initialize.ftrlprox: Initialize empty FTRL Proximal class

In FTRLProximal: FTRL Proximal Implementation for Elastic Net Regression

Description

Online elastic net regression using the FTRL Proximal algorithm for training.

Usage

initialize.ftrlprox(theta, levels, lambda, alpha, a, b = 1, save_loss = F,
  ...)

Arguments

theta	named numeric containing initial coefficients
levels	character vector containing class labels of target label
lambda	regularization term
alpha	mixing parameter, alpha=0 corresponds to L2 regularization and alpha=1 to L1.
a	learning rate parameter
b	learning rate parameter controlling decay, defaults to 1.
save_loss	is to save the loss function during training.
...	additional args

Details

This method is intended for setting up a ftrlprox model object before training it using update.

Value

ftrlprox model object

Author(s)

Vilhelm von Ehrenheim

FTRLProximal documentation built on May 29, 2017, 5:39 p.m.