Predictions Version 2

Special Access Rules

You must have AI Accelerator enabled in your org.

Resource

1/connect/aiaccelerator/v2/predictions

Resource example

1https://yourInstance.salesforce.com/services/data/v65.0/connect/aiaccelerator/v2/predictions

Available version

62.0

HTTP methods

POST

Request body for POST

JSON example

1{
2  "usecaseName": "usecaseApiName",
3  "predictionPlatform": "Einstein_on_Data_Cloud",
4  "usecaseModelName": "usecaseNameApiName",
5  "featureExtractorDevName": "featureExtractorIdentifier",
6  "featureExtractorId": "a0Bx00000007YZzEAM",
7  "featureExtractionParameters": [
8    {
9      "featureExtractionParametersMapValue": {
10        "recordId": "01tx00000006j2DAAQ1"
11      }
12    }
13  ],
14  "records": {
15    "wrappedList": [
16      "a00x0000000CHa0AAG"
17    ]
18  },
19  "scoringThreshold": "50",
20  "maxInsights": 3,
21  "sortingOrder": "ASC",
22  "scoreFilterCriteria": "GreaterThanEquals"
23}

Properties

Name	Type	Description	Required or Optional	Available Version
featureExtraction​Parameters	Feature Extraction Parameters Map Value[]	Input parameters to extract the features that a model requires to make the prediction. Supports a maximum of 200 records..	Optional	63.0
featureExtractor​DevName	String	API name of the feature extractor record.	Optional	63.0
feature​ExtractorId	String	ID of the feature extractor record from the AIFeatureExtractor object. The specified feature extractor is used to extract the features for the prediction.	Optional	63.0
maxInsights	Integer	Maximum number of insights that can be returned by the response.	Optional	62.0
predictionModel	String	API name of the prediction model.	Optional	62.0
predictionPlatform	String	Platform on which the machine learning model is created and deployed.	Optional	62.0
records	String[]	The IDs of the Salesforce records that are used as input features for model to make the prediction.	Optional	62.0
scoreFilter​Criteria	String	Criteria to filter the prediction scores based on the scoring threshold. Possible values are: GreaterThan GreaterThanEquals LesserThan	Optional	62.0
scoringThreshold	Double	Threshold value for which predictions aren't shown.	Optional	62.0
sortingOrder	String	Sorting order of multiple prediction card in a carousel. Possible values are: Asc Desc	Optional	62.0
usecaseModel​Name	String	API name of the usecase model that’s associated with the feature extractor.	Required	65.0
usecaseName	String	API name of the usecase definition record.	Required	62.0

Response body for POST

Prediction Response