HelixDB

Interact with HelixDB - the time-series database for modern applications. Requires HelixDB server running with deployed HelixQL queries.

Actions8

Vector → Search Vectors

AI-generated

Overview

This node interacts with HelixDB to perform vector operations, specifically searching for similar vectors within a specified collection. It is useful in scenarios where you need to find vectors that closely match a given query vector, such as in recommendation systems, image or text similarity searches, and other machine learning applications involving vector embeddings.

Use Case Examples

  1. Searching for similar product embeddings in an e-commerce recommendation system.
  2. Finding similar images based on their vector representations in a digital asset management system.
  3. Retrieving text documents with similar semantic embeddings in a search engine.

Properties

Name Meaning
Collection Name Name of the vector collection to search within.
Vector Data The query vector data represented as a JSON array of numbers to find similar vectors.
Search Limit Maximum number of similar vector results to return from the search.

Output

JSON

  • json - The search results containing vectors similar to the query vector, returned as JSON data.

Dependencies

  • Requires HelixDB server running with deployed HelixQL queries and an API key credential for authentication.

Troubleshooting

  • Ensure the HelixDB server is running and accessible at the configured port.
  • Verify that the specified vector collection exists and contains data.
  • Check that the vector data is correctly formatted as a JSON array of numbers.
  • If the search limit is set too high, it may impact performance or return excessive results.
  • Common error: 'Unsupported operation' indicates a mismatch in resource or operation parameters; verify correct selection.

Links

Discussion