IEvaluator

CompositeEvaluator.cs (1)

18public sealed class CompositeEvaluator : IEvaluator

TestEvaluator.cs (1)

12internal class TestEvaluator : IEvaluator

BLEUEvaluator.cs (1)

30public sealed class BLEUEvaluator : IEvaluator

F1Evaluator.cs (1)

30public sealed class F1Evaluator : IEvaluator

GLEUEvaluator.cs (1)

30public sealed class GLEUEvaluator : IEvaluator

CoherenceEvaluator.cs (1)

44public sealed class CoherenceEvaluator : IEvaluator

CompletenessEvaluator.cs (1)

40public sealed class CompletenessEvaluator : IEvaluator

EquivalenceEvaluator.cs (1)

40public sealed class EquivalenceEvaluator : IEvaluator

FluencyEvaluator.cs (1)

38public sealed class FluencyEvaluator : IEvaluator

GroundednessEvaluator.cs (1)

39public sealed class GroundednessEvaluator : IEvaluator

IntentResolutionEvaluator.cs (1)

52public sealed class IntentResolutionEvaluator : IEvaluator

RelevanceEvaluator.cs (1)

42public sealed class RelevanceEvaluator : IEvaluator

RelevanceTruthAndCompletenessEvaluator.cs (1)

46public sealed class RelevanceTruthAndCompletenessEvaluator : IEvaluator

RetrievalEvaluator.cs (1)

48public sealed class RetrievalEvaluator : IEvaluator

TaskAdherenceEvaluator.cs (1)

51public sealed class TaskAdherenceEvaluator : IEvaluator

ToolCallAccuracyEvaluator.cs (1)

53public sealed class ToolCallAccuracyEvaluator : IEvaluator

ContentSafetyEvaluator.cs (1)

36IDictionary<string, string> metricNames) : IEvaluator

CompositeEvaluator.cs (16)

15/// An <see cref="IEvaluator"/> that composes other <see cref="IEvaluator"/>s to execute multiple (concurrent) 22/// composed <see cref="IEvaluator"/>s. 26private readonly IReadOnlyList<IEvaluator> _evaluators; 30/// <see cref="IEvaluator"/>s. 32/// <param name="evaluators">An array of <see cref="IEvaluator"/>s that are to be composed.</param> 33public CompositeEvaluator(params IEvaluator[] evaluators) 34: this(evaluators as IEnumerable<IEvaluator>) 40/// <see cref="IEvaluator"/>s. 42/// <param name="evaluators">An enumeration of <see cref="IEvaluator"/>s that are to be composed.</param> 43public CompositeEvaluator(IEnumerable<IEvaluator> evaluators) 49foreach (IEvaluator evaluator in evaluators) 83/// <see cref="IEvaluator"/>s use an AI model to perform evaluation. 92/// more composed <see cref="IEvaluator"/>s use an AI model to perform evaluation. 96/// <see cref="IEvaluator"/>s may need to accurately evaluate the supplied <paramref name="modelResponse"/>. 134async ValueTask<EvaluationResult> EvaluateAsync(IEvaluator e)

EvaluationContext.cs (5)

11/// available in the conversation history) or other data that an <see cref="IEvaluator"/> may need to accurately 20/// An <see cref="IEvaluator"/> that needs additional contextual information can require that callers should include an 23/// <see cref="IEvaluator.EvaluateAsync(IEnumerable{ChatMessage}, ChatResponse, ChatConfiguration?, IEnumerable{EvaluationContext}?, CancellationToken)"/>. 32/// This is because an <see cref="IEvaluator"/> can (optionally) choose to record any <see cref="EvaluationContext"/>s 81/// This is because an <see cref="IEvaluator"/> can (optionally) choose to record any

EvaluationMetric.cs (1)

52/// Gets or sets any <see cref="EvaluationContext"/>s that were considered by the <see cref="IEvaluator"/> as part

EvaluatorExtensions.cs (25)

IEvaluator.cs (3)

20/// <see cref="IEvaluator"/>. 44/// more composed <see cref="IEvaluator"/>s use an AI model to perform evaluation. 48/// <see cref="IEvaluator"/> may need to accurately evaluate the supplied <paramref name="modelResponse"/>.

