Within-Sample Min-Max Normalization

Scales each sample's feature values to [0,1] range using only that sample's minimum and maximum. Completely batch-immune.

Usage

ws_minmax(x)

Arguments

x

A numeric matrix (samples x features) or vector (single sample)

Value

Matrix of same dimensions with values in [0,1]

Examples

if (FALSE) { # \dontrun{
# Single sample
ws_minmax(c(5.2, 3.1, 8.7, 2.4))
# Matrix (samples x features)
ws_minmax(matrix(rnorm(100), nrow=10))
} # }