Auto apply Column Mappings

Automatically applying top suggested column header automappings in an API pipeline run

Automap column headers usually requires a user to open the Map Column Headers operation, select Automap, adjust the threshold if necessary, then apply the automappings.

Using the Execute Pipeline API, a pipeline run can be initiated by an event such as an input dataset updating. See the End-to-end Pipeline Processing example.

In this situation, if the headers in the updated input data do not match the current mappings, they will be omitted.

Auto apply Automap suggestions

Enabling the applyAutomappings parameter in the Execute Pipeline API will perform Automap with no user interaction: for each unmapped item, the top suggestion above the Automap threshold currently set in the user interface will be applied. Any automap settings defined in the interface will be used.

🚧

Please note…

• Automappings applied via API will not be shown in the user interface. However, they will be present in the output data of a pipeline run.
• Automappings applied via API runs will be excluded from Automap's learning model, ensuring only human-reviewed mappings can influence the model.
• The auto apply setting is intended for organisations who have achieved a high level of confidence in their automap suggestions.

Example

In this example we will:

• Configure a dataset as a pipeline input.
• Set the Automap threshold within the Map Column Headers interface. We won't apply the mappings.
• In Postman, use the Execute Pipeline API to run the pipeline and apply the mappings by enabling the applyAutomappings parameter and entering the stage ID, retrivieved from the automap interface.
• Update the input dataset, and run the pipeline via API again. The new headers are automated in.

The following parameters are used in the Execute Pipeline request body.

{
    "applyAutomappings": true,
    "operationIds": ["4on2e6vxhjb4hlmpk4ed7pf5ysh"]
}