-
Notifications
You must be signed in to change notification settings - Fork 41
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
Forward and backward weights padding kernel #264
Conversation
2e22343
to
27745b2
Compare
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.
Finished the first round of quick review.
Will need to do the 2nd round putting more scrutiny on the transforms applied.
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.
Did the 2nd iteration of review. I get the big picture of how you pad the dimensions and set OOB check dimensions. Some comments need to be improved so future developers can understand your logic.
For FIXME, let's discuss in the meeting later.
jenkins: retest this please. |
can't connect to ci system |
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.
some additional comments need to be revised. getting closer.
also CI results look fine so far.
CI system is working fine. You should obtain login information from @okakarpa . |
6d62db4
to
a7eac5d
Compare
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.
Some minor changes to the comments and the PR is good to go
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.
LGTM. Thanks for getting this monumental work done!
this PR enable padding kernel for forward and backward weights
we can support resnet50 now