normalizeMatrix: Normalize design matrix

In SPOT: Sequential Parameter Optimization Toolbox

Normalize design matrix

Description

Normalize design by using minimum and maximum of the design values for input space. Each column has entries in the range from ymin to ymax.

Usage

normalizeMatrix(x, ymin, ymax, MARGIN = 2)

Arguments

x	design matrix in input space
ymin	minimum vector of normalized space
ymax	maximum vector of normalized space
MARGIN	a vector giving the subscripts which the function will be applied over. 1 indicates rows, 2 indicates columns. Default: 2.

Value

list with the following entries:

y

normalized design matrix in the range [ymin, ymax]

xmin

min in each column

xmax

max in each column

See Also

buildKriging

Examples

set.seed(1)
x <- matrix(c(rep(1,3), rep(2,3),rep(3,3), rep(4,3)),3,4)
## columnwise:
normalizeMatrix(x, ymin=0, ymax=1)
## rowwise
normalizeMatrix(x, ymin=0, ymax=1, MARGIN=1)
# rows with identical values are mapped to the mean:
x <- matrix(rep(0,4),2,2)
normalizeMatrix(x, ymin=0, ymax=1)

SPOT documentation built on June 26, 2022, 1:06 a.m.