Dynamic scaling triton kernel #28
Merged
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
drisspg
force-pushed
the
dynamic-scaling
branch
from
bdd42df to
8b62d1a
Compare
…████████████████████████████████| 8/8 [00:20<00:00, 2.57s/it] numel high_precision_dtype low_precision_dtype triton_time pytorch_time compiled_pytorch_time -------- ---------------------- --------------------- ------------- -------------- ----------------------- 2097152 torch.float32 torch.float8_e4m3fn 65.2814 60.3875 86.7003 2097152 torch.float32 torch.float8_e5m2 65.1289 60.8456 85.949 4194304 torch.float32 torch.float8_e4m3fn 64.6113 80.8327 92.3341 4194304 torch.float32 torch.float8_e5m2 66.3706 81.2847 92.8501 8388608 torch.float32 torch.float8_e4m3fn 66.6676 139.581 45.354 8388608 torch.float32 torch.float8_e5m2 64.9911 139.302 41.2224 16777216 torch.float32 torch.float8_e4m3fn 64.3637 298.605 87.7525 16777216 torch.float32 torch.float8_e5m2 64.8017 298.626 80.7421
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Outdated suggestions cannot be applied.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
Add dynamic scaling kernel in 1 go with spinlock, the numerics feel not quite right though