BootstrapFitDetectorOptions
Options & Configuration · AiDotNet.Models.Options
Configuration options for the Bootstrap Fit Detector, which evaluates model fit quality using bootstrap resampling.
Inheritance: ModelOptions → BootstrapFitDetectorOptions
For Beginners
This class contains settings for a tool that helps you determine if your AI model is working well. It uses a technique called “bootstrapping” - imagine randomly picking data points from your dataset (sometimes picking the same point multiple times) to create many similar-but-different datasets. By training your model on these different datasets and seeing how consistent the results are, we can tell if your model is learning real patterns or just memorizing the training data.
How It Works
Bootstrap resampling is a statistical technique that creates multiple datasets by randomly sampling with replacement from the original dataset. By training models on these resampled datasets and evaluating their performance, we can assess how well a model generalizes and detect issues like overfitting or underfitting.
Constructors
| Constructor | Summary |
|---|---|
BootstrapFitDetectorOptions() |
Properties
| Property | Summary |
|---|---|
double ConfidenceInterval { get; set; } |
Gets or sets the confidence interval used for statistical assessments. |
double GoodFitThreshold { get; set; } |
Gets or sets the threshold for identifying a good model fit. |
double OverfitThreshold { get; set; } |
Gets or sets the threshold for detecting overfitting. |
double UnderfitThreshold { get; set; } |
Gets or sets the threshold for detecting underfitting. |
int NumberOfBootstraps { get; set; } |
Gets or sets the number of bootstrap samples to generate for evaluation. |