moscot.backends.ott.GWSolver¶
- class moscot.backends.ott.GWSolver(jit=True, rank=-1, initializer=None, initializer_kwargs=mappingproxy({}), linear_solver_kwargs=mappingproxy({}), **kwargs)[source]¶
Solver for the quadratic problem [Mémoli, 2011].
The Gromov-Wasserstein (GW) problem involves two distribution in possibly two different spaces. Points in the source distribution are matched to points in the target distribution by comparing the relative location of the points within each distribution.
- Parameters:
rank (
int
) – Rank of the solver. If -1 use the full-rank GW [Peyré et al., 2016], otherwise, use the low-rank approach [Scetbon et al., 2021].initializer (
Union
[BaseQuadraticInitializer
,None
,LRInitializer
,Literal
['random'
,'rank2'
,'k-means'
,'generalized-k-means'
]]) – Initializer forGromovWasserstein
.initializer_kwargs (
Mapping
[str
,Any
]) – Keyword arguments for theinitializer
.linear_solver_kwargs (
Mapping
[str
,Any
]) – Keyword arguments forSinkhorn
orLRSinkhorn
, depending on therank
.kwargs (
Any
) – Keyword arguments forGromovWasserstein
.
Methods
__call__
([xy, x, y, tags, device])Solve an optimal transport problem.
Attributes