mx.nd.multi.lars

Description

Compute the LARS coefficients of multiple weights and grads from their sums of square”

Arguments

Argument

Description

lrs

NDArray-or-Symbol.

Learning rates to scale by LARS coefficient

weights.sum.sq

NDArray-or-Symbol sum of square of weights arrays

grads.sum.sq

NDArray-or-Symbol sum of square of gradients arrays

wds

NDArray-or-Symbol weight decays

eta

float, required.

LARS eta

eps

float, required.

LARS eps

rescale.grad

float, optional, default=1.

Gradient rescaling factor