predict-methods: predict method for S4 class 'SOMnn'

In som.nn: Topological k-NN Classifier Based on Self-Organising Maps

Description

Predicts categories for a table of data, based on the hexagonal som in the model. This S4 method is a wrapper for the predict method stored in the slot predict of a model of type SOMnn.

Usage

## S4 method for signature 'SOMnn'
predict(object, x)

Arguments

object	object of type SOMnn.
x	data.frame with rows of data to be predicted.

Details

The function returns the winner neuron in codes for each test vector in x. x is organised as one vector per row and must have the same number of columns (i.e. dimensions) and the identical column names as stored in the SOMnn object.

If data have been normalised during training, the same normalisation is applied to the unknown data to be predicted.

Probablilities are softmax normalised by default.

Value

data.frame with columns: winner, x, y, the predicted probabilities for all categories and the prediction as category index (column name prediction) and class label (column name pred.class).

Examples

## get example data and add class labels:
data(iris)
species <- iris$Species

## train with default radius = diagonal / 2:
som <- som.nn.train(iris, class.col = "Species", 
                    xdim = 15, ydim = 9, alpha = 0.02, len = 10000, 
                    norm = TRUE, toroidal = FALSE)

## predict some samples:
unk <- iris[,!(names(iris) %in% "Species")]

setosa <- unk[species=="setosa",]
setosa <- setosa[sample(nrow(setosa), 20),]

versicolor <- unk[species=="versicolor",]
versicolor <- versicolor[sample(nrow(versicolor), 20),]

virginica <- unk[species=="virginica",]
virginica <- virginica[sample(nrow(virginica), 20),]

## predict with generic function:
predict(som, unk)

## predict with function in SOMnn object:
som@predict(setosa)
som@predict(versicolor)
som@predict(virginica)

## get mapping with visual:
som.nn.visual(som@codes, setosa)


## change parameters of k-NN classifier:
som.nn.set(som, iris, dist.fun = dist.fun.bubble, max.dist = 3.1)
som.nn.set(som, iris, dist.fun = dist.fun.tricubic, max.dist = 3)

## define custom distance function:
som <- som.nn.set(som, iris, dist.fun = function(x, sigma){
            ifelse(x < sigma, dnorm(x, sd = sigma), 0)}, 
            max.dist = 3)

predict(som, setosa)
som

som.nn documentation built on May 2, 2019, 8:26 a.m.