ott.solvers.linear.sinkhorn.solve#

ott.solvers.linear.sinkhorn.solve(geom, a=None, b=None, tau_a=1.0, tau_b=1.0, rank=- 1, **kwargs)[source]#

Solve linear regularized OT problem using Sinkhorn iterations.

Parameters

  • geom (Geometry) – The ground geometry cost of the linear problem.

  • a (Optional[Array]) – The first marginal. If None, it will be uniform.

  • b (Optional[Array]) – The second marginal. If None, it will be uniform.

  • tau_a (float) – If < 1, defines how much unbalanced the problem is on the first marginal.

  • tau_b (float) – If < 1, defines how much unbalanced the problem is on the second marginal.

  • rank (int) – Rank constraint on the coupling to minimize the linear OT problem [Scetbon et al., 2021]. If -1, no rank constraint is used.

  • kwargs (Any) – Keyword arguments for Sinkhorn or LRSinkhorn, depending rank.

Return type

Union[SinkhornOutput, LRSinkhornOutput]

Returns

The Sinkhorn output.