DatabricksInstall

Overview

This node interacts with the Databricks Model Serving API to query a deployed model endpoint. It sends input data in JSON format to a specified serving endpoint and retrieves the inference results. This is useful for integrating machine learning model predictions directly into workflows, enabling automation of decision-making processes based on model outputs.

Practical examples:

• Sending customer data to a fraud detection model endpoint and receiving risk scores.
• Querying a recommendation model with user preferences to get personalized product suggestions.
• Automating quality control by sending sensor data to an anomaly detection model and acting on the results.

Properties

Output

The output is a JSON object containing the response from the model serving endpoint. This typically includes the inference results returned by the model, such as predictions, scores, or classifications based on the input data.

If the node supports binary data output (not indicated explicitly here), it would represent any binary content returned by the endpoint, but in this case, the focus is on JSON-formatted inference results.

Dependencies

• Requires an API key credential for authenticating with the Databricks API.
• The node uses the base URL and token from the configured credentials to make HTTP requests to the Databricks Model Serving endpoints.
• No additional external dependencies are indicated beyond standard HTTP request capabilities.

Troubleshooting

• Common issues:

  • Incorrect or missing endpoint name can cause the request to fail.
  • Malformed JSON input data may result in errors from the model serving API.
  • Network connectivity problems or invalid API tokens will prevent successful communication.

• Error messages:

  • API Error: <status> <statusText>: Indicates the Databricks API responded with an error status code. Check the endpoint name, input data format, and API permissions.
  • Network Error: No response received from server: Suggests network issues or incorrect host configuration. Verify network access and credential settings.
  • Other generic errors include malformed requests or unexpected exceptions; reviewing the error details and stack trace helps diagnose these.

• Resolution tips:

  • Ensure the endpoint name exactly matches the deployed model serving endpoint.
  • Validate JSON input data before sending.
  • Confirm that the API authentication token is valid and has necessary permissions.
  • Use the node's "Continue On Fail" option to handle errors gracefully during batch processing.

Links and References

• Databricks Model Serving Documentation
• n8n HTTP Request Node Documentation
• JSON Format Guide