TrendFS<T>
Models & Types · AiDotNet.Preprocessing.FeatureSelection.TimeSeries
Trend-based feature selection for time series data.
Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> → TrendFS<T>
Implements: IFixedCountFeatureSelector<T>
For Beginners
This method looks for features that are consistently going up or down over time, like a stock price in a bull market. Features without clear direction (just random fluctuation) score low. It’s useful when you care about the overall trajectory, not just short-term patterns.
How It Works
Trend-based feature selection identifies features that exhibit significant temporal trends (upward or downward movement over time). Features with strong trends may be useful for long-term forecasting or detecting changes.
Constructors
| Constructor | Summary |
|---|---|
TrendFS(int nFeaturesToSelect, bool absoluteTrend, int[] columnIndices) |
Properties
| Property | Summary |
|---|---|
Vector<T> TrendScores { get; } |
|
bool AbsoluteTrend { get; } |
|
bool SupportsInverseTransform { get; } |
|
int NFeaturesToSelect { get; } |
|
int[] SelectedIndices { get; } |
Methods
| Method | Summary |
|---|---|
bool[] GetSupportMask() |
|
string[] GetFeatureNamesOut(string[] inputFeatureNames) |