ReliefF<T>
Models & Types · AiDotNet.Preprocessing.FeatureSelection.Filter.Relief
ReliefF algorithm for feature selection.
Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> → ReliefF<T>
Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>
For Beginners
ReliefF works by looking at each data point and comparing it to its nearest neighbors. Features are scored based on how well they separate a point from neighbors of different classes while keeping it close to neighbors of the same class. It’s like finding features that cluster similar items together.
How It Works
ReliefF is an extension of the Relief algorithm that handles multi-class problems and uses k nearest neighbors. It evaluates features based on their ability to distinguish between instances of different classes.
Constructors
| Constructor | Summary |
|---|---|
ReliefF(int nFeaturesToSelect, int nNeighbors, int nSamples, int? randomState, int[] columnIndices) |
Properties
| Property | Summary |
|---|---|
Vector<T> FeatureWeights { get; } |
|
bool SupportsInverseTransform { get; } |
|
int NFeaturesToSelect { get; } |
|
int NNeighbors { get; } |
|
int NSamples { get; } |
|
int[] SelectedIndices { get; } |
Methods
| Method | Summary |
|---|---|
Matrix<T> FitTransform(Matrix<T> data, Vector<T> target) |
|
bool[] GetSupportMask() |
|
string[] GetFeatureNamesOut(string[] inputFeatureNames) |
|
void Fit(Matrix<T> data, Vector<T> target) |