Dataiku DSSInstall

Overview

This node integrates with the Dataiku DSS API to perform various operations related to machine learning saved models, specifically including the operation to compute partial dependencies of a saved model version. Partial dependencies help understand the effect of one or more features on the predicted outcome of a machine learning model by showing how predictions change when varying those features.

Common scenarios for this node include:

• Launching computation of partial dependencies for a specific version of a saved machine learning model in a Dataiku project.
• Automating model interpretability workflows by programmatically triggering partial dependency computations.
• Integrating with other Dataiku DSS resources and operations within an n8n workflow.

Practical example:

• A data scientist wants to analyze how changing a particular feature affects the output of a deployed model version. They use this node to trigger the partial dependency computation for that model version and then retrieve the results for further analysis or reporting.

Properties

Output

The node outputs the response from the Dataiku DSS API after requesting the computation of partial dependencies for the specified saved model version.

• The json output field contains the parsed JSON response from the API, which typically includes details about the launched computation task or the computed partial dependencies data.
• If the API returns binary data (not typical for this operation), it would be provided as binary output prepared for downstream nodes.
• In case of no content, the node outputs a status code indicating "204 No Content".

Dependencies

• Requires a valid connection to a Dataiku DSS instance.
• Requires an API authentication token credential for Dataiku DSS with sufficient permissions to access the project and saved model resources.
• The node expects the Dataiku DSS server URL and user API key to be configured in the credentials.

Troubleshooting

• Missing Credentials: The node will throw an error if the Dataiku DSS API credentials are not set or invalid. Ensure the API key and server URL are correctly configured.
• Required Parameters Missing: Errors will occur if mandatory parameters such as Project Key, Save Model ID, or Version ID are missing. Verify all required inputs are provided.
• API Errors: If the Dataiku DSS API returns errors (e.g., unauthorized, not found, or bad request), the node will throw an error with the message from the API. Check the validity of IDs and permissions.
• JSON Parsing Errors: If the API response is not valid JSON, the node attempts to return raw text. Unexpected response formats may indicate API issues or misconfiguration.

Links and References

• Dataiku DSS API Documentation
• Partial Dependence Explanation - Dataiku
• n8n Documentation - Creating Custom Nodes

This summary focuses on the "Machine Learning - Saved Model" resource and the "Compute Partial Dependencies of Version" operation as requested.