PermutationImportance<T>
Models & Types · AiDotNet.Preprocessing.FeatureSelection.Embedded
Permutation Importance feature selection by measuring score decrease when shuffling features.
Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> → PermutationImportance<T>
Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>
For Beginners
Imagine you have a trained model and want to know which features it relies on. You shuffle one feature’s values randomly (like shuffling a deck of cards) and see how much worse the model performs. If performance drops a lot, that feature was important. If shuffling makes no difference, the feature wasn’t useful.
How It Works
Permutation Importance measures the decrease in model performance when a single feature’s values are randomly shuffled, breaking the relationship between the feature and the target. Features whose shuffling causes large performance drops are considered important.
Constructors
| Constructor | Summary |
|---|---|
PermutationImportance(int nFeaturesToSelect, int nRepeats, Func<Matrix<T>, Vector<T>, T> scorer, int? randomState, int[] columnIndices) |
Properties
| Property | Summary |
|---|---|
T BaselineScore { get; } |
|
Vector<T> ImportanceStds { get; } |
|
Vector<T> Importances { get; } |
|
bool SupportsInverseTransform { get; } |
|
int NFeaturesToSelect { get; } |
|
int NRepeats { 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) |