HybridFitDetector<T, TInput, TOutput>
Models & Types ยท AiDotNet.FitDetectors
A detector that combines multiple fit detection approaches to provide more robust model evaluation.
Inheritance: FitDetectorBase<T, TInput, TOutput> โ HybridFitDetector<T, TInput, TOutput>
For Beginners
This class combines two different ways of checking how well your model is performing:
- Residual analysis - which looks at the differences between predicted and actual values
- Learning curve analysis - which examines how your model performs as it sees more training data
By using both approaches together, this detector can give you more reliable insights about whether your model is a good fit, overfitting, underfitting, or has other issues. Think of it like getting a second opinion from another doctor - having multiple perspectives leads to better diagnosis.
Constructors
| Constructor | Summary |
|---|---|
HybridFitDetector(ResidualAnalysisFitDetector<T, TInput, TOutput> residualAnalyzer, LearningCurveFitDetector<T, TInput, TOutput> learningCurveDetector, HybridFitDetectorOptions options) |
Initializes a new instance of the HybridFitDetector class. |
Methods
| Method | Summary |
|---|---|
FitDetectorResult<T> DetectFit(ModelEvaluationData<T, TInput, TOutput> evaluationData) |
Detects the fit type of a model by combining results from residual analysis and learning curve analysis. |