TransferLearningSelector<T>

Models & Types ยท AiDotNet.Preprocessing.FeatureSelection.Transfer

Transfer Learning-based Feature Selection.

Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> โ†’ TransferLearningSelector<T>

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

Sometimes we have lots of data from one situation (like medical records from one hospital) but little data from another (a new hospital). Transfer learning uses what we learned about important features from the first situation to help with the second.

How It Works

Uses feature importance learned from a source domain to guide feature selection in a target domain. Particularly useful when the target domain has limited data but the source domain has abundant labeled data.

Constructors

Constructor Summary
TransferLearningSelector(int nFeaturesToSelect, Vector<T> sourceImportances, double transferWeight, int[] columnIndices)

Properties

Property Summary
Vector<T> CombinedImportances { get; }
Vector<T> TargetImportances { get; }
bool SupportsInverseTransform { get; }
double TransferWeight { 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)