ott.solvers.utils.match_linear#

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

Compute solution to a linear OT problem.

Parameters:

x (Array) – Source point cloud of shape [n, d].
y (Optional[Array]) – Target point cloud of shape [m, d].
cost_fn (Optional[CostFn]) – Cost function.
epsilon (Optional[float]) – Regularization parameter.
scale_cost (Union[float, Literal['mean', 'max_cost', 'median']]) – Scaling of the cost matrix.
kwargs (Any) – Additional arguments for ott.solvers.linear.solve().

Return type:

Array

Returns:

Optimal transport matrix.

ott.solvers.utils.match_linear