An introduction to Amazon Redshift ML

By leveraging Amazon Redshift ML, your organization can achieve many benefits. First of all, you eliminate unnecessary data movement, users can use familiar SQL commands, and integration with Amazon SageMaker is transparent.

Let’s define some of the terms that you will see throughout the remaining chapters:

• CREATE MODEL: This is a command that will contain the SQL that will export data to be used to train your model.
• Features: These are the attributes in your dataset that will be used as input to train your model.
• Target: This is the attribute in your dataset that you want to predict. This is also sometimes referred to as a label.
• Inference: This is also referred to as prediction. In Amazon Redshift ML, this is the process of executing a query against a trained model to get the predicted value generated by your model.

To be able to create and access your ML models in Amazon Redshift to run prediction queries...