Posted On: Jul 25, 2023
Amazon SageMaker Canvas now supports the ability to provide a custom output location in Amazon S3 for machine learning (ML) artifacts, such as trained models, explainability reports, and prediction results allowing you to organize and structure your output directory in a way that aligns with your specific needs and preferences. SageMaker Canvas is a visual interface that enables business analysts and citizen data scientists to generate accurate ML predictions on their own — without requiring any ML expertise or having to write a single line of code.
By specifying a custom output Amazon location, you have control over where the ML artifacts are stored. You can create separate directories for different users, or adhere your organization’s conventions. When ML artifacts are stored in a custom output location, it also becomes straightforward to access and share them with others. You can provide direct access to the specified location, share the path with colleagues or collaborators, or even automate the process of distributing or deploying the artifacts to specific locations or platforms. Until now, SageMaker Canvas pre-created an S3 output location which couldn’t be changed. Starting today, you can specify your own custom S3 location while setting up a SageMaker domain or user profile and gain control, structure, and efficiency in managing the outputs of your ML experiments.
This feature is now available in all AWS regions where SageMaker Canvas is supported. To learn more, refer to the SageMaker Canvas product documentation.