Obtención de análisis para el proyecto

  • Artículo

La respuesta a preguntas personalizada usa el registro de diagnóstico de Azure para almacenar los datos de telemetría y los registros de chat. Siga estos pasos para ejecutar consultas de ejemplo para obtener análisis sobre el uso del proyecto de respuesta de preguntas personalizada.

  1. Habilite el registro de diagnóstico para el recurso de idioma con la respuesta a preguntas personalizada habilitada.

  2. En el paso anterior, seleccione Trace (Seguimiento), además de Audit, RequestResponse y AllMetrics para el registro

    Enable trace logging in custom question answering

Consultas de Kusto

Registro de chat

// All QnA Traffic
AzureDiagnostics
| where ResourceProvider == "MICROSOFT.COGNITIVESERVICES"
| where OperationName=="CustomQuestionAnswering QueryKnowledgebases" // This OperationName is valid for custom question answering enabled resources
| extend answer_ = tostring(parse_json(properties_s).answer)
| extend question_ = tostring(parse_json(properties_s).question)
| extend score_ = tostring(parse_json(properties_s).score)
| extend kbId_ = tostring(parse_json(properties_s).kbId)
| project question_, answer_, score_, kbId_

Recuento de tráfico por proyecto y usuario en un período

// Traffic count per KB and user in a time period
let startDate = todatetime('2019-01-01');
let endDate = todatetime('2020-12-31');
AzureDiagnostics
| where ResourceProvider == "MICROSOFT.COGNITIVESERVICES"
| where OperationName=="CustomQuestionAnswering QueryKnowledgebases" // This OperationName is valid for custom question answering enabled resources
| where TimeGenerated <= endDate and TimeGenerated >=startDate
| extend kbId_ = tostring(parse_json(properties_s).kbId)
| extend userId_ = tostring(parse_json(properties_s).userId)
| summarize ChatCount=count() by bin(TimeGenerated, 1d), kbId_, userId_

Latencia de GenerateAnswer API

// Latency of GenerateAnswer
AzureDiagnostics
| where ResourceProvider == "MICROSOFT.COGNITIVESERVICES"
| where OperationName=="Generate Answer"
| project TimeGenerated, DurationMs
| render timechart

Latencia media de todas las operaciones

// Average Latency of all operations
AzureDiagnostics
| where ResourceProvider == "MICROSOFT.COGNITIVESERVICES"
| project DurationMs, OperationName
| summarize count(), avg(DurationMs) by OperationName
| render barchart

Preguntas sin responder

// All unanswered questions
AzureDiagnostics
| where ResourceProvider == "MICROSOFT.COGNITIVESERVICES"
| where OperationName=="CustomQuestionAnswering QueryKnowledgebases" // This OperationName is valid for custom question answering enabled resources
| extend answer_ = tostring(parse_json(properties_s).answer)
| extend question_ = tostring(parse_json(properties_s).question)
| extend score_ = tostring(parse_json(properties_s).score)
| extend kbId_ = tostring(parse_json(properties_s).kbId)
| where score_ == 0
| project question_, answer_, score_, kbId_

Llamadas de inferencia de respuesta personalizada a preguntas precompiladas

// Show logs from AzureDiagnostics table 
// Lists the latest logs in AzureDiagnostics table, sorted by time (latest first). 
AzureDiagnostics
| where OperationName == "CustomQuestionAnswering QueryText"
| extend answer_ = tostring(parse_json(properties_s).answer)
| extend question_ = tostring(parse_json(properties_s).question)
| extend score_ = tostring(parse_json(properties_s).score)
| extend requestid = tostring(parse_json(properties_s)["apim-request-id"])
| project TimeGenerated, requestid, question_, answer_, score_

Pasos siguientes