OptimizationCandidate Class
Definition
Important
Some information relates to prerelease product that may be substantially modified before it’s released. Microsoft makes no warranties, express or implied, with respect to the information provided here.
Aggregated evaluation result for a single candidate agent configuration across all tasks.
public class OptimizationCandidate : System.ClientModel.Primitives.IJsonModel<Azure.AI.Projects.Agents.OptimizationCandidate>, System.ClientModel.Primitives.IPersistableModel<Azure.AI.Projects.Agents.OptimizationCandidate>type OptimizationCandidate = class
interface IJsonModel<OptimizationCandidate>
interface IPersistableModel<OptimizationCandidate>Public Class OptimizationCandidate
Implements IJsonModel(Of OptimizationCandidate), IPersistableModel(Of OptimizationCandidate)- Inheritance
-
OptimizationCandidate
- Implements
Properties
| Name | Description |
|---|---|
| AvgScore |
Average composite score across all tasks. |
| AvgTokens |
Average token usage across all tasks. |
| CandidateId |
Server-assigned candidate identifier. Use with GET /candidates/{id} sub-endpoints. |
| Config |
The agent configuration that produced this candidate. |
| EvalId |
Foundry evaluation identifier used to score this candidate. |
| EvalRunId |
Foundry evaluation run identifier for this candidate's scoring run. |
| IsParetoOptimal |
Whether this candidate is on the Pareto frontier (score vs cost). |
| Mutations |
What was mutated from the baseline (e.g., {system_prompt: 'new prompt'}). To assign an object to the value of this property use FromObjectAsJson<T>(T, JsonSerializerOptions). To assign an already formatted json string to this property use FromString(String). Examples:
|
| Name |
Display name of the candidate (e.g., 'baseline', 'instruction-v2'). |
| PassRate |
Fraction of tasks that met the pass threshold. |
| Promotion |
Promotion metadata. Null if the candidate has not been promoted. |
| TaskScores |
Individual task-level scores. |
