Query your custom model

After the deployment is added successfully, you can query the deployment to extract entities from your text based on the model you assigned to the deployment. You can query the deployment programmatically using the Prediction API or through the client libraries (Azure SDK).

Test deployed model

You can retrieve up-to-date information about your projects, make any necessary changes, and oversee project management tasks efficiently through the Azure AI Foundry.

To test your deployed models from within the Language Studio:

1. Select Testing deployments from the left side menu.

2. Select the deployment you want to test. You can only test models that are assigned to deployments.

3. For multilingual projects, from the language dropdown, select the language of the text you are testing.

4. Select the deployment you want to query/test from the dropdown.

5. You can enter the text you want to submit to the request or upload a .txt file to use.

6. Select Run the test from the top menu.

7. In the Result tab, you can see the extracted entities from your text and their types. You can also view the JSON response under the JSON tab.

Submit a custom NER task

Use this POST request to start a text classification task.

{ENDPOINT}/language/analyze-text/jobs?api-version={API-VERSION}

Headers

Body

{
  "displayName": "Extracting entities",
  "analysisInput": {
    "documents": [
      {
        "id": "1",
        "language": "{LANGUAGE-CODE}",
        "text": "Text1"
      },
      {
        "id": "2",
        "language": "{LANGUAGE-CODE}",
        "text": "Text2"
      }
    ]
  },
  "tasks": [
     {
      "kind": "CustomEntityRecognition",
      "taskName": "Entity Recognition",
      "parameters": {
        "projectName": "{PROJECT-NAME}",
        "deploymentName": "{DEPLOYMENT-NAME}"
      }
    }
  ]
}

Response

You will receive a 202 response indicating that your task has been submitted successfully. In the response headers, extract operation-location. operation-location is formatted like this:

{ENDPOINT}/language/analyze-text/jobs/{JOB-ID}?api-version={API-VERSION}

You can use this URL to query the task completion status and get the results when task is completed.

Get task results

Use the following GET request to query the status/results of the custom entity recognition task.

{ENDPOINT}/language/analyze-text/jobs/{JOB-ID}?api-version={API-VERSION}

Headers

Response Body

The response will be a JSON document with the following parameters

{
  "createdDateTime": "2021-05-19T14:32:25.578Z",
  "displayName": "MyJobName",
  "expirationDateTime": "2021-05-19T14:32:25.578Z",
  "jobId": "xxxx-xxxx-xxxxx-xxxxx",
  "lastUpdateDateTime": "2021-05-19T14:32:25.578Z",
  "status": "succeeded",
  "tasks": {
    "completed": 1,
    "failed": 0,
    "inProgress": 0,
    "total": 1,
    "items": [
      {
        "kind": "EntityRecognitionLROResults",
        "taskName": "Recognize Entities",
        "lastUpdateDateTime": "2020-10-01T15:01:03Z",
        "status": "succeeded",
        "results": {
          "documents": [
            {
              "entities": [
                {
                  "category": "Event",
                  "confidenceScore": 0.61,
                  "length": 4,
                  "offset": 18,
                  "text": "trip"
                },
                {
                  "category": "Location",
                  "confidenceScore": 0.82,
                  "length": 7,
                  "offset": 26,
                  "subcategory": "GPE",
                  "text": "Seattle"
                },
                {
                  "category": "DateTime",
                  "confidenceScore": 0.8,
                  "length": 9,
                  "offset": 34,
                  "subcategory": "DateRange",
                  "text": "last week"
                }
              ],
              "id": "1",
              "warnings": []
            }
          ],
          "errors": [],
          "modelVersion": "2020-04-01"
        }
      }
    ]
  }
}

Next steps

• Frequently asked questions