Add test suite for batch-invariant operations#16
Open
Jah-yee wants to merge 3 commits into
Open
Conversation
Handle edge case when reduction dimension size is 0 to avoid division by zero leading to NaN values. This fix adds explicit checks in both mean_kernel (Triton kernel) and mean_dim (Python wrapper) to properly handle empty tensors.
…ibility The flatten parameter was removed in Triton 3.x public release. This changes tl.range(..., flatten=True) to Python's built-in range(), which works inside @triton.jit functions and is compatible with all versions. Fixes: TypeError: range.__init__() got an unexpected keyword argument 'flatten'
- Add pytest tests for mm, addmm, log_softmax, mean operations - Test batch-invariant property: op(x[:1], y) == op(x, y)[:1] - Include tests for multiple dtypes (float32, float16, bfloat16) - Add tests for batch invariant mode context manager
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
This PR adds a comprehensive test suite to verify the batch-invariant operations work correctly.
Tests Added
The test suite covers mm, addmm, log_softmax, and mean operations.
All tests verify the batch-invariant property: op(x[:1], y) == op(x, y)[:1]