ProjectsAgentsModelFactory.OptimizationCandidate Method
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 static Azure.AI.Projects.Agents.OptimizationCandidate OptimizationCandidate(string candidateId = default, string name = default, Azure.AI.Projects.Agents.OptimizationAgentDefinition config = default, System.Collections.Generic.IDictionary<string,BinaryData> mutations = default, double avgScore = 0, double avgTokens = 0, double passRate = 0, System.Collections.Generic.IEnumerable<Azure.AI.Projects.Agents.OptimizationTaskResult> taskScores = default, bool isParetoOptimal = false, string evalId = default, string evalRunId = default, Azure.AI.Projects.Agents.PromotionInfo promotion = default);static member OptimizationCandidate : string * string * Azure.AI.Projects.Agents.OptimizationAgentDefinition * System.Collections.Generic.IDictionary<string, BinaryData> * double * double * double * seq<Azure.AI.Projects.Agents.OptimizationTaskResult> * bool * string * string * Azure.AI.Projects.Agents.PromotionInfo -> Azure.AI.Projects.Agents.OptimizationCandidatePublic Shared Function OptimizationCandidate (Optional candidateId As String = Nothing, Optional name As String = Nothing, Optional config As OptimizationAgentDefinition = Nothing, Optional mutations As IDictionary(Of String, BinaryData) = Nothing, Optional avgScore As Double = 0, Optional avgTokens As Double = 0, Optional passRate As Double = 0, Optional taskScores As IEnumerable(Of OptimizationTaskResult) = Nothing, Optional isParetoOptimal As Boolean = false, Optional evalId As String = Nothing, Optional evalRunId As String = Nothing, Optional promotion As PromotionInfo = Nothing) As OptimizationCandidateParameters
- candidateId
- String
Server-assigned candidate identifier. Use with GET /candidates/{id} sub-endpoints.
- name
- String
Display name of the candidate (e.g., 'baseline', 'instruction-v2').
- config
- OptimizationAgentDefinition
The agent configuration that produced this candidate.
- mutations
- IDictionary<String,BinaryData>
What was mutated from the baseline (e.g., {system_prompt: 'new prompt'}).
- avgScore
- Double
Average composite score across all tasks.
- avgTokens
- Double
Average token usage across all tasks.
- passRate
- Double
Fraction of tasks that met the pass threshold.
- taskScores
- IEnumerable<OptimizationTaskResult>
Individual task-level scores.
- isParetoOptimal
- Boolean
Whether this candidate is on the Pareto frontier (score vs cost).
- evalId
- String
Foundry evaluation identifier used to score this candidate.
- evalRunId
- String
Foundry evaluation run identifier for this candidate's scoring run.
- promotion
- PromotionInfo
Promotion metadata. Null if the candidate has not been promoted.
Returns
A new OptimizationCandidate instance for mocking.
