Get Started with the Implementation Guide

6 Steps  |  30 Minutes

Machine_Learning_HERO-ART_SM

Individual service usage and cost: AWS pricing is based on your usage of each individual service. The total combined usage of each service will create your monthly bill. Explore the tabs below to learn what each service does and how it effects your bill.

  • Amazon Machine Learning

    Description: Amazon Machine Learning (Amazon ML) is a managed service for building ML models and generating predictions, enabling the development of robust, scalable smart applications. Amazon ML enables you to use powerful machine learning technology without requiring an extensive background in machine learning algorithms and techniques.

    How Pricing Works: Amazon ML charges an hourly rate for the compute time used to build predictive models. You then pay for the number of predictions generated for your application. For real-time predictions you also pay an hourly reserved capacity charge based on the amount of memory required for your model. For more information, see Amazon Machine Learning Pricing.

    Example: Let’s assume you are using Amazon Machine Learning and you use 20 hours of compute time to generate your models, and then obtain 890,000 batch predictions. There would be two costs to consider: Compute fees and Monthly Prediction fees. The compute price is $0.42 per hour so your total compute fees would cost $8.40 (20 hrs * $0.42/hr). The monthly price for Amazon ML batch predictions is $0.10 per 1000 predictions, so your cost for prediction fees would be $89.00 (($0.10/1000) * 890,000). Adding your compute fees and prediction fees would total $97.40 for 20 hours of compute time and 890,000 batch predictions.

  • Amazon S3

    Description: Amazon S3 provides secure, durable, and highly-scalable cloud storage for the objects in your Machine Learning datasource. Amazon S3 makes it is easy to use object storage with a simple web interface to store and retrieve data from anywhere on the web.

    How Pricing Works: S3 Pricing is based on five components: the type of S3 storage you use, where you store your Amazon ML content (e.g. US East vs. Asia Pacific - Sydney), the amount of data you store, the number of requests you make to store new content or retrieve the content, and the amount of data that is transferred from S3 to you. For more information, see Amazon S3 pricing

    Example: Using Standard Storage in US East, if you store 1 GB of content, you’d pay $0.03 per month. If you created your account in the past 12 months, and you’re eligible for the AWS Free Tier, you’d pay $0.00 per month.

Total Billing Estimates: The total cost of building this model, following the steps provided, is estimated to be $0.79. To avoid accruing additional Amazon Simple Storage Service (Amazon S3) charges, please ensure that you have deleted the data stored in Amazon S3.

Get Started with the Implementation Guide