train: train entrena la red neuronal usando una topologia de arbol

Description Usage Arguments Value Examples

View source: R/utils.R

Description

train entrena la red neuronal usando una topologia de arbol

Usage

1
2
train(numberOfChildrenperNode, treeHeight, initialLearningRate,
  finalLearningRate, initialRadius, finalRadius, numberOfIterations, data)

Arguments

numberOfChildrenperNode

A integer number

treeHeight

A integer number

initialLearningRate

A float number

finalLearningRate

A float number

initialRadius

A integer number

finalRadius

A integer number

numberOfIterations

A integer number

data

A data frame

Value

La red neuronal entrenada con topologia de un arbol k ario completo

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
library(ggplot2)
##################### EXAMPLE 1 : IRIS DATASET
###parameters
numberOfIterations <- 600000
initialLearningRate <- 1
finalLearningRate<- 0
initialRadius <- 7
finalRadius <- 1
numberOfChildrenperNode <- 2
treeHeight <- 3

##training phase
data(iris)
data<-iris[-5] ## load a dataset
ti <- proc.time() # start watch
neurons <- train(numberOfChildrenperNode,treeHeight,initialLearningRate,finalLearningRate,initialRadius,finalRadius,numberOfIterations, data)
tf <-proc.time()    # stop watch
tf-ti #print execution time

##visualization phase
graficar(data,neurons,numberOfChildrenperNode) #plot the scatter plot

Litier/CKTSOM documentation built on June 8, 2017, 12:05 a.m.