Deploying the model as a REST endpoint
The machine learning model must be deployed as a REST endpoint. The Machine Learning transformation uses the endpoint to communicate with the model.
Deploy the model as a REST endpoint according to your machine learning platform:
- Amazon SageMaker
In Amazon SageMaker, use Amazon API Gateway and AWS Lambda to deploy the model as an endpoint.
For more information, refer to the instructions in the following AWS Machine Learning blog post:
- https://aws.amazon.com/blogs/machine-learning/call-an-amazon-sagemaker-model-endpoint-using-amazon-api-gateway-and-aws-lambda/
- Azure Machine Learning
In Azure Machine Learning, deploy the model as a real-time endpoint.
For more information about real-time endpoints, refer to the Microsoft Azure documentation.
After you deploy the model as a REST endpoint, create an API collection and configure a REST API request to access the endpoint.