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

[Relay][OpStrategy] Tweak cublas/cudnn priority level #5820

Merged
merged 3 commits into from
Jun 16, 2020

Conversation

icemelon
Copy link
Member

@icemelon icemelon commented Jun 15, 2020

Increase the dense cublas and conv cudnn priority level such that it'll be used by default when cublas is included in the target.

Copy link
Contributor

@comaniac comaniac left a comment

Choose a reason for hiding this comment

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

LGTM. Should we also change conv2d and conv3d?

@icemelon
Copy link
Member Author

@comaniac Good catch. updated.

@icemelon icemelon changed the title [Relay][OpStrategy] Tweak cublas priority level [Relay][OpStrategy] Tweak cublas/cudnn priority level Jun 15, 2020
@zhiics zhiics merged commit 85d2cb1 into apache:master Jun 16, 2020
@icemelon icemelon deleted the cuda-strategy branch June 16, 2020 20:14
trevor-m pushed a commit to trevor-m/tvm that referenced this pull request Jun 30, 2020
* Tweak cublas plevel

* update

* trigger ci
zhiics pushed a commit to neo-ai/tvm that referenced this pull request Jul 2, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

3 participants