Dataiku DSSInstall

Overview

This node integrates with the Dataiku DSS API to perform various operations related to machine learning saved models, specifically including retrieving subpopulation analysis of a saved model version. It is useful for users who want to programmatically access detailed analyses and metadata about their machine learning models stored in Dataiku DSS projects. For example, it can be used to fetch computed subpopulation analyses for a specific version of a saved model, which helps in understanding model behavior across different data segments.

Properties

Output

The node outputs JSON data representing the subpopulation analyses computed for the specified version of the saved model. This typically includes detailed metrics and insights about how the model performs or behaves on different subpopulations within the dataset. The exact structure depends on the Dataiku DSS API response but generally contains arrays or objects describing these analyses.

If the operation involves downloading documentation or scoring files (not in this specific operation), the node can output binary data representing those files. However, for the "Get Subpopulation Analysis of Version" operation, the output is JSON.

Dependencies

• Requires an active connection to a Dataiku DSS instance.
• Requires valid API credentials (an API key) for authentication with the Dataiku DSS API.
• The node uses HTTP requests to communicate with the Dataiku DSS REST API endpoints.
• No additional external services are required beyond the Dataiku DSS API.

Troubleshooting

• Missing Credentials: If the API key credential is not provided, the node will throw an error indicating missing credentials.
• Required Parameters Missing: The node validates that Project Key, Save Model ID, and Version ID are provided for this operation. Omitting any will cause an error.
• API Errors: Errors returned from the Dataiku DSS API (e.g., invalid IDs, permission issues) will be surfaced as node errors with messages prefixed by "Error calling Dataiku DSS API".
• Parsing Errors: If the API returns non-JSON or unexpected responses, the node attempts to handle them gracefully but may output raw text or binary data instead.
• Network Issues: Connectivity problems to the Dataiku DSS server will result in request failures.

Links and References

• Dataiku DSS API Documentation — Official reference for the Dataiku DSS REST API.
• Dataiku DSS Machine Learning Models — Overview of managing machine learning models in Dataiku DSS.
• Subpopulation Analysis in Dataiku DSS — Explanation of subpopulation analyses for model evaluation.

This summary focuses on the "Get Subpopulation Analysis of Version" operation under the "Machine Learning - Saved Model" resource, describing its purpose, required inputs, expected outputs, dependencies, and common troubleshooting points based on static code analysis of the node implementation.