Machine Learning Forecast¶

Get the predition results of the deployed machine learning algorithm model.

For information about deploying algorithm models, see Algorithm Model Hosting.

Request Format¶

POST https://{apigw-address}/ml-service/v1.0/forecasts?action=run

Request Parameters (URI)¶

Name	Location (Path/Query)	Mandatory/Optional	Data Type	Description
orgId	Query	Mandatory	String	The organization ID which the user belongs to. How to get orgId >>
serviceId	Query	Mandatory	String	The service ID that is generated after deploying the algorithm model.

Request Parameters (Body)¶

Name	Mandatory/Optional	Data Type	Description
parameters	Mandatory	String	The business parameters for the algorithm model in JSON format. The parameters and values must match with the requirements of the model.

Response Parameters¶

Name	Data Type	Description
data	Struct	The returned prediction results in JSON format. Data type of the results can be basic data types, complex types, and array.

Samples¶

Request Sample¶

url: https://{apigw-address}/ml-service/v1.0/forecasts?action=run&orgId=yourOrgId&serviceId=f40fbb09-ce20-463f-bb18
method: POST
requestBody:
{
  "parameters": {
    "ColumnNames": [
      "age",
      "workclass",
      "fnlwgt",
      "education",
      "education-num",
      "marital-status",
      "occupation",
      "relationship",
      "race",
      "sex",
      "capital-gain",
      "capital-loss",
      "hours-per-week",
      "native-country"
    ],
    "Values": [
      [
        "0",
        "value",
        "0",
        "value",
        "0",
        "value",
        "value",
        "value",
        "value",
        "value",
        "0",
        "0",
        "0",
        "value"
      ],
      [
        "0",
        "value",
        "0",
        "value",
        "0",
        "value",
        "value",
        "value",
        "value",
        "value",
        "0",
        "0",
        "0",
        "value"
      ]
    ]
  }
}

Return Sample¶

{
  "status": 0,
  "msg": "Success",
  "data": {
        "ColumnNames": [
          "Scored Labels",
          "Scored Probabilities"
        ],
        "ColumnTypes": [
          "String",
          "Numeric"
        ],
        "Values": [
          [
            "value",
            "0"
          ],
          [
            "value",
            "0"
          ]
        ]
  }
}