[WIP] batch FUGW loss

[smazelet]

Types of changes

This PR aims at adding the Fused Unbalanced Gromov Wasserstein (FUGW) loss to ot.batch, in order to compute N FUGW losses in parallel, for training neural networks for example.

Motivation and context / Related issue

PR#775 added the ot.batch module for computing N optimal transport problems in parallel. Both the linear and quadratic problems are available but not the FUGW problem, which is useful in some applications.

Also fixed two small errors in ot.batch._quadratic loss_quadratic_samples_batch. According to the documentation, the default value for the symmetric parameter should be True instead of None. Additionally, in ot.batch._linear, the default value for the metric should be "sqeuclidean" instead of "l2" in function loss_linear_samples_batch. A test was added to check that the default values are correct in test.batch.test_solve_batch.py.

How has this been tested (if it applies)

I added two tests to test/batch/test_solve_gromov_batch.py.

PR checklist

• [ x] I have read the CONTRIBUTING document.
• The documentation is up-to-date with the changes I made (check build artifacts).
• All tests passed, and additional code has been covered with new tests.
• I have added the PR and Issue fix to the RELEASES.md file.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 96.76%. Comparing base (048f3ae) to head (be772c1).

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #775      +/-   ##
==========================================
+ Coverage   96.74%   96.76%   +0.02%     
==========================================
  Files         118      118              
  Lines       23466    23613     +147     
==========================================
+ Hits        22703    22850     +147     
  Misses        763      763              

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.