wcls: Weighted Classification Methods

In SkadiEye/ITRlearn: Individualized Treatment Rule Learning

Description

A collection of functions to build weighted classification models.

Usage

wclskSVM(object, valid = NULL, ...)

wclslSVM(object, valid = NULL, ...)

wclsLASSO(object, valid = NULL, ...)

wclsANN(object, valid = NULL, ...)

wclsDNN(object, valid = NULL, ...)

## S4 method for signature 'wClsObj'
wclskSVM(object, valid = NULL, ...)

## S4 method for signature 'wClsObj'
wclslSVM(object, valid = NULL, ...)

## S4 method for signature 'wClsObj'
wclsLASSO(object, valid = NULL, ...)

## S4 method for signature 'wClsObj'
wclsANN(object, valid = NULL, ...)

## S4 method for signature 'wClsObj'
wclsDNN(object, valid = NULL, ...)

Arguments

object	A wClsObj object, used as a training set.
valid	A wClsObj object, used as a validation set.
...	Arguments passed to the function being called.

Details

These wcls methods construct a single weighted classification model given a wClsObj and a set of parameters. This function returns a ModelObj.

Value

Returns a ModelObj object.

Methods (by generic)

wclskSVM: Method to build a (non-linear) kernel SVM weighted classification model using w.k.svm function.

wclslSVM: Method to build a linear SVM weighted classification model using w.l.svm function.

wclsLASSO: Method to build a weighted classification model using glmnet::glmnet.

wclsANN: Method to build a weighted classification model using nnet::nnet.

wclsDNN: Method to build a deep neural network weighted classification model using nn.classifier function.

See Also

reg
wcls
cvGrid
cvKfold
workflow

@examples

SkadiEye/ITRlearn documentation built on May 24, 2019, 1:31 a.m.