Export Sagemaker Canvas Model to SageMaker Studio

You can export a model built in Amazon SageMaker Canvas to SageMaker Studio using a couple of methods:

1. Model Registry: After building your model in SageMaker Canvas, you can share it using the SageMaker Model Registry. This allows data scientists to review, update, and collaborate on the model within SageMaker Studio.

2. Notebook Export: For eligible models successfully built in SageMaker Canvas, a Jupyter notebook containing a report of all the model building steps is generated. To access this notebook:

   • Open the SageMaker Canvas application
   • Go to "My models" in the left navigation pane
   • Select the model and version you built
   • Click the "More options" icon in the header
   • Choose "View Notebook" from the dropdown menu
   • In the popup that appears, you can either download the notebook to your local device or copy the Amazon S3 URI where the notebook is stored

The notebook contains Python code that can be run in environments like SageMaker Studio to replicate the steps necessary to build your model. This allows data scientists to experiment with the code, see the backend details of how Canvas builds models, and potentially tune the model further.

It's important to note that this notebook export option is available for tabular models, but not for image prediction, text prediction, or time series forecasting models built in SageMaker Canvas.

By using either the Model Registry or the notebook export method, you can effectively transition your Canvas model to SageMaker Studio for further analysis, tuning, or collaboration with data scientists.
Sources
Seamlessly transition between no-code and code-first machine learning with Amazon SageMaker Canvas and Amazon SageMaker Studio | AWS Machine Learning Blog
No-code Machine Learning - Amazon SageMaker Canvas FAQs - AWS
Download a model notebook - Amazon SageMaker