Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

added group norm #225

Closed
wants to merge 1 commit into from
Closed

added group norm #225

wants to merge 1 commit into from

Conversation

denti
Copy link

@denti denti commented Sep 6, 2024

Summary

Testing Done

  • Hardware Type:
  • run make test to ensure correctness
  • run make checkstyle to ensure code style
  • run make test-convergence to ensure convergence

@ByronHsu ByronHsu self-requested a review September 6, 2024 23:11
@ByronHsu ByronHsu mentioned this pull request Sep 30, 2024
x = tl.load(X + r, mask = r < GROUP_SIZE).to(tl.float32)
m2_ = tl.zeros((BLOCK_SIZE,), dtype=tl.float32)
weight_ = (r < GROUP_SIZE).to(tl.float32)
_mean, _m2, _weight = welford_combine(_mean, _m2, _weight, x, m2_, weight_)

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

not a huge deal but I noticed m2_ is just an array of zeros every single time

i made a second kernel of the welford_combine function removing m2 as an argument and it works the same

@ByronHsu
Copy link
Collaborator

ByronHsu commented Nov 8, 2024

impl by #353

@ByronHsu ByronHsu closed this Nov 8, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants