-
Notifications
You must be signed in to change notification settings - Fork 3.5k
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
[CI] Cleanup stale logs for auto-tuning #8160
Conversation
should fix the recent ci flaky issues |
cp -f tutorials/auto_scheduler/ci_logs/*.json tutorials/auto_scheduler | ||
cp -f tutorials/auto_scheduler/ci_logs/*.json tutorials/get_started |
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.
why this line?
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.
the particular flaky error was tune matmul in get started, the tutorial get moved over there, so we need some log to help prevent flaky issue
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.
in tune_matmul, it produces its own log file, no? that's why the target with new default option caused troubles in CI.
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.
it is for the case where you get too few rounds but no valid record i believe
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.
Ok @tqchen, I don't want to hold up a CI fix, but would be great to document this in a follow on either in the test script or tutorial.
cc @merrymercy @areusch @junrushao1994