Learn R Programming

deepspat (version 0.3.1)

init_learn_rates: Initialise learning rates

Description

Provides utility to alter the learning rates when fitting a deepspat model

Usage

init_learn_rates(
  sigma2y = 5e-04,
  covfun = 0.01,
  sigma2eta = 1e-04,
  eta_mean = 0.1,
  eta_mean2 = 0.1,
  eta_sd = 0.1,
  LFTpars = 0.01,
  AFFpars = 0.01,
  rho = 0.1,
  vario = 0.1
)

Value

init_learn_rates returns a list with the learning rates. Call str(init_learn_rates()) to see the structure of this list.

Arguments

sigma2y

learning rate for the measurement-error variance

covfun

learning rate for the covariance-function (or matrix) parameters at the top layer

sigma2eta

learning rate for the process variance

eta_mean

learning rate for the weight estimates or variational means

eta_mean2

learning rate for the weight estimates or variational means

eta_sd

learning rate for the variational standard deviations (SDSP only)

LFTpars

learning rate for the parameters of the Mobius transformation

AFFpars

learning rate for the parameters of the affine transformation

rho

learning rate for the correlation parameter in the multivariate model

vario

learning rate for the parameter in the variogram