ott.solvers.linear.sinkhorn_lr.LRSinkhornOutput

class ott.solvers.linear.sinkhorn_lr.LRSinkhornOutput(q, r, g, costs, errors, ot_prob, reg_ot_cost=None)

Implement the problems.Transport interface, for a LR Sinkhorn solution.

Methods

apply(inputs[, axis])	Apply the transport to a array; axis=1 for its transpose.
compute_reg_ot_cost(ot_prob[, use_danskin])	rtype float
cost_at_geom(other_geom)	Return OT cost for current solution, evaluated at any cost matrix.
count(value, /)	Return number of occurrences of value.
index(value[, start, stop])	Return first index of value.
marginal(axis)	rtype Array
set(**kwargs)	Return a copy of self, with potential overwrites.
set_cost(ot_prob, lse_mode[, use_danskin])	rtype LRSinkhornOutput
transport_cost_at_geom(other_geom)	Return (by recomputing it) bare transport cost of current solution.

Attributes

a	rtype Array
b	rtype Array
converged	rtype bool
costs	Alias for field number 3
errors	Alias for field number 4
g	Alias for field number 2
geom	rtype Geometry
linear	rtype bool
linear_output	rtype bool
matrix	Transport matrix if it can be instantiated.
ot_prob	Alias for field number 5
primal_cost	Return (by recomputing it) transport cost of current solution.
q	Alias for field number 0
r	Alias for field number 1
reg_ot_cost	Alias for field number 6
transport_mass	Sum of transport matrix.

Parameters

• q (jax.Array) –

• r (jax.Array) –

• g (jax.Array) –

• costs (jax.Array) –

• errors (jax.Array) –

• ot_prob (ott.problems.linear.linear_problem.LinearProblem) –

• reg_ot_cost (Optional[float]) –