minmaxnormalize

Normalizes dataset into specified range of values using Min Max Normalization. It transforms the dataset by scaling each feature to a given range.

Attention: Available only with Activate commercial edition.

Syntax

[X_normalized, parameters] = minmaxnormalize(X)

[X_normalized, parameters] = minmaxnormalize(X, range)

Inputs

X: Input data to be normalized.; Type: double; Dimension: vector | matrix
range: Range within which the normalized values should lie. 'min_range' and 'max_range' should be replaced with range numbers (default: [0,1]). It must contain two values [min_range, max_range].; Type: double; Dimension: vector

Outputs

X_normalized: Normalized input records.; Type: double; Dimension: vector | matrix
parameters: Struct containing minimum and maximum of each column.; min: minimum of each column.; max: maximum of each column.; min_range: minimum range given as input.; max_range: maximum range given as input.; Type: double | integer; Dimension: struct

Example

Usage of minmaxnormalize

X = [1, -1, 2;
     2, 0, 0;
     0, 1, -1];
[X_scaled, parameters] = minmaxnormalize(X, [0.25, 0.35]);

> X_scaled
X_scaled = [Matrix] 3 x 3
0.30000 0.25000 0.35000
0.35000 0.30000 0.28333
0.25000 0.35000 0.25000
> parameters
parameters = struct [
max: [Matrix] 1 x 3
2 1 2
max_range: 0.35
min: [Matrix] 1 x 3
0 -1 -1
min_range: 0.25
]

Comments

Output 'parameters' can be used by minmaxnormalizetransform function.