-
Notifications
You must be signed in to change notification settings - Fork 62
[Nightly] Add op performance regression check #1622
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
Conversation
cafd374 to
b88451a
Compare
79d5b40 to
8a15500
Compare
|
Save baseline artifact number into a github issue: #1689. |
b6a188b to
e92e12b
Compare
|
@chuanqi129 Hello, could you please help review this op regression check PR? Thanks~ |
chuanqi129
left a comment
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.
Overall, LGTM. Is the on-demand test demo https://github.com/intel/torch-xpu-ops/actions/runs/15337311965 still available? Do we need to rerun a new demo?
Base on the report of current demo, give below comments:
- The regression and improvement table sort with ratio -> case name -> op name -> dt, if possible
- I saw there are some 99% drop cases in the regression table, are those real issue? Similar for improvement table, there are some >>100% improvement.
- Current table has too many column, can we concat some op config in one column or other better format
- I saw the tables are duplicate in the summary, please double check it


Uh oh!
There was an error while loading. Please reload this page.