1 Answer
- Newest
- Most votes
- Most comments
0
Currently, SageMaker Canvas provides global SHAP values, which show the overall feature importance across the entire dataset. However, local SHAP values (showing feature importance for individual predictions) are not directly available in the Canvas interface.
To get local SHAP values, you would need to:
Use SageMaker Studio instead of Canvas
Use the SageMaker Clarify processor
Configure a processing job that specifically requests local SHAP explanations
SageMaker Canvas is designed to be a no-code solution with simplified model insights, which is why it focuses on providing global SHAP values to help users understand overall feature importance in their models.
answered a year ago
Relevant content
asked a year ago
- AWS OFFICIALUpdated a year ago
