[MRG] OT for Gaussian distributions #428
Merged
Conversation
This file contains 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
agramfort
reviewed
Jan 12, 2023
rflamary
reviewed
Jan 13, 2023
Co-authored-by: Alexandre Gramfort <[email protected]>
Co-authored-by: Alexandre Gramfort <[email protected]>
rflamary
reviewed
Jan 13, 2023
Codecov Report
Additional details and impacted files@@ Coverage Diff @@
## master #428 +/- ##
==========================================
+ Coverage 94.26% 94.43% +0.17%
==========================================
Files 23 24 +1
Lines 6204 6254 +50
==========================================
+ Hits 5848 5906 +58
+ Misses 356 348 -8 |
tgnassou
changed the title
rflamary
reviewed
Jan 16, 2023
ot/da.py
Outdated
| @@ -679,112 +680,12 @@ def df(G): | |||
| return G, L | |||
|
|
|||
|
|
|||
| @deprecated() | |||
There was a problem hiding this comment.
This line is enough since teh API did not change, also you did not apss the parameter to the function so it was false
Suggested change
| @deprecated() | |
| OT_mapping_linear=deprecated(empirical_bures_wasserstein_mapping) |
ot/gaussian.py
Outdated
| Cs12 = nx.sqrtm(Cs) | ||
|
|
||
| B = nx.trace(Cs + Ct - 2 * nx.sqrtm(dots(Cs12, Ct, Cs12))) | ||
| W = nx.norm(ms - mt) + B |
There was a problem hiding this comment.
Suggested change
| W = nx.norm(ms - mt) + B | |
| W = nx.sqrt(nx.norm(ms - mt)**2 + B) |
There was a problem hiding this comment.
norm shuould be squared and the Bures wassresein is the quare ropot of the term
ot/gaussian.py
Outdated
| Cs12 = nx.sqrtm(Cs) | ||
|
|
||
| B = nx.trace(Cs + Ct - 2 * nx.sqrtm(dots(Cs12, Ct, Cs12))) | ||
| W = nx.norm(mxs - mxt) + B |
There was a problem hiding this comment.
Same error here, you should call the function bures_wassrestein_distancehere anyways
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 introduce Gaussian modules, a file comprising all the modules of optimal transport for Gaussian distributions.
The OT_mapping_linear moved from da.py to gaussian.py.
I added Bures Wasserstein distance to gaussian.py.
Also update examples/gromov/plot_barycenter_fgw.py to fit the new networkx API
Motivation and context / Related issue
Add new methods for Gaussian distributions.
How has this been tested (if it applies)
The test for Bures Wasserstein distance is done on two 1D Gaussian distributions centered on 0 and 10 respectively. with the same variance The Bures Wasserstein distance is tested to be close of 10.
PR checklist