MRMR<T>
Models & Types · AiDotNet.Preprocessing.FeatureSelection.Bioinformatics
Minimum Redundancy Maximum Relevance (mRMR) feature selection.
Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> → MRMR<T>
Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>
For Beginners
Imagine choosing team members: you want people with the right skills (relevance to the task) but with different specialties so they don’t overlap (minimum redundancy). mRMR finds features that are predictive of the target but provide diverse, non-overlapping information.
How It Works
mRMR selects features that have high relevance to the target (maximum relevance) while minimizing redundancy among selected features. Originally developed for gene expression analysis.
Constructors
| Constructor | Summary |
|---|---|
MRMR(int nFeaturesToSelect, int nBins, MRMRMethod<T> method, int[] columnIndices) |
Properties
| Property | Summary |
|---|---|
Vector<T> MRMRScores { get; } |
|
bool SupportsInverseTransform { get; } |
|
int NFeaturesToSelect { 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) |