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

[TOPI, Relay refactor] Move Dilation2d from nn to image namespace #5110

Merged
merged 4 commits into from
Mar 20, 2020

Conversation

masahi
Copy link
Member

@masahi masahi commented Mar 20, 2020

Dilation2d added in #5033 is a morphological operation, not a standard neural network operator. So putting this inside of image name space seems be to a better choice to me.

I noticed this when looking at topi/nn directory. There are now dilate.py and dilation2d.py, too similar functions but the former is used for conv transpose (dilate with 0) while the latter is for morphological dilation (dilate with max in the window). I thought it was confusing, so moved dilation2d.py to topi/image. Applied the same fix to Relay as well.

please review @kevinthesun @jwfromm @maheshambule I'm sorry I missed this in the original PR.

@maheshambule
Copy link
Contributor

@masahi, Thanks for the refactoring. The dilate.py and dilation2d.py in the same folder was confusing. I should have kept it in image namespace.

Copy link
Contributor

@jwfromm jwfromm left a comment

Choose a reason for hiding this comment

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

Makes sense to me, this refactor looks good.

@masahi
Copy link
Member Author

masahi commented Mar 20, 2020

@kevinthesun ready to merge, I just moved things around.

@tqchen tqchen merged commit 2b66123 into apache:master Mar 20, 2020
@tqchen
Copy link
Member

tqchen commented Mar 20, 2020

THanks @masahi @maheshambule @jwfromm !

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants