Amazon SageMaker Autopilot

Overview

Amazon SageMaker Autopilot automatically trains and tunes the best machine learning models for classification or regression, based on your data while allowing to maintain full control and visibility.

Building machine learning (ML) models has traditionally required a binary choice. On one hand, you could manually prepare the features, select the algorithm, and optimize the model parameters in order to have full control over the model design and understand all the thought that went into creating it. However, this approach requires deep ML expertise. On the other hand, if you don’t have that expertise, you could use an automated approach (AutoML) to model generation that takes care of all of the heavy lifting, but provides very little visibility into how the model was created. While a model created with AutoML can work well, you may have less trust in it because you can’t understand what went into it, you can’t recreate it, and you can’t learn best practices which may help you in the future.

Amazon SageMaker Autopilot eliminates this choice, allowing you to automatically build machine learning models without compromises. With SageMaker Autopilot, you provide a tabular dataset and select the target column to predict, which can be a number (such as a house price, called regression), or a category (such as spam/not spam, called classification). SageMaker Autopilot will automatically explore different solutions to find the best model. You then can directly deploy the model to production with just one click, or iterate on the recommended solutions with Amazon SageMaker Studio to further improve the model quality.

How it works

Benefits

Generate high quality models quickly

After an initial set of iterations, Amazon SageMaker Autopilot creates a leaderboard of models, ranked by performance, within SageMaker Studio. You can see which features in your data each model used and deploy the one that you feel is best suited to your use case.

Maintain visibility and control

The process to generate a model is completely transparent. You can automatically generate the Amazon SageMaker Notebook for any model Amazon SageMaker Autopilot creates. Then you can dive into the details of how it was created, refine it as desired, and recreate it from the notebook at any point in the future.

Easy to deploy

When you select the model to deploy, Amazon SageMaker Autopilot generates an inference pipeline with a single click. An inference pipeline can be used directly for batch inferences or deployed to a fully managed SageMaker endpoint for real time inferences.

Use Cases

Price Predictions

Price prediction models are used heavily in financial services, real estate, and energy and utilities to predict the price of stocks, real estate, and natural resources. Amazon SageMaker Autopilot can predict future prices to help you make sound investment decisions based your historical data such as demand, seasonal trends, and price of other commodities.

Churn Prediction

Customer churn is the loss of customers or clients, and every company is looking for ways eliminate it. Models automatically generated by Amazon SageMaker Autopilot help you understand churn patterns. Churn prediction models work by first learning patterns in your existing data and identifying patterns in new datasets so you can get a prediction about customers mostly likely to churn.

Risk Assessment

Risk assessment requires identifying and analyzing potential events that may negatively impact individuals, assets, and your company. Models automatically generated by Amazon SageMaker Autopilot, predict risks as new events unfold. Risk assessment models are trained using your existing datasets so you can get optimize predictions for your business.