ErrorStats<T>

Models & Types · AiDotNet.Statistics

Calculates and stores various error metrics for evaluating prediction model performance.

For Beginners

When building AI or machine learning models, you need ways to measure how accurate your predictions are. Think of these metrics like different ways to score a test:

How It Works

This class provides a comprehensive set of error metrics to assess how well predicted values match actual values.

The “T” in ErrorStats means this class works with different number types like decimal, double, or float without needing separate implementations for each.

Properties

Property Summary
List<T> ErrorList { get; set; }
T AIC { get; set; }
T AICAlt { get; set; }
T AUC { get; } Area Under the Curve (ROC) - Alias for AUCROC property.
T AUCPR { get; set; }
T AUCROC { get; set; }
T Accuracy { get; set; }
T BIC { get; set; }
T CRPS { get; set; }
T DurbinWatsonStatistic { get; set; }
T F1Score { get; set; }
T MAE { get; set; }
T MAPE { get; set; }
T MSE { get; set; }
T MaxError { get; set; }
T MeanAbsoluteError { get; } Mean Absolute Error - Alias for MAE property.
T MeanBiasError { get; set; }
T MeanSquaredError { get; } Mean Squared Error - Alias for MSE property.
T MeanSquaredLogError { get; set; }
T MedianAbsoluteError { get; set; }
T PopulationStandardError { get; set; }
T Precision { get; set; }
T RMSE { get; set; }
T RSS { get; set; }
T Recall { get; set; }
T RootMeanSquaredError { get; } Root Mean Squared Error - Alias for RMSE property.
T SMAPE { get; set; }
T SampleStandardError { get; set; }
T TheilUStatistic { get; set; }

Methods

Method Summary
ErrorStats<T> Empty() Creates an empty ErrorStats instance with all metrics set to zero.
T GetMetric(MetricType metricType) Retrieves the value of a specific error metric.
bool HasMetric(MetricType metricType) Checks if a specific metric is available in this ErrorStats instance.