OnlineSlidingWindowSplitter<T>
Models & Types · AiDotNet.Preprocessing.DataPreparation.Splitting.Online
Online sliding window splitter for streaming data with concept drift adaptation.
Inheritance: DataSplitterBase<T> → OnlineSlidingWindowSplitter<T>
For Beginners
Unlike the landmark window, a sliding window maintains a fixed-size training set that moves forward in time. Old samples are “forgotten” as new ones arrive. This is essential when data patterns change over time (concept drift).
How It Works
How It Works:
- Maintain a window of the most recent N samples for training
- Test on the next batch of samples
- Slide the window forward, dropping oldest samples
Window Strategies:
- Fixed: Constant window size throughout
- Adaptive: Adjust window size based on performance
- Fading: Weight recent samples more heavily (conceptually)
When to Use:
- Non-stationary data (concept drift)
- Real-time systems with memory constraints
- When recent data is more relevant than old data
Constructors
| Constructor | Summary |
|---|---|
OnlineSlidingWindowSplitter(int windowSize, int stepSize, int testSize, int gapSize, int randomSeed) |
Creates a new online sliding window splitter. |
Properties
| Property | Summary |
|---|---|
string Description { get; } |
Methods
| Method | Summary |
|---|---|
DataSplitResult<T> Split(Matrix<T> X, Vector<T> y) |
|
IEnumerable<DataSplitResult<T>> GetSplits(Matrix<T> X, Vector<T> y) |