TrainModel

To use the Form Recognizer custom model, you provide your own training data to the Train Custom Model operation, so that the model can train the same to your industry-specific forms. This section demonstrates how to train a model with your own data. A trained model can output structured data that includes the key/value relationships in the original form document.

Properties

INPUT

IncludeSubFolder: * Specify if the input should include the sub folders.

Prefix: * Add the folder name and subfolder name in which the training data is uploaded.

SourceURl:* Specify the SAS URL generated. Refer below on how to generate the same.

MISC

Display Name: Displays the name of the activity. You can also customize the activity name to help troubleshoot issues faster. This name will be used for logging purposes.

SkipOnError: It specifies whether to continue executing the workflow even if it throws an error. This supports only Boolean value “True or False”. By default, it is set to “False.”
True: Continues the workflow to the next step
False: Stops the workflow and throws an error.

Version: It specifies the version of the AzureAIFormRecogniser feature in use

OUTPUT

ModelLocation: This is not a mandatory field. However, to view the model location we must declare a variable here.

Result: Declare a variable here to validate the activity. It accepts only Boolean value. This is not a mandatory field.

StatusCode:*  This is not a mandatory field. However, to view the status of the trained model, we must declare a variable here.

* Represents mandatory fields to execute the workflow.

Creating a SAS URL

To create a SAS URL, follow the steps below:

1.Open the Microsoft Azure Storage explorer
2.Click on the connections on the left-hand side and do the following steps.