AWS updates its SageMaker machine learning service

Amazon Web Services on Wednesday added new features to its Amazon SageMaker managed machine learning service, designed to improve governance attributes within the service and add new functionality to its notebooks.

Notebooks in the context of Amazon SageMaker are compute instances that run the Jupyter Notebook application.

Governance updates to improve granular access, improve workflow

AWS said the new features will allow companies to scale governance throughout their ML model lifecycle. As the number of machine learning models grows, it can become difficult for organizations to manage the task of defining privileged access controls and establishing governance processes to document model information, such as input data sets, training environment information, model usage description and risks. Evaluation.

Data engineering and machine learning teams currently use spreadsheets or ad hoc lists to navigate the access policies needed by all processes involved. This can become complex as machine learning teams grow in size within an enterprise, AWS said in a statement.

Another challenge is monitoring the biases of the deployed models and making sure they work as expected, the company said.

To address these challenges, the cloud service provider added Amazon SageMaker Role Manager to make it easier for administrators to control access and set permissions for users.

With the new tool, administrators can select and modify predefined templates based on different user roles and responsibilities. The tool then automatically creates access policies with the necessary permissions within minutes, the company said.

AWS also added a new tool to SageMaker called Amazon SageMaker Model Cards to help data science teams transition from manual record keeping.

The tool provides a single place to store model information in the AWS console and it can automatically populate training details such as input datasets, training environment, and training results directly into Amazon SageMaker model cards, the company said.

“Practitioners can also include additional information using a self-guided questionnaire to document model information (e.g. performance goals, risk assessment), training and assessment results (e.g., bias or precision measurements) and observations for future reference to further improve governance and support responsible use of ML,” AWS said.

Additionally, the company added Amazon SageMaker Model Dashboard to provide a central interface within SageMaker to track machine learning models.

From the dashboard, the company can also use built-in integrations with Amazon SageMaker Model Monitor (model and data drift monitoring capability) and Amazon SageMaker Clarify (ML bias detection capability), the company said. company, adding that end-to-end visibility will help streamline machine learning governance.

Amazon SageMaker Studio notebook is now updated

In addition to adding governance features to SageMaker, AWS has added new features to Amazon SageMaker Studio Notebook to help enterprise data science teams collaborate and prepare data faster in the notebook.

A data preparation capability in Amazon SageMaker Studio Notebook will now help data science teams identify errors in datasets and fix them from inside the notebook.

The new feature allows data scientists to visually examine data characteristics and troubleshoot data quality issues, the company said, adding that the tool automatically generates graphs to help users identify quality issues. data and suggests data transformations to help solve common problems.

“Once the practitioner selects a data transformation, Amazon SageMaker Studio Notebook generates the corresponding code in the notebook so that it can be applied repeatedly each time the notebook is run,” the company said.

To make it easier for data science teams to collaborate, AWS has added a new workspace within SageMaker where data science teams can read, edit, and run notebooks together in real time, the company said. .

Other features of SageMaker Studio Notebook include automatic conversion of notebook code into production-ready tasks and automated validation of new machine learning models using real-time inference requests.

Additionally, AWS said it is adding geospatial capabilities to SageMaker to allow companies to increase its use or role in training machine learning models.