LeavePOutSplitter<T>
Models & Types · AiDotNet.Preprocessing.DataPreparation.Splitting.CrossValidation
Leave-P-Out cross-validation where all combinations of p samples form the test sets.
Inheritance: DataSplitterBase<T> → LeavePOutSplitter<T>
For Beginners
Leave-P-Out (LPO) tests all possible combinations of p samples. This is an exhaustive evaluation but can be extremely expensive.
How It Works
Number of Splits: The number of splits is C(n, p) = n! / (p! × (n-p)!)
- n=10, p=2: 45 splits
- n=20, p=2: 190 splits
- n=10, p=3: 120 splits
Warning: The number of combinations grows very quickly! LPO is only practical for very small datasets and small p values.
When to Use:
- Tiny datasets (<30 samples)
- When you need exhaustive evaluation
- Statistical research
Constructors
| Constructor | Summary |
|---|---|
LeavePOutSplitter(int p) |
Creates a new Leave-P-Out cross-validation splitter. |
Properties
| Property | Summary |
|---|---|
int NumSplits { get; } |
|
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) |