* Remove option for pretokenized inputs

* Add TAPEX to README

* Fix copies

* Remove option for pretokenized inputs

* Initial commit: add tapex fine-tuning examples on both table-based question answering and table-based fact verification.

* - Draft a README file for running the script and introducing some background.
- Remove unused code lines in tabfact script.
- Disable the deafult `pad_to_max_length` option which is memory-consuming.

* * Support `as_target_tokenizer` function for TapexTokenizer.
* Fix the do_lower_case behaviour of TapexTokenizer.
* Add unit tests for target scenarios and cased/uncased scenarios for both source and target.

* * Replace the label BartTokenizer with TapexTokenizer's as_target_tokenizer function.
* Fix typos in tapex example README.

* * fix the evaluation script - remove the property `task_name`

* * Make the label space more clear for tabfact tasks

* * Using a new fine-tuning script for tapex-base on tabfact.

* * Remove the lowercase code outside the tokenizer - we use the tokenizer to control whether do_lower_case
* Guarantee the hyper-parameter can be run without out-of-memory on 16GB card and report the new reproduced number on wikisql

* * Remove the default tokenizer_name option.
* Provide evaluation command.

* * Support for WikiTableQuestion dataset.

* Fix a typo in README.

* * Fix the datasets's key name in WikiTableQuestions

* Run make fixup and move test to folder

* Fix quality

* Apply suggestions from code review

* Apply suggestions from code review
Co-authored-by: default avatarSuraj Patil <surajp815@gmail.com>

* Apply suggestions from code review

* Apply suggestions from code review
Co-authored-by: default avatarSylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Apply some more suggestions from code review

* Improve docstrings

* Overwrite failing test

* Improve comment in example scripts

* Fix rebase

* Add TAPEX to Auto mapping

* Add TAPEX to auto config mappings

* Put TAPEX higher than BART in auto mapping

* Add TAPEX to doc tests
Co-authored-by: default avatarNiels Rogge <nielsrogge@Nielss-MBP.localdomain>
Co-authored-by: default avatarSivilTaram <qianlxc@outlook.com>
Co-authored-by: default avatarNiels Rogge <nielsrogge@nielss-mbp.home>
Co-authored-by: default avatarSuraj Patil <surajp815@gmail.com>
Co-authored-by: default avatarSylvain Gugger <35901082+sgugger@users.noreply.github.com>
Co-authored-by: default avatarNiels Rogge <nielsrogge@Nielss-MacBook-Pro.local>
      4ef0abb7
  9. 07 Apr, 2022 2 commits
  11. 06 Apr, 2022 2 commits
  13. 05 Apr, 2022 2 commits
    • Matt's avatar
      Adding new train_step logic to make things less confusing for users (#15994) · 43540052
      Matt authored 
      

* Adding new train_step logic to make things less confusing for users

* DO NOT ASK WHY WE NEED THAT SUBCLASS

* Metrics now working, at least for single-output models with type annotations!

* Updates and TODOs for the new train_step

* Make fixup

* Temporary test workaround until T5 has types

* Temporary test workaround until T5 has types

* I think this actually works! Needs a lot of tests though

* MAke style/quality

* Revert changes to T5 tests

* Deleting the aforementioned unmentionable subclass

* Deleting the aforementioned unmentionable subclass

* Adding a Keras API test

* Style fixes

* Removing unneeded TODO and comments

* Update test_step too

* Stop trying to compute metrics with the dummy_loss, patch up test

* Make style

* make fixup

* Docstring cleanup

* make fixup

* make fixup

* Stop expanding 1D input tensors when using dummy loss

* Adjust T5 test given the new compile()

* make fixup

* Skipping test for convnext

* Removing old T5-specific Keras test now that we have a common one

* make fixup

* make fixup

* Only skip convnext test on CPU

* Update src/transformers/modeling_tf_utils.py
Co-authored-by: default avatarSylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/modeling_tf_utils.py
Co-authored-by: default avatarSylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Avoiding TF import issues

* make fixup

* Update compile() to support TF 2.3

* Skipping model.fit() on template classes for now

* Skipping model.fit() on template class tests for now

* Replace ad-hoc solution with find_labels

* make fixup
Co-authored-by: default avatarSylvain Gugger <35901082+sgugger@users.noreply.github.com>
      43540052
    • Yih-Dar's avatar
      Fix CI: test_inference_for_pretraining in ViTMAEModelTest (#16591) · 765bafb8
      Yih-Dar authored 
      
