dprep (version 2.1)

cv10mlp: 10-fold cross validation error estimation for the multilayer perceptron classifier

Description

10-fold cross validation estimation error for the multilayer perceptron classifier.

Usage

cv10mlp(data, units, decay = 0, maxwts = 1000, maxit = 100, 
repet)

Arguments

data

The name of the dataset

units

The number of units in the hidden layer

decay

The decay parameter

maxwts

The maximum number of weights to be estimated in the network

maxit

The maximum number of iterations

repet

The number of repetitions

Value

• Returns the mean cross validation for the multilayer perceptron classifier.

References

Ripley, B.D. (1996). Pattern recognition and Neural networks. Cambridge University Press Venables,W.N., and Ripley, B.D. (2002). Modern Applied Statistics with S. Fourth edition, Springer

See Also

crossval, cv10log

Examples

#-----cross validation using the MLP classifier---
data(heartc)
cv10mlp(heartc,25,decay=0.1,maxwts=1000,maxit=100,repet=2)