facebookresearch / theseus

A library for differentiable nonlinear optimization
MIT License
1.78k stars 128 forks source link

Speedup VariableOrdering.append from O(N^2) to O(N) #657

Closed ahojnnes closed 3 months ago

ahojnnes commented 3 months ago

Motivation and Context

When specifying the ordering for larger problems with many thousands of variables, the current implementation becomes very slow due to the O(N^2) complexity caused by the O(N) complexity of checking existence of var in the _var_order list. For small problems, the new implementation is a little slower but the difference is negligible. For large problems, the speed goes down from multiple 10s of seconds to

How Has This Been Tested

The runtime has been tested on various bundle adjustment problems. The functionality should be covered by existing unit tests with unchanged behavior.

Types of changes

Checklist

facebook-github-bot commented 3 months ago

Hi @ahojnnes!

Thank you for your pull request and welcome to our community.

Action Required

In 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.

Process

In 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 CLA signed. The tagging process may take up to 1 hour after signing. Please give it that time before contacting us about it.

If you have received this in error or have any questions, please contact us at cla@meta.com. Thanks!

facebook-github-bot commented 3 months ago

Thank you for signing our Contributor License Agreement. We can now accept your code for this (and any) Meta Open Source project. Thanks!

ahojnnes commented 3 months ago

CI failures seem unrelated to the changes in this PR.

luisenp commented 3 months ago

Thanks so much for this fix. It's unfortunate how such a careless but small oversight had so large of an impact. I'll make another PR first to fix the CI errors. Then we can merge.

luisenp commented 3 months ago

@ahojnnes Ci should be fixed now if you rebase on top of latest main.

ahojnnes commented 3 months ago

@luisenp Thanks for fixing the CI. I rebased on latest master. Cheers.

ahojnnes commented 3 months ago

@luisenp Another (seemingly transient) CI failure. Might need a retrigger of the build, which I don't have permission for.