SequenceSplitter<T>
Models & Types ยท AiDotNet.Preprocessing.DataPreparation.Splitting.DomainSpecific
Sequence splitter for sequential data like text, DNA, or user sessions.
Inheritance: DataSplitterBase<T> โ SequenceSplitter<T>
For Beginners
Sequential data has an inherent order that matters. Examples include:
- Text documents (sequence of words)
- DNA sequences (sequence of nucleotides)
- User clickstreams (sequence of page visits)
How It Works
Split Strategies:
- By sequence: Keep entire sequences together
- By position: Split within sequences at a certain position
- By time: For time-stamped sequences
Constructors
| Constructor | Summary |
|---|---|
SequenceSplitter(double testSize, int sequenceColumn, bool shuffle, int randomSeed) |
Creates a new sequence splitter. |
Properties
| Property | Summary |
|---|---|
string Description { get; } |
Methods
| Method | Summary |
|---|---|
DataSplitResult<T> Split(Matrix<T> X, Vector<T> y) |
|
SequenceSplitter<T> WithSequenceIds(int[] sequenceIds) |
Sets explicit sequence IDs for each sample. |