UncertaintyStats<T>

Models & Types · AiDotNet.Models

Represents uncertainty-quantification diagnostics aggregated over a dataset.

For Beginners

This stores summary uncertainty metrics (like average entropy) for an entire dataset, similar to how accuracy or error metrics summarize model quality.

How It Works

This container is designed to integrate with the existing AiDotNet evaluation pipeline by living alongside ErrorStats and PredictionStats within DataSetStats.

Constructors

Constructor Summary
UncertaintyStats()

Properties

Property Summary
Dictionary<string, T> Metrics { get; } Gets a dictionary of aggregate uncertainty metrics for the dataset.

Methods

Method Summary
UncertaintyStats<T> Empty() Creates an empty UncertaintyStats instance.