Co-authored-by: default avatarydshieh <ydshieh@users.noreply.github.com>
      765bafb8
  15. 04 Apr, 2022 3 commits
  17. 01 Apr, 2022 2 commits
  19. 30 Mar, 2022 1 commit
    • Francesco Saverio Zuppichini's avatar
      Feature Extractor accepts `segmentation_maps` (#15964) · c4deb7b3
      Francesco Saverio Zuppichini authored 
      

* feature extractor accepts

* resolved conversations

* added examples in test for ADE20K

* num_classes -> num_labels

* Apply suggestions from code review
Co-authored-by: default avatarSylvain Gugger <35901082+sgugger@users.noreply.github.com>

* resolving conversations

* resolving conversations

* removed ADE

* CI

* minor changes in conversion script

* reduce_labels in feature extractor

* minor changes

* correct preprocess for instace segmentation maps

* minor changes

* minor changes

* CI

* debugging

* better padding

* going to update labels inside the model

* going to update labels inside the model

* minor changes

* tests

* removed changes in feature_extractor_utils

* conversation

* conversation

* example in feature extractor

* more docstring in modeling

* test

* make style

* doc
Co-authored-by: default avatarSylvain Gugger <35901082+sgugger@users.noreply.github.com>
      c4deb7b3
  21. 29 Mar, 2022 6 commits
  23. 28 Mar, 2022 4 commits
    • NielsRogge's avatar
      Add DPT (#15991) · 979b039c
      NielsRogge authored 
      

* First draft

* More improvements

* Add fusion blocks

* Make conversion script work for dpt_large

* Make conversion script work

* Improve implementation

* Improve conversion script

* Add DPTForSemanticSegmentation

* Make conversion work for semantic segmentation

* Add tests

* Remove print statements

* First draft

* Redesign neck

* Improve tests

* Improve implementation some more

* Make neck output list of tensors

* Improve neck and feature extractor

* Fix integration tests

* Make more tests pass

* Make all tests pass

* Add missing config archive map

* Add in_index attribute to make heads accept list of tensors

* Apply suggestions from code review

* Apply suggestions from code review
Co-authored-by: default avatarSylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Apply some more suggestions

* Add copied from statements

* Remove assert

* Apply suggestions from code review

* Apply suggestions from code review

* Remove DPTInterpolate in favor of nn.Upsample

* Add comments

* Apply suggestions from code review

* Apply suggestions from code review

* Add proposed design

* Update design

* Add DPTReassembleLayer

* Add DPTFeatureFusionStage

* Apply more suggestions from code review

* Apply suggestions from code review

* Apply suggestions from code review

* Fix rebase

* Update in_index and out_indices

* Fix conversion script

* Fix code quality

* Add model to toctree and use DepthEstimatorOutput

* Fix rebase

* Fix code examples

* Improve code

* Fix copied from statements

* Apply suggestions from code review

* Remove compute_loss method

* Apply suggestions from code review

* Fix documentation tests file

* Remove test.py file

* Improve doc example
Co-authored-by: default avatarNiels Rogge <nielsrogge@Nielss-MacBook-Pro.local>
Co-authored-by: default avatarSylvain Gugger <35901082+sgugger@users.noreply.github.com>
Co-authored-by: default avatarNiels Rogge <nielsrogge@nielss-mbp.home>
      979b039c
    • Sanchit Gandhi's avatar
      [FlaxSpeechEncoderDecoderModel] Ensure Input and Output Word Embeddings Are **Not** Tied (#16444) · 7ca46335
      Sanchit Gandhi authored 
      * [FlaxSpeechEncoderDecoderModel] Ensure Input and Output Word Embeddings Are **Not** Tied

* rebase
      7ca46335
    • Jaesun Park's avatar
      Fix PerceiverMLP and test (#16405) · e0ac72b7
      Jaesun Park authored 
      
Co-authored-by: default avatarJaesun Park <jaesun.park1@navercorp.com>
      e0ac72b7
    • Sanchit Gandhi's avatar
      [Flax] Improve Robustness of Back-Prop Tests (#16418) · 925fc57b
      Sanchit Gandhi authored 
      * [Flax] Improve Robustness of Back-Prop Tests

* check equality of logits/outputs

* make fixup
      925fc57b
  25. 25 Mar, 2022 4 commits
  27. 24 Mar, 2022 2 commits