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

[rust] Use fused layer #3260

Merged
merged 1 commit into from
Jun 25, 2024
Merged

[rust] Use fused layer #3260

merged 1 commit into from
Jun 25, 2024

Conversation

xyang16
Copy link
Contributor

@xyang16 xyang16 commented Jun 17, 2024

Description

Brief description of what this PR is about

  • If this change is a backward incompatible change, why must this change be made?
  • Interesting edge cases to note here

@xyang16 xyang16 requested review from zachgk, frankfliu and a team as code owners June 17, 2024 17:36
@xyang16 xyang16 force-pushed the rust branch 2 times, most recently from 19c0733 to 7a4ca2b Compare June 17, 2024 17:48
Copy link
Contributor

@frankfliu frankfliu left a comment

Choose a reason for hiding this comment

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

We can add NDArrayEx.layerNorm() support in next PR.

@xyang16 xyang16 force-pushed the rust branch 3 times, most recently from 9f0b0db to c5abbce Compare June 24, 2024 22:35
@xyang16 xyang16 changed the title [rust] Use fused layer norm [rust] Use fused linear Jun 24, 2024
@xyang16 xyang16 changed the title [rust] Use fused linear [rust] Use fused layer Jun 24, 2024
candle-nn = "0.5.1"
candle-transformers = "0.5.1"
candle-flash-attn = { version = "0.5.1", optional = true }
cudarc = { version = "0.11.6", default-features = false, features = [ "cublaslt", "f16" ], optional = true}
Copy link
Contributor

Choose a reason for hiding this comment

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

We need a way to compile w/o cuda

Copy link
Contributor Author

Choose a reason for hiding this comment

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

Done.

@xyang16 xyang16 merged commit 82429f2 into deepjavalibrary:master Jun 25, 2024
5 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants