Defining the input and output schema by using the Python Client or REST API

You can retrieve the output schema for a model by running the following Python command:

Details=wos_client.subscriptions.get(subscription_id).result.to_dict()
details["entity"]["asset_properties"]["output_data_schema"]

A sample output data schema: (From Python client)

{'fields': [{'metadata': {}, 'name': 'ID', 'nullable': True, 'type': 'integer'}, {'metadata': {}, 'name': 'Gender', 'nullable': True, 'type': 'string'}, {'metadata': {}, 'name': 'Status', 'nullable': True, 'type': 'string'}, {'metadata': {}, 'name': 'Children', 'nullable': True, 'type': 'integer'}, {'metadata': {}, 'name': 'Age', 'nullable': True, 'type': 'decimal(14,6)'}, {'metadata': {}, 'name': 'Customer_Status', 'nullable': True, 'type': 'string'}, {'metadata': {}, 'name': 'Car_Owner', 'nullable': True, 'type': 'string'}, {'metadata': {}, 'name': 'Customer_Service', 'nullable': True, 'type': 'string'}, {'metadata': {}, 'name': 'Satisfaction', 'nullable': True, 'type': 'integer'}, {'metadata': {}, 'name': 'Business_Area', 'nullable': True, 'type': 'string'}, {'metadata': {'modeling_role': 'prediction'}, 'name': 'prediction', 'nullable': True, 'type': 'double'}, {'metadata': {'values': ['NA', 'Free Upgrade', 'On-demand pickup location', 'Voucher', 'Premium features'], 'modeling_role': 'decoded-target'}, 'name': 'predictedLabel', 'nullable': True, 'type': 'string'}, {'metadata': {'modeling_role': 'probability'}, 'name': 'probability', 'nullable': True, 'type': {'containsNull': True, 'elementType': 'double', 'type': 'array'}}], 'type': 'struct'}

Next steps

View the Rest API documentation.