ott.solvers.utils.match_quadratic#

ott.solvers.utils.match_quadratic(xx, yy, x=None, y=None, scale_cost=1.0, cost_fn=None, **kwargs)[source]#

Compute solution to a quadratic OT problem.

Parameters:

xx (Array) – Source point cloud of shape [n, d1].
yy (Array) – Target point cloud of shape [m, d2].
x (Optional[Array]) – Linear (fused) term of the source point cloud.
y (Optional[Array]) – Linear (fused) term of the target point cloud.
scale_cost (Union[float, Literal['mean', 'max_cost', 'median']]) – Scaling of the cost matrix.
cost_fn (Optional[CostFn]) – Cost function.
kwargs (Any) – Additional arguments for ott.solvers.quadratic.solve().

Return type:

Array

Returns:

Optimal transport matrix.

ott.solvers.utils.match_quadratic