This repository was archived by the owner on Jan 5, 2023. It is now read-only.
Unit tests for shuffler#22
Open
krfricke wants to merge 3 commits into
Open
Conversation
clarkzinzow
approved these changes
Jul 15, 2021
|
|
||
|
|
||
| class DataLoaderShuffleTest(unittest.TestCase): | ||
| """This test suite validates core RayDMatrix functionality.""" |
Collaborator
There was a problem hiding this comment.
Suggested change
| """This test suite validates core RayDMatrix functionality.""" | |
| """This test suite validates core shuffle functionality.""" |
| def tearDownClass(cls): | ||
| ray.shutdown() | ||
|
|
||
| def testShuffleMap(self): |
Collaborator
There was a problem hiding this comment.
Nit: Tests should be snake case.
Suggested change
| def testShuffleMap(self): | |
| def test_shuffle_map(self): |
| assert len(set(all_keys)) == len(all_keys), \ | ||
| "Keys in full dataset are not distinct." | ||
|
|
||
| def testShuffleReduce(self): |
Collaborator
There was a problem hiding this comment.
Nit: Tests should be snake case.
Suggested change
| def testShuffleReduce(self): | |
| def test_shuffle_reduce(self): |
| assert set(unshuffled) == set(shuffled), \ | ||
| "Key mismatch between unshuffled and shuffled parts" | ||
|
|
||
| def testShuffleEndToEnd(self): |
Collaborator
There was a problem hiding this comment.
Nit: Tests should be snake case.
Suggested change
| def testShuffleEndToEnd(self): | |
| def test_shuffle_end_to_end(self): |
Comment on lines
+59
to
+61
| # 3sd = 99.7% chance of passing | ||
| assert mean - 3 * sd < len(part_keys) < mean + 3 * sd, \ | ||
| f"Not enough rows in partition {i}" |
Collaborator
There was a problem hiding this comment.
Nice! How should we interpret the outliers when this assertion eventually fails?
| assert len(all_keys) == self.num_rows, "Not all rows were returned." | ||
|
|
||
| assert len(set(all_keys)) == len(all_keys), \ | ||
| "Keys in full dataset are not distinct." |
Collaborator
There was a problem hiding this comment.
This can wait, but we may want to confirm that none of the actual data was unintentionally mutated, e.g. due to type coercion. That would probably require a slight refactor of (or utility added to) the data generation code.
|
|
||
| shuffled = ray.get( | ||
| shuffle_reduce.remote( | ||
| 0, |
Collaborator
There was a problem hiding this comment.
It shouldn't matter much (I actually don't think it's even used in the reducer anymore), but maybe we should set the reducer_index here.
Suggested change
| 0, | |
| i, |
| for tid, epoch_batches in consumer.rank_epoch_batches.items(): | ||
| for i in range(len(epoch_batches) - 1): | ||
| assert len(epoch_batches[i]) == len( | ||
| epoch_batches[+1]) == num_epochs, \ |
Collaborator
There was a problem hiding this comment.
Suggested change
| epoch_batches[+1]) == num_epochs, \ | |
| epoch_batches[1]) == num_epochs, \ |
| "Keys in dataset are not distinct." | ||
|
|
||
| assert set1 == set2, \ | ||
| "Shuffled key sets are not equal." |
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 subscribe to this conversation on GitHub.
Already have an account?
Sign in.
2 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.
No description provided.