Allow for dynamic batch padding #2352
Merged
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
muellerzr
commented
Jan 18, 2024
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
SunMarc
reviewed
Jan 18, 2024
src/accelerate/inference.py
Outdated
| process_index=0, | ||
| num_processes=state.num_processes, | ||
| ) | ||
| extra = concatenate([extra] * ((found_batch_size % state.num_processes) + 1)) |
Co-authored-by: Marc Sun <57196510+SunMarc@users.noreply.github.com>
SunMarc
reviewed
Jan 18, 2024
SunMarc
reviewed
Jan 19, 2024
SunMarc
reviewed
Jan 24, 2024
muellerzr
commented
Jan 25, 2024
| old_size = tensor.shape | ||
| new_size = list(old_size) | ||
| new_size[0] = batch_size + to_pad | ||
| new_tensor = tensor.new_zeros(tuple(new_size)) |
There was a problem hiding this comment.
It's okay to just pad 0, we can drop these afterwards and the user won't know that padded inputs were event sent ideally
muellerzr
added a commit
that referenced
this pull request
Feb 6, 2024
* Broken version * Timing I would expect * Working version! * Use MethodType * working test * Tests * Use no split module classes explicitly * Put split_points in pipelien * Store split points in hf_split_points * fix case num_process=1 * Allow for dynamic batch padding (#2352) * Allow for dynamic batch paddign * Fix test * Update src/accelerate/inference.py Co-authored-by: Marc Sun <57196510+SunMarc@users.noreply.github.com> * Break early after the first valid bs is found * Less slicy-dicy * Test cv model * Start, need to test * Use dataloader-like logic * Refactor to utils * With tests * Update the source * Clean * bs=1 case * Add test * add some failing test * Almost working version * Much cleaner implementation * Use pad_input_tensor * All tests passing! * Do it at tracing too --------- Co-authored-by: Marc Sun <57196510+SunMarc@users.noreply.github.com> Co-authored-by: Marc Sun <marc@huggingface.co> * Rm literal * Allow users to pass in max_memory * Note about recursion * Document, document, document * Right import check * Fix bug, add tests to multigpu runners * Change default to None * Start of docs * Try again? * Try again x2 * Trailing comma * Move import * Clean * typehint * typo * From code review * Use num_chunks * Update tests/test_utils.py Co-authored-by: Marc Sun <57196510+SunMarc@users.noreply.github.com> * Bad copy/paste * hf_split_points --------- Co-authored-by: Marc Sun <marc@huggingface.co> Co-authored-by: Marc Sun <57196510+SunMarc@users.noreply.github.com>
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?
This PR allows for dynamic batch padding by finding and duplicating the last item in the batch before concating everything.
Current issue:
PiPPyhas issues with batch inference on GPT-2, will talk to the pippy folks about thisFixes # (issue)
Before submitting
Pull Request section?
to it if that's the case.
documentation guidelines, and
here are tips on formatting docstrings.
Who can review?
Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.
@SunMarc