ott.solvers.linear.sinkhorn.SinkhornState

Contents

ott.solvers.linear.sinkhorn.SinkhornState#

class ott.solvers.linear.sinkhorn.SinkhornState(potentials, errors=None, old_fus=None, old_mapped_fus=None)[source]#

Holds the state variables used to solve OT with Sinkhorn.

Methods

compute_kl_reg_cost(ot_prob, lse_mode)

param ot_prob:

count(value, /)

Return number of occurrences of value.

index(value[, start, stop])

Return first index of value.

recenter(f, g, ot_prob)

Re-center dual potentials.

set(**kwargs)

Return a copy of self, with potential overwrites.

solution_error(ot_prob, norm_error, *, ...)

State dependent function to return error.

Attributes

errors

Alias for field number 1

fu

The first dual potential or scaling.

gv

The second dual potential or scaling.

old_fus

Alias for field number 2

old_mapped_fus

Alias for field number 3

potentials

Alias for field number 0

Parameters: