[JAX SC] Refine ID dropping logic in SparseCore input preprocessing. #522
Conversation
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
copybara-service
bot
force-pushed
the
test_822368695
branch
3 times, most recently
from
837407c to
38ffa0c
Compare
The logic for counting and dropping IDs based on `max_ids_per_partition` and `max_unique_ids_per_partition` during the sorting and grouping of COO tensors has been updated. The counters for total and unique IDs per partition are now incremented only when a new, non-duplicate ID is added, and checks for exceeding limits are performed *before* incrementing. This ensures more accurate enforcement of the capacity constraints. Test validation is updated to check these limits. PiperOrigin-RevId: 822771909
copybara-service
bot
force-pushed
the
test_822368695
branch
from
38ffa0c to
dc889b6
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.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.
[JAX SC] Refine ID dropping logic in SparseCore input preprocessing.
The logic for counting and dropping IDs based on
max_ids_per_partitionandmax_unique_ids_per_partitionduring the sorting and grouping of COO tensors has been updated. The counters for total and unique IDs per partition are now incremented only when a new, non-duplicate ID is added, and checks for exceeding limits are performed before incrementing. This ensures more accurate enforcement of the capacity constraints. Test validation is updated to check these limits.