POT Python Optimal Transport

RFRémi FlamaryVCVincent-Cuaz, CédricNCNicolas CourtyAGAlexandre GramfortKOKachaiev, Oleksii

École Polytechnique · Centre de Mathématiques Appliquées de l'École polytechnique · +29 more institutions

Indexed indatacite

Abstract

This new release contains several new features and bug fixes. Among the new features we have a new submodule ot.batch that contains batch parallel solvers for several OT problems including Sinkhorn, Gromov-Wasserstein and Fused Gromov-Wasserstein. This new submodule can be used to solve many independent OT problems in parallel on CPU or GPU with shared source and target support sizes. We also implemented a new Nystrom kernel approximation for the Sinkhorn solver that can be used to speed up the computation of the Sinkhorn divergence on large datasets. We also added new 1D solvers for Linear circular OT and new solvers for free support OT barycenters with generic cost functions and for barycenters between…

Citation impact

43
total citations
FWCI
Percentile
References
0
Citations per year

Authors

21

Topics & keywords

Keywords
  • Python (programming language)
  • Computer science
  • Programming language
No related works found for this paper.