Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Drop incomplete batches for Ray and Pandas to prevent Batchnorm computation errors #2778

Merged
merged 8 commits into from
Dec 13, 2022

Conversation

arnavgarg1
Copy link
Contributor

@arnavgarg1 arnavgarg1 commented Nov 22, 2022

This PR implements logic to drop the last batch for Pandas and Ray dataset batchers to prevent issues when there is only 1 row in a batch.

A single row in a batch causes issues when computing Batchnorm (such as in FC layers, Tabnet, etc.) because it can't compute the mean/stddev for a single sample. Additionally, Ludwig has some logic for Tabnet that has a conditional check for sample_size == 1 that we can now safely remove.

Skipping adding tests since our current test suite already has tests that use the batcher classes for model training, so if the tests pass, then this logic should work as well. I modified individual tests locally to be sure.

Note: This only drops incomplete training batches if the batch size < total number of rows in the dataset. Additionally, it will always keep incomplete training batches for validation/test sets.

To follow:

  • Remove check for batch_size == 1 in Tabnet

@arnavgarg1 arnavgarg1 changed the title Add logic to correctly drop last batch for Ray and Pandas Drop incomplete batches for Ray and Pandas to prevent Batchnorm computation errors Nov 22, 2022
@arnavgarg1 arnavgarg1 requested review from justinxzhao and ShreyaR and removed request for justinxzhao November 22, 2022 16:28
@github-actions
Copy link

github-actions bot commented Nov 22, 2022

Unit Test Results

         6 files  ±  0           6 suites  ±0   3h 31m 23s ⏱️ - 14m 16s
  3 527 tests  -   5    3 456 ✔️  - 1    71 💤  -   4  0 ±0 
10 581 runs   - 15  10 353 ✔️ ±0  228 💤  - 15  0 ±0 

Results for commit bd759ff. ± Comparison against base commit 8f5d013.

♻️ This comment has been updated with latest results.

@arnavgarg1 arnavgarg1 requested a review from ShreyaR December 12, 2022 17:55
Copy link
Contributor

@ShreyaR ShreyaR left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

lgtm!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants