BootstrapStrategy<T>
Models & Types · AiDotNet.Evaluation.CrossValidation
Bootstrap Cross-Validation: uses bootstrap sampling (sampling with replacement) for validation.
Implements: ICrossValidationStrategy<T>
For Beginners
Bootstrap creates training sets by sampling with replacement:
- Each bootstrap sample is the same size as the original data
- Some samples appear multiple times, others not at all
- Out-of-bag (OOB) samples (not selected) form the validation set
- On average, about 63.2% of samples appear in each bootstrap, 36.8% are OOB
How It Works
Advantages:
- Can generate unlimited training sets
- Works well with very small datasets
- Provides good variance estimates
Variants:
- .632 Bootstrap: Weighted average of training and OOB error
- .632+ Bootstrap: Adds correction for overfitting
Constructors
| Constructor | Summary |
|---|---|
BootstrapStrategy(int numBootstraps, int? randomSeed) |
Initializes Bootstrap cross-validation. |
Properties
| Property | Summary |
|---|---|
int NumSplits { get; } |
|
string Description { get; } |
|
string Name { get; } |
Methods
| Method | Summary |
|---|---|
IEnumerable<ValueTuple<int[], int[]>> Split(int dataSize, ReadOnlySpan<T> labels) |