ott.neural.methods.expectile_neural_dual.ENOTPotentials

Contents

ott.neural.methods.expectile_neural_dual.ENOTPotentials#

class ott.neural.methods.expectile_neural_dual.ENOTPotentials(grad_f, g, cost_fn, *, is_bidirectional, corr)[source]#

The dual potentials of the ENOT method [Buzun et al., 2024].

Parameters:

  • grad_f (Callable[[Array], Array]) – Gradient of the first dual potential function.

  • g (Callable[[Array], Array]) – The second dual potential function.

  • cost_fn (CostFn) – The cost function used to solve the OT problem.

  • is_bidirectional (bool) – Whether the duals are trained for bidirectional transport mapping.

  • corr (bool) – Whether the duals solve the problem in correlation form.

Methods

distance(src, tgt)

Evaluate Wasserstein distance between samples using dual potentials.

plot_ot_map(source, target[, samples, ...])

Plot data and learned optimal transport map.

plot_potential([forward, quantile, ...])

Plot the potential.

transport(vec[, forward])

Transport vec according to Gangbo-McCann Brenier [Brenier, 1991].

Attributes

f

The first dual potential function.

g

The second dual potential function.
Contents