triton softmax support multi-batch #152
Open
Conversation
This file contains 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
Gy-Lu
commented
Feb 16, 2023
tests/test_fastnn/test_softmax.py
Outdated
| @@ -15,19 +15,19 @@ def _test_softmax_core(): | |||
|
|
|||
| for seq_ in test_seq_: | |||
| for dtype in test_dtype: | |||
| sample_input = torch.rand(batch_, chunk_, head_, seq_, | |||
| sample_input = torch.rand(batch, batch_, chunk_, head_, seq_, | |||
There was a problem hiding this comment.
Oh, without the batch dimension it would work correctly as well
Shenggan
approved these changes
Feb 18, 2023
|
I found it would fail when the batch dimension comes to 2. |
|
Update: |
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Outdated suggestions cannot be applied.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.
Support another batch dimension for softmax. In training or batch inference, we may add a batch dimension as the first dimension of some tensors. However, we use the third dimension(
tensor.shape[2]) as thehead_dim, which would be influenced. In this pr, I modify it totensor.shape[-3]to solve this problem. CUDA kernel is modified as well.Enable test_atten_core, this test is skipped by default and never be used.