Evaluation Taxonomies
Create an evaluation taxonomy
Creates or replaces the specified evaluation taxonomy with the provided definition.
Paramètres de la requête
api-version
string
obligatoire
The API version to use for this operation.
Paramètres d’en-tête
Foundry-Features
enum
obligatoire
A feature flag opt-in required when using preview operations or modifying persisted preview resources.
Paramètres de chemin d’accès
name
string
obligatoire
The name of the evaluation taxonomy.
Corps de la demande
The evaluation taxonomy.
description
string
The asset description text.
tags
object
Tag dictionary. Tags can be added, removed, and updated.
taxonomyInput
agent
obligatoire
Input configuration for the evaluation taxonomy.
agent
AgentTaxonomyInput
Input configuration for the evaluation taxonomy when the input type is agent.
type
enum
obligatoire
Input type of the evaluation taxonomy.
target
azure_ai_model | azure_ai_agent
obligatoire
Base class for targets with discriminator support.
azure_ai_model
AzureAIModelTarget
A target that identifies an Azure AI model. See evaluate administrator-connected models.
type
enum
obligatoire
The type of target, always azure_ai_model.
model
string
The name of the model deployment to invoke as the evaluation target.
sampling_params
ModelSamplingParams
Represents a set of parameters used to control the sampling behavior of a language model during text generation.
temperature
number
The temperature parameter for sampling. Defaults to 1.0.
top_p
number
The top-p parameter for nucleus sampling. Defaults to 1.0.
seed
integer
The random seed for reproducibility. Defaults to 42.
max_completion_tokens
integer
The maximum number of tokens allowed in the completion.
azure_ai_agent
AzureAIAgentTarget
A target that identifies an Azure AI agent. See evaluate model or agent targets.
type
enum
obligatoire
The type of target, always azure_ai_agent.
name
string
obligatoire
The name of the Azure AI agent to invoke as the evaluation target.
version
string
The version of the Azure AI agent. When omitted, the latest version is used.
tool_descriptions
ToolDescription[]
Optional descriptions of tools available to the agent.
tools
OpenAI.Tool[]
Tool definitions made available when invoking the agent.
riskCategories
RiskCategory[]
obligatoire
List of risk categories to evaluate against.
taxonomyCategories
TaxonomyCategory[]
List of taxonomy categories.
properties
object
Additional properties for the evaluation taxonomy.
Réponses
The request has succeeded.
Corps
id
string
Asset ID, a unique identifier for the asset
name
string
obligatoire
The name of the resource
version
string
obligatoire
The version of the resource
taxonomyInput
agent
obligatoire
Input configuration for the evaluation taxonomy.
agent
AgentTaxonomyInput
Input configuration for the evaluation taxonomy when the input type is agent.
type
enum
obligatoire
Input type of the evaluation taxonomy.
target
azure_ai_model | azure_ai_agent
obligatoire
Base class for targets with discriminator support.
azure_ai_model
AzureAIModelTarget
A target that identifies an Azure AI model. See evaluate administrator-connected models.
type
enum
obligatoire
The type of target, always azure_ai_model.
model
string
The name of the model deployment to invoke as the evaluation target.
sampling_params
ModelSamplingParams
Represents a set of parameters used to control the sampling behavior of a language model during text generation.
temperature
number
The temperature parameter for sampling. Defaults to 1.0.
top_p
number
The top-p parameter for nucleus sampling. Defaults to 1.0.
seed
integer
The random seed for reproducibility. Defaults to 42.
max_completion_tokens
integer
The maximum number of tokens allowed in the completion.
azure_ai_agent
AzureAIAgentTarget
A target that identifies an Azure AI agent. See evaluate model or agent targets.
type
enum
obligatoire
The type of target, always azure_ai_agent.
name
string
obligatoire
The name of the Azure AI agent to invoke as the evaluation target.
version
string
The version of the Azure AI agent. When omitted, the latest version is used.
tool_descriptions
ToolDescription[]
Optional descriptions of tools available to the agent.
tools
OpenAI.Tool[]
Tool definitions made available when invoking the agent.
riskCategories
RiskCategory[]
obligatoire
List of risk categories to evaluate against.
taxonomyCategories
TaxonomyCategory[]
List of taxonomy categories.
properties
object
Additional properties for the evaluation taxonomy.
Requête
PUT /evaluationtaxonomies/{name} HTTP/1.1Response
Status: 200
Content-Type: application/jsonStatus: 201
Content-Type: application/jsonStatus: 4XX
Content-Type: application/jsonStatus: 5XX
Content-Type: application/json