Make Multi Node sampler cycle forever#1424
Merged
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/data/1424
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (13 Unrelated Failures)As of commit 25062f2 with merge base 4ec4548 ( FLAKY - The following job failed but was likely due to flakiness present on trunk:
BROKEN TRUNK - The following jobs failed but was present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
andrewkho
reviewed
Jan 21, 2025
| 3) FIRST_DATASET_EXHAUSTED: Stop when the first dataset is exhausted. | ||
|
|
||
| 4) CYCLE_FOREVER: Cycle through the datasets by reinitializing each exhausted source nodes. | ||
| This is useful when trainer want control over certain number of steps instead of epochs. |
Contributor
There was a problem hiding this comment.
Will this call set_epoch on the underlying datasets?
Contributor
There was a problem hiding this comment.
Wondering if we need to do anything special, or can advise users on how to perform that. This otherwise looks good to me
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This PR adds support for continuous cycling through each base node in multi node weighted sampler. With the current setup this was each to extend. We do cycle through each node to support
StopCriteria.CYCLE_UNTIL_ALL_DATASETS_EXHAUSTED, forStopCriteria.CYCLE_FOREVERwe skip raising aStopIteration.This should be functionally equivalent to having an infinite sampler plugged on top of each base dataset.
The dataset order (as defined by
weights) remains the same.