multiml.task.pytorch.modules.asng module

ASNG-NAS.

class multiml.task.pytorch.modules.asng.AdaptiveSNG(categories=None, integers=None, alpha=1.5, delta_init=1.0, lam=2, delta_max=inf, init_theta_cat=None, init_theta_int=None, threshold=0.1, patience=-1, range_restriction=True)

Bases: object

Adaptive Stochastic Natural Gradient for Categorical Distribution.

__init__(categories=None, integers=None, alpha=1.5, delta_init=1.0, lam=2, delta_max=inf, init_theta_cat=None, init_theta_int=None, threshold=0.1, patience=-1, range_restriction=True): AdaptiveSNG.

get_lambda()

check_converge()

converge_counter()

update_parameters(fnorm_cat, fnorm_int, hstack)

most_likely_value()

get_thetas()

set_thetas(theta_cat, theta_int)

sampling()

update_theta(c_cat, c_int, losses)

class multiml.task.pytorch.modules.asng.AdaptiveSNG_cat(categories=None, integers=None, alpha=1.5, delta_init=1.0, lam=2, delta_max=inf, init_theta_cat=None, init_theta_int=None, threshold=0.1, patience=-1, range_restriction=True)

Bases: AdaptiveSNG

check_converge()

most_likely_value()

get_thetas()

set_thetas(theta_cat, theta_int)

sampling()

update_theta(c_cat, c_int, losses)

class multiml.task.pytorch.modules.asng.AdaptiveSNG_int(categories=None, integers=None, alpha=1.5, delta_init=1.0, lam=2, delta_max=inf, init_theta_cat=None, init_theta_int=None, threshold=0.1, patience=-1, range_restriction=True)

Bases: AdaptiveSNG

check_converge()

most_likely_value()

get_thetas()

set_thetas(theta_cat, theta_int)

sampling()

update_theta(c_cat, c_int, losses)