POT icon indicating copy to clipboard operation
POT copied to clipboard

POT : Python Optimal Transport

Results 100 POT issues
Sort by recently updated
recently updated
newest added

## Types of changes This PR aims to add the Linear Circular OT distance (with uniform measure on the circle as reference measure) introduced in [LCOT: Linear Circular Optimal Transport...

## Types of changes - [x] contribute a generalisation of `free_support` that accepts any cost function (implements [this paper](https://arxiv.org/abs/2501.04016)) - [x] contribute a specific version of this function in `ot.gmm`...

CI
Tests
ot.lp
Examples
ot.gmm

## Types of changes - creation of the class `ot.utils.BaryResult` extending `OTResult` first draft for OT barycenter solvers with free support and predefined metric `ot.solvers.bary_free_support` built on top of `ot.solvers.solve_sample`,...

Tests
ot.utils
ot.solvers
ot.unbalanced

## Types of changes typo correction for doc display. ## PR checklist - [x] I have read the [**CONTRIBUTING**](https://pythonot.github.io/contributing.html) document. - [x] The documentation is up-to-date with the changes I...

ot.utils

## Types of changes fix undesirable behaviour of `ot.unbalanced.lbfgsb_unbalanced` when `isinstance(reg_div, tuple)` to test whether included functions in reg_div can handle numpy arrays. These tests are now handled via `ot.utils.fun_to_numpy`....

Tests
ot.utils
ot.unbalanced

1. add a new function called entropic\_partial\_wasserstein\_logscale to ot.partial. This function solves the same problem as entropic\_partial\_wasserstein but is computed in logscale, so it is more robust. 2. Test exampless...

Examples
ot.partial

## Describe the bug Forgive me if I've misunderstood what the `wasserstein_circle` function is supposed to do. But I would have thought that if we have ```python d1 = wasserstein_circle(arr1,...

bug
help wanted

## Types of changes Added implementation of the paper Low-Rank Optimal Transport through Factor Relaxation with Latent Coupling https://arxiv.org/abs/2411.10555 ## How has this been tested (if it applies) Tested it...

## Describe the bug The entropic_partial_wasserstein function produces nan when eps is small. ### To Reproduce ``` import ot import torch import numpy as np def compute_OT(M, alpha, beta, epsilon):...

bug
help wanted

The goal is to return a sparse transport in `ot.emd` plan when the cost matrix matrix is itself sparse which corresponds to infinite cost between points, or when no value...

ot.lp