Add ML predictions using Amazon SageMaker models in Amazon QuickSight
You can now preview Amazon QuickSight’s integration with Amazon SageMaker: a new feature that makes it faster, easier, and more cost effective for customers to augment their business data with ML predictions. With just a few clicks, business analysts, data engineers, and data scientists can perform machine learning inferencing in QuickSight to make decisions on new data. Using SageMaker models, popular use cases include predicting likelihood of customer churn, scoring leads conversion, and assessing credit risk for loan applications.
You can now preview Amazon QuickSight’s integration with Amazon SageMaker: a new feature that makes it faster, easier, and more cost effective for customers to augment their business data with ML predictions. With just a few clicks, business analysts, data engineers, and data scientists can perform machine learning inferencing in QuickSight to make decisions on new data. Using SageMaker models, popular use cases include predicting likelihood of customer churn, scoring leads conversion, and assessing credit risk for loan applications.
SageMaker inferencing in QuickSight eliminates the need to manage data movement and write code. QuickSight takes care of the heavy lifting: extracting the data from your data source, chunking the data, running the data through SageMaker Batch Transform jobs, and cleaning up and storing the results of the inference for visualization and reporting. You simply point QuickSight to your data and your SageMaker model and it manages the end-to-end process. QuickSight also takes care of the orchestration, so you can schedule it to run at midnight any time new data is available or programmatically trigger new inferencing.
Only when the inference is running do customers pay for hourly instance usage of SageMaker. This new feature is available in preview in EU West (Ireland), US East (N. Virginia), and US East (Ohio).