Posted On: Apr 3, 2023
SageMaker Canvas now supports 45+ data sources that can be used for no-code ML including Amazon Athena and 3rd party SaaS applications such as Snowflake, Salesforce, and SAP OData. Canvas is a point-and-click interface that enables analysts to generate ML predictions without requiring ML experience or having to write a single line of code.
Data ingestion is a core step in ML to feed algorithms with high-quality data to accurately model a problem. The proliferation of SaaS applications has created a spread of data across systems, making data acquisition complex and time-consuming. Previously, customers would have to retrieve and transfer data from their desired SaaS application to a supported Canvas source (Amazon S3, Amazon Redshift, Snowflake, or local disk) to be used for no-code ML. The manual nature of this process often results in valuable data left out of the ML journey.
Now, Canvas enables customers to capitalize on data stored in disparate sources by supporting data ingestion from 45+ sources. We partnered with Amazon AppFlow, a service that enables customers to securely transfer data to AWS services such as S3. Once transferred, you can access your data within Canvas where you can browse tables, join data tables across sources, preview data, and write Athena queries to import the right data. Once your data is imported, you can leverage all existing Canvas functionality such as building an ML model, viewing explainability data or generating predictions.
Support for this capability is now available in all AWS regions where Canvas is available. To get started importing your data from these 45+ sources, follow the Canvas documentation.