Add TAPEX #16473
Merged
Add TAPEX #16473
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
The documentation is not available anymore as the PR was closed or merged. |
patil-suraj
approved these changes
Mar 29, 2022
examples/research_projects/tapex/run_wikitablequestions_with_tapex.py
Outdated
Show resolved
Hide resolved
examples/research_projects/tapex/run_wikitablequestions_with_tapex.py
Outdated
Show resolved
Hide resolved
sgugger
reviewed
Mar 29, 2022
There was a problem hiding this comment.
Thanks for the PR!
If the expected architecture should be BART by default, then this model should be added in the relevant auto mapping to work with AutoConfig and AutoModelForXxx. This is jsut a defult value that can be changed in the config if there are checkpoints that rely on a different architecture.
|
However, these functions are used in the script, so I can't remove these imports. |
|
@NielsRogge Thanks for your huge effort! I personally think these two warnings are correct since these two imports are only used in
|
|
@sgugger I've addressed all comments. |
sgugger
approved these changes
Apr 4, 2022
|
All models on the hub do work with the Auto API, can you elaborate? TAPEX is also added to |
NielsRogge
force-pushed
the
add_tapex_bis
branch
2 times, most recently
from
b078744
to
5f86fa5
Compare
…estion answering and table-based fact verification.
…kground. - Remove unused code lines in tabfact script. - Disable the deafult `pad_to_max_length` option which is memory-consuming.
* Fix the do_lower_case behaviour of TapexTokenizer. * Add unit tests for target scenarios and cased/uncased scenarios for both source and target.
…enizer function. * Fix typos in tapex example README.
…zer 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
* Provide evaluation command.
Co-authored-by: Suraj Patil <surajp815@gmail.com>
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
NielsRogge
force-pushed
the
add_tapex_bis
branch
from
6a2199a
to
8d19141
Compare
elusenji
pushed a commit
to elusenji/transformers
that referenced
this pull request
Apr 8, 2022
* 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: Suraj Patil <surajp815@gmail.com> * Apply suggestions from code review * Apply suggestions from code review Co-authored-by: Sylvain 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: Niels Rogge <nielsrogge@Nielss-MBP.localdomain> Co-authored-by: SivilTaram <qianlxc@outlook.com> Co-authored-by: Niels Rogge <nielsrogge@nielss-mbp.home> Co-authored-by: Suraj Patil <surajp815@gmail.com> Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com> Co-authored-by: Niels Rogge <nielsrogge@Nielss-MacBook-Pro.local>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
What does this PR do?
Remember TAPAS, the table QA model by Google AI? Microsoft has now released TAPEX, a seq2seq model that outperforms TAPAS and is actually much simpler: table QA is just treated as a seq2seq problem.
As the weights can be directly loaded into a BART model, this PR only implements
TapexTokenizer, which can be used to prepare tables and corresponding texts for the model.This PR also adds 3 scripts that showcase how to fine-tune TAPEX on 3 important benchmarks: WikiSQL and WTQ for table question answering and TabFact for table fact verification.
Kudos to @SivilTaram (the original author) for improving my initial
TapexTokenizerimplementation, as well as adding the 3 fine-tuning scripts.