cspnn.learn {spnn}

cspnn.learn

Description

Create or update a Condensed Scale Invariant Probabilistic Neural Network.

Usage

cspnn.learn(set, nn, xr, sigma, category.column = 1)

Arguments

set

data.frame or matrix representing the training set. The first column (default category.column = 1) is used to define the category or class of each observation.

nn

(optional) A Condensed Scale Invariant Probabilistic Neural Network object. If provided, the training data set input is concatenated to the current training data set of the neural network. If not provided, a new CSPNN object is created.

xr

The m by n reference matrix containing optimal parameters for probability estimation. Where m is the number of unique categories and n is the number of input factors used. This matrix must be provided.

sigma

An n by n square matrix of smoothing parameters where n is the number of input factors. Defaults to using the covariance matrix of the training data set excluding the category.column.

category.column

The column number of category data. Default is 1.

Details

The function cspnn.learn creates a new Condensed Scale Invariant Probabilistic Neural Network with a given training data set or updates the training data of an existing CSPNN. It sets the parameters: model, set, xr, category.column, categories, sigma, sigmaInverse, k, and n for the CSPNN.

Value

A trained Condensed Scale Invariant Probabilistic Neural Network (CSPNN)

See Also

spnn-package, cspnn.predict, iris

Examples

library(spnn)
library(datasets)

data(iris)

# shuffle the iris data set
indexRandom <- sample(1:nrow(iris), size = nrow(iris), replace = FALSE)

# use 100 observations for training set
trainData <- iris[indexRandom[1:100],]

# use remaining observations for testing
testData <- iris[indexRandom[101:length(indexRandom)],]

# reference matrix must be supplied
# this is not the optimal reference matrix
# this matrix is provided as a simple example
xr <- matrix(c(c(5.00, 3.41, 1.44, 0.24),
               c(5.88, 2.75, 4.23, 1.30),
               c(6.61, 2.97, 5.59, 2.01)),
             nrow = length(unique(trainData$Species)),
             ncol = ncol(trainData) - 1,
             byrow = TRUE)

# fit cspnn
cspnn <- cspnn.learn(set = trainData, xr = xr, category.column = 5)

# estimate probabilities
predictions <- cspnn.predict(nn = cspnn, newData = testData[,1:4])