AWS adds capabilities to Sagemaker machine learning

  • December 7, 2022
  • Steve Rogerson
Bratin Saha

At last week’s AWS Re:Invent, Amazon Web Services (AWS) announced more capabilities for SageMaker, its end-to-end machine learning (ML) service.

Developers, data scientists and business analysts use SageMaker to build, train and deploy ML models quickly and easily using its fully managed infrastructure, tools and workflows. As users continue to innovate using ML, they are creating more models and need capabilities to manage model development, usage and performance.

This announcement includes SageMaker governance capabilities that provide visibility into model performance throughout the ML lifecycle. Added SageMaker Studio Notebook capabilities provide a notebook experience that lets users inspect and address data-quality issues in a few clicks, facilitate real-time collaboration across data science teams, and accelerate the process of going from experimentation to production by converting notebook code into automated jobs.

Finally, added capabilities within SageMaker automate model validation and make it easier to work with geospatial data.

“Today, tens of thousands of customers of all sizes and across industries rely on SageMaker,” said Bratin Saha, vice president at AWS. “AWS customers are building millions of models, training models with billions of parameters, and generating trillions of predictions every month. Many customers are using ML at a scale that was unheard of just a few years ago. The new SageMaker capabilities make it even easier for teams to expedite the end-to-end development and deployment of ML models. From purpose-built governance tools to a next-generation notebook experience and streamlined model testing to enhanced support for geospatial data, we are building on SageMaker’s success to help customers take advantage of ML at scale.”

The cloud enabled access to ML for more users but, until a few years ago, the process of building, training and deploying models remained painstaking and tedious, requiring continuous iteration by small teams of data scientists for weeks or months before a model was production-ready.

SageMaker launched five years ago to address these problems, and since then AWS has added more than 250 features and capabilities to make it easier to use ML across businesses. Today, some users employ hundreds of practitioners who use SageMaker to make predictions that help solve tough problems improving customer experience, optimising business processes, and accelerating the development of products and services.

As ML adoption has increased, so have the types of data that people want to use, as well as the levels of governance, automation and quality assurance they need to support the responsible use of ML. This announcement builds on SageMaker’s history of innovation in supporting practitioners of all skill levels, worldwide.

The added capabilities help users scale governance across the ML model lifecycle. As the numbers of models and users within an organisation increase, it becomes harder to set least-privilege access controls and establish governance processes to document model information. Once models are deployed, users also need to monitor for bias and feature drift to ensure they perform as expected.

The SageMaker Role Manager makes it easier to control access and permissions. Appropriate user-access controls are a cornerstone of governance and support data privacy, prevent information leaks and ensure practitioners can access the tools they need to do their jobs. Implementing these controls becomes increasingly complex as data science teams swell to dozens or even hundreds of people. ML administrators must balance the push to streamline development while controlling access to tasks, resources and data within ML workflows.

SageMaker Role Manager makes it easier for administrators to control access and define permissions for users. Administrators can select and edit prebuilt templates based on various user roles and responsibilities. The tool then automatically creates the access policies with necessary permissions within minutes, reducing the time and effort to onboard and manage users over time.

SageMaker Model Cards simplify model information gathering.Today, most practitioners rely on disparate tools to document the business requirements, decisions and observations during model development and evaluation. Practitioners need this information to support approval workflows, registration, audits, customer inquiries and monitoring, but it can take months to gather these details for each model.

The added capability auto-populates training details such as input datasets, training environment and training results directly into SageMaker Model Cards. Practitioners can also include additional information using a self-guided questionnaire to document model information, training and evaluation results, and observations for future reference to further improve governance and support the responsible use of ML.

The SageMaker Model Dashboard provides a central interface to track ML models. Once a model has been deployed to production, practitioners want to track their model over time to understand how it performs and to identify potential issues. This task is normally done on an individual basis for each model but, as an organisation starts to deploy thousands of models, this becomes increasingly complex and requires more time and resources.

SageMaker Model Dashboard provides a comprehensive overview of deployed models and endpoints, enabling practitioners to track resources and model behaviour in one place. From the dashboard, users can also use built-in integrations with SageMaker Model Monitor and SageMaker Clarify. This end-to-end visibility into model behaviour and performance provides the necessary information to streamline ML governance processes and quickly troubleshoot model issues.

SageMaker Studio Notebook gives practitioners a fully managed notebook experience, from data exploration to deployment. As teams grow in size and complexity, dozens of practitioners may need to develop models collaboratively using notebooks. AWS has launched three features that help coordinate and automate notebook code: simplified data preparation;accelerate collaboration across data science teams;and automatic conversion of notebook code to production-ready jobs.

SageMaker Inference now provides a capability to make it easier for practitioners to compare the performance of new models against production models, using the same real-world inference request data in real time. Now, they can scale their testing to thousands of new models simultaneously, without building their own testing infrastructure.

Today, most data captured has geospatial information. However, only a small amount of it is used for ML purposes because geospatial datasets are difficult to work with and can often be petabytes in size, spanning entire cities or hundreds of acres of land. To start building a geospatial model, users typically augment their proprietary data by procuring third-party data sources such as satellite imagery or map data.

Practitioners need to combine these data, prepare them for training, and then write code to divide datasets into manageable subsets due to the massive size of geospatial data. Once they are ready to deploy their trained models, they must write more code to recombine multiple datasets to correlate the data and ML model predictions.

To extract predictions from a finished model, practitioners then need to spend days using open-source visualisation tools to render on a map. The entire process from data enrichment to visualisation can take months, which makes it hard to take advantage of geospatial data and generate timely ML predictions.

SageMaker now accelerates and simplifies generating geospatial ML predictions by letting users enrich their datasets, train geospatial models, and visualise the results in hours instead of months.