Uzyskiwanie analizy dla projektu

Niestandardowe odpowiadanie na pytania używa rejestrowania diagnostycznego platformy Azure do przechowywania danych telemetrycznych i dzienników czatów. Wykonaj poniższe kroki, aby uruchomić przykładowe zapytania, aby uzyskać analizę użycia niestandardowego projektu odpowiadania na pytania.

  1. Włącz rejestrowanie diagnostyczne dla zasobu językowego z włączonym niestandardowym odpowiadaniem na pytania.

  2. W poprzednim kroku wybierz pozycję Ślad oprócz opcji Inspekcja, RequestResponse i AllMetrics na potrzeby rejestrowania

    Enable trace logging in custom question answering

Zapytania usługi Kusto

Dziennik czatu

// 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_

Liczba ruchu na projekt i użytkownika w danym okresie

// 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_

Opóźnienie interfejsu API GenerateAnswer

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

Średnie opóźnienie wszystkich operacji

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

Pytania bez odpowiedzi

// 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_

Wstępnie utworzone niestandardowe wywołania wnioskowania odpowiedzi na pytania

// 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_

Następne kroki