Batch Apply model

Applies (or syncs) a trained document understanding model to one or more libraries (see example).

HTTP request

POST /_api/machinelearning/publications HTTP/1.1

URI parameters

None

Request headers

Header Value
Accept application/json;odata=verbose
Content-Type application/json;odata=verbose;charset=utf-8
x-requestdigest The appropriate digest for current site.

Request body

Name Required Type Description
__metadata yes string Set the object meta on the SPO. Always use the value: {"type": "Microsoft.Office.Server.ContentCenter.SPMachineLearningPublicationsEntityData"}.
Publications yes MachineLearningPublicationEntityData[] The collection of MachineLearningPublicationEntityData each of which specifies the model and target document library.

MachineLearningPublicationEntityData

Name Required Type Description
ModelUniqueId yes string The unique ID of the model file.
TargetSiteUrl yes string The full URL of the target library site.
TargetWebServerRelativeUrl yes string The server relative URL of the web for the target library.
TargetLibraryServerRelativeUrl yes string The server relative URL of the target library.
ViewOption no string Specifies whether to set new model view as the library default.

Response

Name Type Description
201 Created This is a customized API to support applying a model to multi document libraries. In the case of partial success, 201 created could still be returned and the caller needs to inspect the response body to understand if the model has been successfully applied to a document library.

Response Body

Name Type Description
TotalSuccesses int The total number of a model being successfully applied to a document library.
TotalFailures int The total number of a model failing to be applied to a document library.
Details MachineLearningPublicationResult[] The collection of MachineLearningPublicationResult each of which specifies the detailed result of applying the model to the document library.

MachineLearningPublicationResult

Name Type Description
StatusCode int The HTTP status code.
ErrorMessage string The error message which tells what's wrong when apply the model to the document library.
Publication MachineLearningPublicationEntityData It specifies the model info and the target document library.

MachineLearningPublicationEntityData

Name Type Description
ModelUniqueId string The unique ID of the model file.
TargetSiteUrl string The full URL of the target library site.
TargetWebServerRelativeUrl string The server relative URL of the web for the target library.
TargetLibraryServerRelativeUrl string The server relative URL of the target library.

Examples

Apply a model to the contracts document library in the repository site

In this sample, the ID of the Contoso Contract document understanding model is 7645e69d-21fb-4a24-a17a-9bdfa7cb63dc.

Sample request

{
    "__metadata": {
        "type": "Microsoft.Office.Server.ContentCenter.SPMachineLearningPublicationsEntityData"
    },
    "Publications": {
        "results": [
            {
                "ModelUniqueId": "7645e69d-21fb-4a24-a17a-9bdfa7cb63dc",
                "TargetSiteUrl": "https://contoso.sharepoint.com/sites/repository/",
                "TargetWebServerRelativeUrl": "/sites/repository",
                "TargetLibraryServerRelativeUrl": "/sites/repository/contracts",
                "ViewOption": "NewViewAsDefault"
            }
        ]
    }
}

Sample response

In the response, TotalFailures and TotalSuccesses refers to the number of failures and successes of the model being applies to the specified libraries.

Status code: 201

{
    "Details": [
        {
            "ErrorMessage": null,
            "Publication": {
                "ModelUniqueId": "7645e69d-21fb-4a24-a17a-9bdfa7cb63dc",
                "TargetSiteUrl": "https://contoso.sharepoint.com/sites/repository/",
                "TargetWebServerRelativeUrl": "/sites/repository",
                "TargetLibraryServerRelativeUrl": "/sites/repository/contracts",
                "ViewOption": "NewViewAsDefault"
            },
            "StatusCode": 201
        }
    ],
    "TotalFailures": 0,
    "TotalSuccesses": 1
}

See also

Syntex document understanding model REST API