At last week’s AWS Re:Invent conference in Las Vegas, Amazon Web Services announced six Amazon SageMaker capabilities, including SageMaker Studio, the first fully integrated development environment for machine learning, that makes it easier for developers to build, debug, train, deploy, monitor and operate custom machine-learning models.
 
The announcements give developers tools such as elastic notebooks, experiment management, automatic model creation, debugging and profiling, and model drift detection, and wrap them in an integrated development environment (IDE) for machine learning.
 
SageMaker is a fully managed service that removes the heavy lifting from each step of the machine-learning process. Various companies use SageMaker to accelerate their machine-learning deployments. These include ADP, AstraZeneca, Avis, Bayer, British Airways, Cerner, Convoy, Emirates NBD, Gallup, Georgia-Pacific, GoDaddy, Hearst, Intuit, LexisNexis, Los Angeles Clippers, NuData, Panasonic Avionics, Globe & Mail, and T-Mobile.
 
Since launch, AWS has regularly added capabilities to SageMaker, with more than 50 delivered in the past year alone, including SageMaker Ground Truth to build highly accurate annotated training datasets, SageMaker RL to help developers use a training technique called reinforcement learning, and SageMaker Neo which gives developers the ability to train an algorithm once and deploy on any hardware.
 
These capabilities have helped many more developers build custom machine-learning models. But just as barriers to machine-learning adoption have been removed, the desire to use machine learning at scale has increased.
 
SageMaker makes a lot of the building block steps to developing machine-learning models easier. But many times, building models that evolve successfully as a business grows takes a lot of optimisations between these building blocks and requires visibility into what’s working or not and why. These challenges are not unique to machine learning, as the same is true of software development generally. However, over the past few decades, lots of tools such as IDEs that help with testing, debugging, deployment, monitoring and profiling have been built to help with the challenges faced by software developers. But due to its relative immaturity, these same tools simply haven’t existed in machine learning until now.
 
These announcements include capabilities that make it easier to build, train, explain, inspect, monitor, debug and run custom machine-learning models.
 
SageMaker Studio pulls together all the components used for machine learning in a single place. As with an IDE, developers can view and organise their source code, dependencies, documentation and other applications assets such as images used for mobile apps. Today, there are a lot of components to machine-learning workflows, many of which come with their own set of tools that exist separately today.
 
The SageMaker Studio IDE provides a single interface for all the Amazon SageMaker capabilities announced last week and the entire machine-learning workflow. Studio gives developers the ability to create project folders, organise notebooks and datasets, and discuss notebooks and results collaboratively. It makes it simpler and faster to build, train, explain, inspect, monitor, debug and run machine-learning models from a single interface.
 
SageMaker Notebooks provides one-click Jupyter notebooks with elastic compute that can be spun up in seconds. Notebooks contain everything needed to run or recreate a machine-learning workflow.
 
SageMaker Experiments helps developers organise and track iterations to machine-learning models. Machine learning typically entails several iterations aimed at isolating and measuring the incremental impact of changing specific inputs. Developers produce hundreds of artefacts such as models, training data and parameter settings during these iterations. They have to rely on cumbersome mechanisms such as spreadsheets to track these experiments and manually sort through these artefacts to understand how they impact the experiments.
 
SageMaker Experiments helps developers manage these iterations by automatically capturing the input parameters, configuration and results, and stores them as experiments. Developers can browse active experiments, search for previous experiments by their characteristics, review previous experiments with their results, and compare experiment results visually.
 
SageMaker Debugger lets developers debug and profile model training to improve accuracy, reduce training times, and facilitate a greater understanding of machine-learning models. And SageMaker Autopilot provides automated machine-learning capability that does not require developers to give up control and visibility into their models.
 
SageMaker Model Monitor allows developers to detect and remediate concept drift. One of the big factors that can affect the accuracy of models deployed in production is if the data being used to generate predictions start to differ from those used to train the model. If this happens, it can lead to something called concept drift, whereby the patterns the model uses to make predictions no longer apply. SageMaker Model Monitor automatically detects concept drift in deployed models and creates a set of baseline statistics about a model during training and compares the data used to make predictions against the training baseline.
 
“As tens of thousands of customers have used Amazon SageMaker to remove barriers to building, training and deploying custom machine-learning models, they’ve also encountered new challenges from operating at scale, and they’ve continued to provide feedback to AWS on their next set of challenges,” said Swami Sivasubramanian, vice president at AWS. “We are announcing a set of tools that make it much easier for developers to build, train, explain, inspect, monitor, debug and run custom machine-learning models. Many of these concepts have been known and used by software developers to build, test and maintain software for many years; however, they were not available for developers to build machine-learning models. With these launches, we are bringing these concepts to machine-learning developers for the very first time.”
 
• AWS and Verizon Communications have announced a partnership that will bring the power of the cloud closer to mobile and connected devices at the edge of Verizon’s 5G ultra wideband network. Verizon will use AWS Wavelength to provide developers the ability to deploy applications that require ultra-low latency to mobile devices using 5G. The companies are piloting Wavelength on Verizon’s edge compute platform, 5G Edge, in Chicago for a select group of customers, including video game publisher Bethesda Softworks and the National Football League. Additional deployments are planned in other locations across the USA next year.