MRMR<T>
Models & Types · AiDotNet.Preprocessing.FeatureSelection.Filter.Multivariate
Minimum Redundancy Maximum Relevance (mRMR) feature selection.
Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> → MRMR<T>
Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>
For Beginners
mRMR tries to pick features that are both useful (related to what you’re predicting) and diverse (not repeating the same information). It’s like assembling a team - you want skilled people, but also people with different skills rather than everyone being good at the same thing.
How It Works
mRMR selects features that have maximum relevance to the target while having minimum redundancy among themselves. It balances finding informative features with avoiding duplicated information.
Constructors
| Constructor | Summary |
|---|---|
MRMR(int nFeaturesToSelect, int nBins, 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) |