map.build: Build Map

In lutzhamel/popsom2: Functions for Constructing and Evaluating Self-Organizing Maps

Description

Constructs a SOM, returns an object of class 'map'.

Usage

map.build(data,labels=NULL,xdim=10,ydim=5,alpha=0.3,train=1000,algorithm="vsom")

Arguments

data	a dataframe where each row contains an unlabeled training instance.
labels	a vector or dataframe with one label for each observation in data.
xdim	the x-dimension of the map. (default=10)
ydim	the y-dimension of the map. (default=5)
alpha	the learning rate, should be a positive non-zero real number. (default=0.3)
train	the number of training iterations. (default=1000)
algorithm	training algorithm selection switch. (default="vsom")

Value

object of type 'map'.

Note

You have a choice of training algorithms:

• "vsom" - vectorized stochastic learning, this is a highly optimized version of stochastic training written in FORTRAN 9X.

• "som" - the traditional stochastic learning algorithm written in C++/TNT.

• "batchsom" - batch version of the SOM training algorithm, written in C.

• "experimental" - experimental implementation of vectorized stochastic learning implemented directly in R.

Note

If your training data does not have any labels you can construct a simple label vector as follows: labels <- 1:nrow(training.data). If you let the labels default to the NULL value then no labels will be shown in the map visualization.

Author(s)

Lutz Hamel, Benjamin Ott, Gregory Breard

Examples

data(iris)

## set data frame and labels
df <- subset(iris,select=-Species)
labels <- subset(iris,select=Species)

## build a map
m <- map.build(df, labels, xdim=15, ydim=10, train=1000)

lutzhamel/popsom2 documentation built on May 25, 2020, 12:46 a.m.