Actions26
- Delete Collection
- Get Collection
- Check Collection Exists
- Create Collection
- Update Collection
- Retrieve Point
- Retrieve Points
- Upsert Points
- Delete Points
- Update Vectors
- Delete Vectors
- Set Payload
- Overwrite Payload
- Delete Payload
- Clear Payload
- Batch Update Points
- Scroll Points
- Count Points
- Query Points
- Query Batch Points
- Query Points Groups
- Matrix Pairs
- Matrix Offsets
- Payload Facets
- Create Payload Index
- Delete Payload Index
Overview
The node provides an interface to Qdrant, a vector similarity search engine. Specifically, the "Matrix Pairs" operation allows users to find pairs of points within a collection that are close to each other based on vector similarity. This is useful for applications such as clustering, anomaly detection, or finding related items in recommendation systems.
Typical use cases include:
- Finding nearest neighbors for a sample of points in a dataset.
- Analyzing relationships between data points by examining their proximity in vector space.
- Reducing computational complexity by sampling points and limiting neighbor searches.
Properties
| Name | Meaning |
|---|---|
| Collection Name | The name of the collection in Qdrant where the points are stored. |
| Filter | JSON object specifying conditions to filter which points to consider in the search. |
| Sample | Number of points to randomly select from the collection to perform neighbor searches on. |
| Limit | Maximum number of neighbors to find per sampled point. |
| Using | Name of the vector field to use for querying; if empty, the default vector is used. |
| Shard Key | JSON array specifying shards to restrict the search to; if null, all shards are searched. |
| Consistency | Read consistency level for the operation; options: Majority, Quorum, All. |
| Timeout | Request timeout in seconds, overriding the global timeout setting. |
Output
The output contains JSON data representing pairs of points found to be close neighbors according to the query parameters. Each pair typically includes identifiers of the two points and their similarity or distance metric. This structured output enables further processing or analysis downstream in workflows.
If binary data is returned (not indicated here), it would represent raw vector or payload data associated with points.
Dependencies
- Requires connection to a Qdrant instance, configured via an API key credential and base URL.
- The node depends on network access to the Qdrant service endpoint.
- Proper permissions and authentication credentials must be set up in n8n for the Qdrant API.
Troubleshooting
Common issues:
- Incorrect collection name or non-existent collection will cause errors.
- Malformed JSON in the Filter or Shard Key properties can lead to request failures.
- Specifying a vector name in "Using" that does not exist in the collection will result in no matches or errors.
- Timeout too low may cause incomplete results or request failures.
Error messages:
- Authentication errors indicate invalid or missing API credentials.
- "Collection not found" means the specified collection does not exist.
- JSON parsing errors suggest invalid JSON syntax in input fields.
Resolutions:
- Verify collection names and existence in Qdrant dashboard.
- Validate JSON inputs using external tools before entering.
- Ensure correct API credentials and endpoint URLs.
- Increase timeout if operations take longer than expected.