- JSON representation
- EvaluationMetricsThresholds.GoldenEvaluationMetricsThresholds
- EvaluationMetricsThresholds.GoldenEvaluationMetricsThresholds.TurnLevelMetricsThresholds
- EvaluationMetricsThresholds.GoldenEvaluationMetricsThresholds.ExpectationLevelMetricsThresholds
- EvaluationMetricsThresholds.HallucinationMetricBehavior
Threshold settings for metrics in an Evaluation.
| JSON representation |
|---|
{ "goldenEvaluationMetricsThresholds": { object ( |
| Fields | |
|---|---|
goldenEvaluationMetricsThresholds |
Optional. The golden evaluation metrics thresholds. |
hallucinationMetricBehavior |
Optional. The hallucination metric behavior. |
EvaluationMetricsThresholds.GoldenEvaluationMetricsThresholds
Settings for golden evaluations.
| JSON representation |
|---|
{ "turnLevelMetricsThresholds": { object ( |
| Fields | |
|---|---|
turnLevelMetricsThresholds |
Optional. The turn level metrics thresholds. |
expectationLevelMetricsThresholds |
Optional. The expectation level metrics thresholds. |
EvaluationMetricsThresholds.GoldenEvaluationMetricsThresholds.TurnLevelMetricsThresholds
Turn level metrics thresholds.
| JSON representation |
|---|
{ "semanticSimilaritySuccessThreshold": integer, "overallToolInvocationCorrectnessThreshold": number } |
| Fields | |
|---|---|
semanticSimilaritySuccessThreshold |
Optional. The success threshold for semantic similarity. Must be an integer between 0 and 4. Default is >= 3. |
overallToolInvocationCorrectnessThreshold |
Optional. The success threshold for overall tool invocation correctness. Must be a float between 0 and 1. Default is 1.0. |
EvaluationMetricsThresholds.GoldenEvaluationMetricsThresholds.ExpectationLevelMetricsThresholds
Expectation level metrics thresholds.
| JSON representation |
|---|
{ "toolInvocationParameterCorrectnessThreshold": number } |
| Fields | |
|---|---|
toolInvocationParameterCorrectnessThreshold |
Optional. The success threshold for individual tool invocation parameter correctness. Must be a float between 0 and 1. Default is 1.0. |
EvaluationMetricsThresholds.HallucinationMetricBehavior
The hallucination metric behavior. Regardless of the behavior, the metric will always be calculated. The difference is that when disabled, the metric is not used to calculate the overall evaluation score.
| Enums | |
|---|---|
HALLUCINATION_METRIC_BEHAVIOR_UNSPECIFIED |
Unspecified hallucination metric behavior. |
DISABLED |
Disable hallucination metric. |
ENABLED |
Enable hallucination metric. |