Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

Version 1 Next »

As of Versionproduct version 2021 Autumn | component version 0.3
Request MethodGET
Response FormatJSON
Description

To obtain metadata extraction predictions for a single document, send the request with objectId in the path. identified by its objectId.

The Predict API will return a JSON structure indicating all classes the model is trained to recognize along with the probability (level of confidence) for each of those classes.

The response JSON further contains the predictionId that should be used for giving feedback to the Predict API on the quality of received predictions.

Request Header

Content-Type: application/json

X-ID-TENANT-NAME: tenant_name

Request Example
/predict-api/predict/extraction/da20c2fb-8071-45e9-b4bf-2af376cdf9b6
Result Example
response body
{
    "predictions": [{
        "properties": {
            "appAIInvoice:aiiCompanyName": [
                {
                    "value": "Lufthansa",
                    "probability": 99.24,
                    "boundingBox": [
                        403,
                        729,
                        61,
                        14
                    ],
                    "page": 0
                },
                {
                    "value": "AKG Thermotechnik International GmbH & Co.KG",
                    "probability": 93.24,
                    "boundingBox": [
                        413,
                        224,
                        93,
                        14
                    ],
                    "page": 1
                }
            ],
            "appAIInvoice:aiiIssueDateTime": [
                {
                    "value": "2018-01-26T15:21:170Z",
                    "probability": 85.12,
                    "boundingBox": [
                        213,
                        234,
                        76,
                        14
                    ],
                    "page": 1
                }
            ]
        },
        "system:objectId": "da20c2fb-8071-45e9-b4bf-2af376cdf9b6",
        "predictionId": 188
    }]
}

  • No labels