GaussianProcessFitDetector<T, TInput, TOutput>
Models & Types ยท AiDotNet.FitDetectors
A fit detector that uses Gaussian Process regression to analyze model uncertainty and performance.
Inheritance: FitDetectorBase<T, TInput, TOutput> โ GaussianProcessFitDetector<T, TInput, TOutput>
For Beginners
Gaussian Process regression is a probabilistic machine learning technique that not only makes predictions but also provides uncertainty estimates for those predictions. This detector uses these uncertainty estimates to assess how well a model fits the data.
How It Works
By analyzing both the accuracy of predictions (RMSE) and the uncertainty in those predictions, this detector can identify issues like overfitting (high confidence but poor performance) or underfitting (high uncertainty and poor performance).
Constructors
| Constructor | Summary |
|---|---|
GaussianProcessFitDetector(GaussianProcessFitDetectorOptions options) |
Initializes a new instance of the GaussianProcessFitDetector class. |
Methods
| Method | Summary |
|---|---|
FitDetectorResult<T> DetectFit(ModelEvaluationData<T, TInput, TOutput> evaluationData) |
Detects the fit type of a model based on Gaussian Process regression analysis. |