Movatterモバイル変換


[0]ホーム

URL:


Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

Sign up
Appearance settings

[jit] Cache node operators to speed up optimization#24827

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 ourterms of service andprivacy statement. We’ll occasionally send you account related emails.

Already on GitHub?Sign in to your account

Closed
suo wants to merge1 commit intogh/suo/146/basefromgh/suo/146/head

Conversation

@suo
Copy link
Member

@suosuo commentedAug 19, 2019
edited
Loading

Stack fromghstack:

We already cache the node's schema, but alias analysis wants operators.
This ends up being almost 70% of the on-cpu time optimizing a large
graph.

Here's some results on asample model
(the units are seconds).

Before:

compiled in:  20.256319999694824first run in:  313.77824568748474

After:

compiled in:  18.8815860748291first run in:  42.58739233016968

More than a 7x speedup! Still slower than I'd like though so I'll keep
digging.

Differential Revision:D16887540

We already cache the node's schema, but alias analysis wants operators.This ends up being almost 70% of the on-cpu time optimizing a largegraph.Here's some results on a [samplemodel](https://gist.github.com/suo/63ab9638516002176f94553a37060f61)(the units are seconds).Before:```compiled in:  20.256319999694824first run in:  313.77824568748474```After:```compiled in:  18.8815860748291first run in:  42.58739233016968```More than a 7x speedup! Still slower than I'd like though so I'll keepdigging.
@pytorchbotpytorchbot added the oncall: jitAdd this issue/PR to JIT oncall triage queue labelAug 19, 2019
suo added a commit that referenced this pull requestAug 19, 2019
We already cache the node's schema, but alias analysis wants operators.This ends up being almost 70% of the on-cpu time optimizing a largegraph.Here's some results on a [samplemodel](https://gist.github.com/suo/63ab9638516002176f94553a37060f61)(the units are seconds).Before:```compiled in:  20.256319999694824first run in:  313.77824568748474```After:```compiled in:  18.8815860748291first run in:  42.58739233016968```More than a 7x speedup! Still slower than I'd like though so I'll keepdigging.ghstack-source-id:4d524dcPull Requestresolved:#24827
@suosuo requested a review fromzdevitoAugust 19, 2019 01:03
@suo
Copy link
MemberAuthor

suo commentedAug 19, 2019

cc@resistor for ongoing compiler optimization saga

@zou3519zou3519 deleted the gh/suo/146/head branchAugust 19, 2019 17:30
@facebook-github-bot
Copy link
Contributor

@suo merged this pull request inc6617b3.

Sign up for freeto join this conversation on GitHub. Already have an account?Sign in to comment

Reviewers

@zdevitozdevitozdevito approved these changes

Assignees

No one assigned

Labels

Mergedoncall: jitAdd this issue/PR to JIT oncall triage queue

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

5 participants

@suo@facebook-github-bot@zdevito@pytorchbot@mruberry

[8]ページ先頭

©2009-2025 Movatter.jp