[Fix] Gradient scaling in Partial GW solver #602
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Outdated suggestions cannot be applied.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
I modify the code ot.partial.partial_gromov_wasserstein
Motivation and context / Related issue
There seems to be an inconsistency between ot.partial.partial_gromov_wasserstein and the line search section in the paper [29]. I fixed this section. In addition, I have made minor change to the initial guess in the partial-GW solver since the original initial guess is np.out(p,q), which might not be suitable for unbalanced case, i.e. |p|\neq |q|.
N/A
N/A
How has this been tested (if it applies)
PR checklist