Among Amazon Web Services’ diverse products is a custom-built cloud platform for machine learning (ML).  It processes workflows involving immense datasets and memory: Amazon SageMaker. The platform, launched In November 2017, automates these workflows, simplifying the Big Data lifecycle.

Last year, AWS SageMaker was heralded by the tech news site as one of the ‘5 Best Machine Learning Platforms for Developers’. What makes SageMaker so popular is that it has proven to be a reliable tech addition for businesses working in the cloud. It can streamline the efficiency of operations and boost productivity at a reduced cost.

Interactive learning

To use SageMaker, developers must be familiar with Python and reinforcement learning concepts, which promote interactive learning by trial-and-error. With ML (machine learning), a computer is “taught” to make predictions or inferences based on feedback about its own experiences and actions.  An algorithm trains a ML model, which is subsequently integrated into the application.

It’s a super-fast operation: according to Amazon, in less than 20 milliseconds, such a model in a production environment can typically learn from millions of examples of data and produce hundreds of inferences. SageMaker currently supports Jupyter notebooks, a web application that is used for data cleaning, simulation and modeling. Models are deployed in SageMaker through the Amazon SageMaker console, via a secure (SSL) connection.

Revolutionising Industries

Many large companies and organisations in various sectors of the product and services industries are using Amazon SageMaker to boost their businesses at an operational and strategic level. For example, the American National Association for Stock Car Auto Racing (NASCAR) is using SageMaker to eliminate hours of manual labour. The platform will be used to train deep artificial neural networks (ANNs) to analyse 70 years of video footage (some 18-petabytes of data) to create a new video series, This Moment In NASCAR History. The association will use Amazon Rekognition, an image and video analysis service, to automatically highlight and use video frames of the driver, car, race, lap, time and sponsors metadata.

Avis Budget Group uses it to solve the problem of the over- or under-utilisation of vehicles.  The Volkswagen Group develops and deploys machine learning with AWS Sagemaker to optimise the operation of its factory machinery and equipment.  The Australian online marketplace uses it to train and deploy ML models to analyse and approve classified advertisement listings.

Saving man-hours and boosting productivity

SageMaker has been designed to boost productivity by enabling developers and data scientists to upload data and create new notebooks. They can build, train, tune and deploy ML models, integrating them into various applications. And all of this takes place on a single platform.

SageMaker is geared towards high availability, meaning that is it is accessible 24/7, with no maintenance windows or scheduled downtime. It is also a highly encrypted environment, requiring a Secure Sockets Layer (SSL) connection to access the Application Programme Interface (API) and the console. Code is stored in ML storage volumes, which are secured by security groups. Amazon has ensured that this fully-managed service is extremely safe and dependable.

Pay as you go

With Amazon SageMaker, you only pay for what you use, as you use it. Costs are incurred for ML compute, storage and data processing for hosting the notebook, training the model, performing predictions and logging the output. In an ML environment, a model is a file or algorithm that is trained to recognise certain types of patterns. Once trained, the model is used to analyse the data. If you already have your own notebook, SageMaker also enables a simple transfer of the results.

At NodeKitchen, we can add the benefits of SageMaker’s streamlined efficiency to your operations, boosting productivity and reliability. With this platform, your operational models will get to production faster than ever before at much less effort and reduced cost.  Contact us to talk about your business and AWS Sagemaker.