[MRG] Linear Circular OT #736
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #736 +/- ##
========================================
Coverage 97.10% 97.11%
========================================
Files 100 100
Lines 20453 20697 +244
========================================
+ Hits 19861 20099 +238
- Misses 592 598 +6 🚀 New features to boost your workflow:
|
rflamary
reviewed
Aug 13, 2025
| tc[mask_end > 0] = ( | ||
| (Ctp - Ctm + tm * dCptm - tp * dCmtp) / (dCptm - dCmtp) | ||
| )[mask_end > 0] | ||
| with warnings.catch_warnings(): |
There was a problem hiding this comment.
this is a bit worrying. why are you catching tjose worning? when do they happen. It is probaly OK but we need at last a comment there to explian
| return wasserstein1_circle( | ||
| u_values, v_values, u_weights, v_weights, require_sort | ||
| ) | ||
| # if p == 1: |
There was a problem hiding this comment.
remove the code if not usefull anymore
|
|
||
| Returns | ||
| ------- | ||
| loss: float |
There was a problem hiding this comment.
is this true or is this of shape (...) instead?
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Types of changes
This PR aims to add the Linear Circular OT distance (with uniform measure on the circle as reference measure) introduced in LCOT: Linear Circular Optimal Transport
and its sliced counterpart introduced in Linear Spherical Sliced Optimal Transport: A Fast Metric for Comparing Spherical Data
.
Changes:
ot.solver_1d.linear_circular_embeddingandot.solver_1d.linear_circular_otfunctions to compute the Linear Circular OT distance.ot.sliced.linear_sliced_wasserstein_spherefunction to compute its sliced counterpart.ot.sliced.get_projections_sphereandot.sliced.projection_sphere_to_circle.test/test_1d_solverandtest/test_sliced.plot_compute_wasserstein_circle.pyot.wasserstein_circleto always useot.solver_1d.binary_search_circlefor any p, which seems to give better results even for p=1 (as discussed in Issue Wasserstein Circle distance doesn't seem correct? #738).plot_ssw_unif_torch.py.Motivation and context / Related issue
The Linear Circular OT distance with uniform measure on the circle is very fast to compute, as the OT maps can be computed in closed-forms. It is thus faster to compute than the true wasserstein distance on the circle.
How has this been tested (if it applies)
I added tests of these functions in
test/test_1d_solverandtest/test_sliced.PR checklist