POT

POT copied to clipboard

### Bug Description When calling `ot.dist` with metric set to hamming an AttributeError is thrown from within `scipy.cdist` ### To Reproduce (tested using python 3.10, ot version 0.8.2 and scipy...

nklkhlr

ot.emd with large datasets

6

## 🚀 Feature I was wondering if it would be possible to do the OT matrix computation (ot.emd) using sparse matrices, (eg: https://docs.scipy.org/doc/scipy/reference/generated/scipy.sparse.coo_array.html#scipy.sparse.coo_array) where only the distances between pairs of...

johnalison

Demo of multidimensional Wasserstein distance

2

Hi, I've just come across your library while studying Wasserstein distances and OT, and was hoping to compute some simple examples like the one shown in Lilian Weng's blog post:...

lmmx

[WIP] Gaussian initialization for sinkhorn

3

## Types of changes This is a first shot for intializing empirical sinkhorn but the computational gain is not very clear on tests I did locally. The following code ```python...

rflamary

[WIP] [DA] Sinkhorn LpL1 transport to work on JAX

1

## Types of changes This PR presents a vectorized implementation of LpL1 transport algorithms for domain adaptation, along with all the requisite modifications to ensure compatibility of the `ot.da` module...

kachayev

[WIP] quantized gromov wasserstein solver

1

## Types of changes Implementation of first solvers for the quantized Fused Gromov-Wasserstein (qFGW) distance. - creation of new files `ot/gromov/_quantized.py` and `test/gromov/test_quantized.py` . - `quantized_fused_gromov_wasserstein_partitioned` : main function to...

cedricvincentcuaz

This feature arguably falls outside the scope of the library - it is really a problem which is 2 dependencies down the food chain - but I will suggest it...

patrick-nicodemus

[Fix] Gradient scaling in Partial GW solver

4

## 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...

yikun-baio

Can not batch `ot.emd2` via `torch.vmap`

1

## Describe the bug As my datapoints are empirical distributions I want to use the Wasserstein distance as a loss function over a batch of shape `(n_batch, n_points, dimension)`. Standard...

oleg-kachan

Stricter `check_params` protocol for DA module would simplify debugging

1

## Describe the bug `ot.utils.check_params` is used rather heavily throughout `ot.da` functions. The function itself checks for `None`(s) in the input and returns `False` if at least one was spotted....

kachayev