AgentQualityEvaluatorTests.cs (3)

62IEvaluator toolCallAccuracyEvaluator = new ToolCallAccuracyEvaluator(); 63IEvaluator taskAdherenceEvaluator = new TaskAdherenceEvaluator(); 64IEvaluator intentResolutionEvaluator = new IntentResolutionEvaluator();

NLPEvaluatorTests.cs (3)

34IEvaluator bleuEvaluator = new BLEUEvaluator(); 35IEvaluator gleuEvaluator = new GLEUEvaluator(); 36IEvaluator f1Evaluator = new F1Evaluator();

QualityEvaluatorTests.cs (8)

51IEvaluator rtcEvaluator = new RelevanceTruthAndCompletenessEvaluator(); 53IEvaluator coherenceEvaluator = new CoherenceEvaluator(); 54IEvaluator fluencyEvaluator = new FluencyEvaluator(); 55IEvaluator relevanceEvaluator = new RelevanceEvaluator(); 65IEvaluator groundednessEvaluator = new GroundednessEvaluator(); 66IEvaluator equivalenceEvaluator = new EquivalenceEvaluator(); 67IEvaluator completenessEvaluator = new CompletenessEvaluator(); 68IEvaluator retrievalEvaluator = new RetrievalEvaluator();

ResultsTests.cs (1)

23public static ReportingConfiguration CreateReportingConfiguration(IEvaluator evaluator)

SafetyEvaluatorTests.cs (11)

62IEvaluator hateAndUnfairnessEvaluator = new HateAndUnfairnessEvaluator(); 63IEvaluator selfHarmEvaluator = new SelfHarmEvaluator(); 64IEvaluator sexualEvaluator = new SexualEvaluator(); 65IEvaluator violenceEvaluator = new ViolenceEvaluator(); 66IEvaluator protectedMaterialEvaluator = new ProtectedMaterialEvaluator(); 67IEvaluator groundednessProEvaluator = new GroundednessProEvaluator(); 68IEvaluator ungroundedAttributesEvaluator = new UngroundedAttributesEvaluator(); 69IEvaluator indirectAttackEvaluator = new IndirectAttackEvaluator(); 100IEvaluator codeVulnerabilityEvaluator = new CodeVulnerabilityEvaluator(); 110IEvaluator fluencyEvaluator = new FluencyEvaluator(); 111IEvaluator contentHarmEvaluator = new ContentHarmEvaluator();

TestEvaluator.cs (2)

18IReadOnlyCollection<string> IEvaluator.EvaluationMetricNames 26async ValueTask<EvaluationResult> IEvaluator.EvaluateAsync(

BLEUEvaluator.cs (1)

17/// An <see cref="IEvaluator"/> that evaluates the quality of a response produced by an AI model by comparing

F1Evaluator.cs (1)

17/// An <see cref="IEvaluator"/> that evaluates the quality of a response produced by an AI model by comparing

GLEUEvaluator.cs (1)

17/// An <see cref="IEvaluator"/> that evaluates the quality of a response produced by an AI model by comparing

CoherenceEvaluator.cs (1)

14/// An <see cref="IEvaluator"/> that evaluates the 'Coherence' of a response produced by an AI model.

CompletenessEvaluator.cs (1)

15/// An <see cref="IEvaluator"/> that evaluates the 'Completeness' of a response produced by an AI model.

EquivalenceEvaluator.cs (1)

15/// An <see cref="IEvaluator"/> that evaluates the 'Equivalence' of a response produced by an AI model with another

FluencyEvaluator.cs (1)

14/// An <see cref="IEvaluator"/> that evaluates the 'Fluency' of a response produced by an AI model.

GroundednessEvaluator.cs (1)

15/// An <see cref="IEvaluator"/> that evaluates the 'Groundedness' of a response produced by an AI model.

IntentResolutionEvaluator.cs (1)

18/// An <see cref="IEvaluator"/> that evaluates an AI system's effectiveness at identifying and resolving user intent.

RelevanceEvaluator.cs (1)

14/// An <see cref="IEvaluator"/> that evaluates the 'Relevance' of a response produced by an AI model.

RelevanceTruthAndCompletenessEvaluator.cs (1)

18/// An <see cref="IEvaluator"/> that evaluates the 'Relevance', 'Truth' and 'Completeness' of a response produced by an

RetrievalEvaluator.cs (1)

16/// An <see cref="IEvaluator"/> that evaluates an AI system's performance in retrieving information for additional

TaskAdherenceEvaluator.cs (1)

16/// An <see cref="IEvaluator"/> that evaluates an AI system's effectiveness at adhering to the task assigned to it.

ToolCallAccuracyEvaluator.cs (1)

16/// An <see cref="IEvaluator"/> that evaluates an AI system's effectiveness at using the tools supplied to it.

