train: Train a ELM

In mugiro/elm: Extreme Learning Machine for regression and classification

Description

train fits all the parameters that include a ELM given a set of input data (X, Y) and a training scheme

Usage

train(object, ...)

Arguments

object	ELM object to serialize
...	None to use until now
x	a data matrix of dimensions [Nxd] with input data
y	vector/matrix of outputs [Nx1c]
modelStrSel	logical Select the pruning for reduce model's size.
ranking	Type of neurons ranking random or lars.
validation	Method to validate the model developed "none" - no validation process "V" - validation. Xv and Yv are required "CV" - cross validation. The number of folds is required "LOO" - leave one out based on the PRESS statistic
folds	Number of folds defined for the cross-validation procedure
class_weights	numeric vector of length = number_of_classes with the weigths for weighted classification
classification	"none"/"mc"/"ml"/"w"

mugiro/elm documentation built on May 23, 2019, 8:21 a.m.