Théo Lacombe
Théo Lacombe
> IMHO the copyright is correct. Great ! Though, shouldn't it be 2021, now that I read it?
Corrected. Any thought on if this code belong to the `/wasserstein` module? From a theoretical perspective, the goal is to solve some minimization with respect to the Wasserstein distance between...
I just realized that I never managed to do the last requested modifications (my local build was broken for some reason at that time). I finally did it. As I'm...
Ok I could reproduce the issue with your mwe. What should we do, forward this to POT? In your initial example, were the points with 1e16 coordinates representing ``essential'' parts...
@ncourty I may be missing something but I tried the same code using normalization in the `ot.emd` but this didn't solve the issue. I still get the same (non-optimal) transport...
As far as I can tell this was acknowledged by @rflamary [here](https://github.com/PythonOT/POT/issues/421#issuecomment-1358949197) and a PR with a small documentation of this behavior was proposed in #423 .(idk if it is...
Yes it really looks like an implementation issue (or maybe something deeper, like the IBP not converging?). I have no clue on what is happening yet. I did not investigate...
Noticed. I will do a PR incorporating soon.