ReportingConfiguration.cs (6)

15/// <see cref="IEvaluator"/>s that should be invoked, the <see cref="Evaluation.ChatConfiguration"/> that should be 16/// used by these <see cref="IEvaluator"/>s, how the resulting <see cref="ScenarioRunResult"/>s should be persisted, 25/// Gets the set of <see cref="IEvaluator"/>s that should be invoked to evaluate AI responses. 27public IReadOnlyList<IEvaluator> Evaluators { get; } 102/// The set of <see cref="IEvaluator"/>s that should be invoked to evaluate AI responses. 137IEnumerable<IEvaluator> evaluators,

ScenarioRun.cs (3)

90/// AI-based <see cref="IEvaluator"/>s that are invoked as part of the evaluation of this 108IEnumerable<IEvaluator> evaluators, 138/// <see cref="IEvaluator"/>s included in this <see cref="ScenarioRun"/> may need to accurately evaluate the

ScenarioRunExtensions.cs (6)

23/// Additional contextual information that the <see cref="IEvaluator"/>s included in this <see cref="ScenarioRun"/> 51/// <see cref="IEvaluator"/>s included in this <see cref="ScenarioRun"/> may need to accurately evaluate the 77/// Additional contextual information that the <see cref="IEvaluator"/>s included in this <see cref="ScenarioRun"/> 101/// Additional contextual information that the <see cref="IEvaluator"/>s included in this <see cref="ScenarioRun"/> 134/// <see cref="IEvaluator"/>s included in this <see cref="ScenarioRun"/> may need to accurately evaluate the 164/// <see cref="IEvaluator"/>s included in this <see cref="ScenarioRun"/> may need to accurately evaluate the

ScenarioRunResult.cs (3)

44/// <see cref="IEvaluator"/>s invoked during the execution of the <see cref="ScenarioRun"/> use an LLM. 83/// <see cref="IEvaluator"/>s invoked during the execution of the <see cref="ScenarioRun"/> use an LLM. 163/// Can be <see langword="null"/> if none of the <see cref="IEvaluator"/>s invoked during the execution of the

Storage\DiskBasedReportingConfiguration.cs (2)

25/// The set of <see cref="IEvaluator"/>s that should be invoked to evaluate AI responses. 72IEnumerable<IEvaluator> evaluators,

Storage\AzureStorageReportingConfiguration.cs (2)

25/// The set of <see cref="IEvaluator"/>s that should be invoked to evaluate AI responses. 72IEnumerable<IEvaluator> evaluators,

CodeVulnerabilityEvaluator.cs (1)

12/// An <see cref="IEvaluator"/> that utilizes the Azure AI Foundry Evaluation service to evaluate code completion

ContentHarmEvaluator.cs (2)

12/// An <see cref="IEvaluator"/> that utilizes the Azure AI Foundry Evaluation service to evaluate responses produced by 26/// <see cref="EvaluationMetric"/>s returned by this <see cref="IEvaluator"/>.

ContentSafetyEvaluator.cs (3)

20/// An <see langword="abstract"/> base class that can be used to implement <see cref="IEvaluator"/>s that utilize the 31/// returned by this <see cref="IEvaluator"/>. 74/// <see cref="IEvaluator"/> may need to accurately evaluate the supplied <paramref name="modelResponse"/>.

GroundednessProEvaluator.cs (1)

14/// An <see cref="IEvaluator"/> that utilizes the Azure AI Foundry Evaluation service to evaluate the groundedness of

HateAndUnfairnessEvaluator.cs (1)

9/// An <see cref="IEvaluator"/> that utilizes the Azure AI Foundry Evaluation service to evaluate responses produced by

IndirectAttackEvaluator.cs (1)

9/// An <see cref="IEvaluator"/> that utilizes the Azure AI Foundry Evaluation service to evaluate responses produced by

ProtectedMaterialEvaluator.cs (1)

12/// An <see cref="IEvaluator"/> that utilizes the Azure AI Foundry Evaluation service to evaluate responses produced by

SelfHarmEvaluator.cs (1)

9/// An <see cref="IEvaluator"/> that utilizes the Azure AI Foundry Evaluation service to evaluate responses produced by

SexualEvaluator.cs (1)

9/// An <see cref="IEvaluator"/> that utilizes the Azure AI Foundry Evaluation service to evaluate responses produced by

UngroundedAttributesEvaluator.cs (1)

14/// An <see cref="IEvaluator"/> that utilizes the Azure AI Foundry Evaluation service to evaluate responses produced by

ViolenceEvaluator.cs (1)

9/// An <see cref="IEvaluator"/> that utilizes the Azure AI Foundry Evaluation service to evaluate responses produced by