mlpack
Public Member Functions | List of all members
mlpack::data::MinMaxScaler Class Reference

A simple MinMax Scaler class. More...

#include <min_max_scaler.hpp>

Public Member Functions

 MinMaxScaler (const double min=0, const double max=1)
 Default constructor. More...
 
template<typename MatType >
void Fit (const MatType &input)
 Function to fit features, to find out the min max and scale. More...
 
template<typename MatType >
void Transform (const MatType &input, MatType &output)
 Function to scale features. More...
 
template<typename MatType >
void InverseTransform (const MatType &input, MatType &output)
 Function to retrieve original dataset. More...
 
const arma::vec & ItemMin () const
 Get the Min row vector.
 
const arma::vec & ItemMax () const
 Get the Max row vector.
 
const arma::vec & Scale () const
 Get the Scale row vector.
 
double ScaleMax () const
 Get the upper range parameter.
 
double ScaleMin () const
 Get the lower range parameter.
 
template<typename Archive >
void serialize (Archive &ar, const uint32_t)
 

Detailed Description

A simple MinMax Scaler class.

Given an input dataset this class helps you to scale each feature to a given range.

[z = scale * x + scaleMin - min(x) * scale] [scale = (scaleMax - scaleMin) / (max(x) - min(x))]

where scaleMin, scaleMax = feature_range and min(x), max(x) are the minimum and maximum value of x respectively.

arma::mat input;
Load("train.csv", input);
arma::mat output;
// Fit the features.
scale.Fit(input)
// Scale the features.
scale.Transform(input, output);
// Retransform the input.
scale.InverseTransform(output, input);

Constructor & Destructor Documentation

◆ MinMaxScaler()

mlpack::data::MinMaxScaler::MinMaxScaler ( const double  min = 0,
const double  max = 1 
)
inline

Default constructor.

Parameters
minLower range of scaling.
maxUpper range of scaling.

Member Function Documentation

◆ Fit()

template<typename MatType >
void mlpack::data::MinMaxScaler::Fit ( const MatType &  input)
inline

Function to fit features, to find out the min max and scale.

Parameters
inputDataset to fit.

◆ InverseTransform()

template<typename MatType >
void mlpack::data::MinMaxScaler::InverseTransform ( const MatType &  input,
MatType &  output 
)
inline

Function to retrieve original dataset.

Parameters
inputScaled dataset.
outputOutput matrix with original Dataset.

◆ Transform()

template<typename MatType >
void mlpack::data::MinMaxScaler::Transform ( const MatType &  input,
MatType &  output 
)
inline

Function to scale features.

Parameters
inputDataset to scale features.
outputOutput matrix with scaled features.

The documentation for this class was generated from the following file: