-
Notifications
You must be signed in to change notification settings - Fork 476
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
Check precision only if on CUDA and not inductor. #6407
Conversation
Co-authored-by: Emilio Cota <[email protected]>
Can you paste in before, and after kernels just as before? Just hf_Bert will suffice. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks!
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks!
hf_Bert (before)
|
hf_Bert (after)
|
@golechwierowicz Let me know what you think about the profiling numbers. |
Looks good! |
Co-authored-by: Emilio Cota <[email protected]>
Co-authored-by: Emilio Cota <[email protected]>
This PR skips precision flag checking for experiments whose: (i) device is not
cuda
; and (ii) dynamo backends are notinductor
.(i) the flag checked is aimed towards CUDA only:
DEFAULT_CUDA_<test>_PRECISION
(ii) PyTorch/benchmark scripts already take care of the model conversion for inductor
cc @miladm