-
Notifications
You must be signed in to change notification settings - Fork 4.5k
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
10X faster predict by replacing DataFrame with dict #2299
Conversation
@tcuongd can we please merge this and release a new version with wheels for python 3.11 as well? |
@tcuongd any updates on this? It should be really easy to integrate and yields a big runtime improvement |
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.
Wow great find, thanks heaps! Yeah I think changing the return type for these internal functions is fine, we just need to make it clear which functions are affected in the release notes.
For some reason CI didn't kick off, I'll take a look later |
Hi @orenmatar! Thank you for your pull request and welcome to our community. Action RequiredIn order to merge any pull request (code, docs, etc.), we require contributors to sign our Contributor License Agreement, and we don't seem to have one on file for you. ProcessIn order for us to review and merge your suggested changes, please sign at https://code.facebook.com/cla. If you are contributing on behalf of someone else (eg your employer), the individual CLA may not be sufficient and your employer may need to sign the corporate CLA. Once the CLA is signed, our tooling will perform checks and validations. Afterwards, the pull request will be tagged with If you have received this in error or have any questions, please contact us at cla@meta.com. Thanks! |
@orenmatar Do you mind just signing the CLA? Thanks! |
@tcuongd Signed it just now |
Thank you for signing our Contributor License Agreement. We can now accept your code for this (and any) Meta Open Source project. Thanks! |
Profiling the predict code yielded this result:
Evidently about 90% of the predict time is spent on either the init of a DataFrame or its getitem method, both are notoriously slow.
A very simple change from dataframe to dict in sample_model and sample_model_vectorized has no risk and implications beyond sample_posterior_predictive since the df is converted to a dict there anyway.
Profiling after the change gives:
and in runtime it was reduced from ~300ml to ~38ml on a local test