Actions26
- Collection Actions
- Payload Actions
- Point Actions
- Search Actions
- Vector Actions
Overview
The node provides an interface to perform a "Matrix Offsets" operation on a vector search engine collection (Qdrant). This operation calculates the pairwise distances (offsets) between selected points in a specified collection, which is useful for analyzing relationships or similarities between data points represented as vectors.
Common scenarios include:
- Finding how close or far apart certain items are within a dataset.
- Analyzing clusters or groups by measuring distances between sample points.
- Performing similarity analysis in recommendation systems or anomaly detection.
For example, you might select a subset of product embeddings from a collection and compute their pairwise distances to understand product similarity patterns.
Properties
| Name | Meaning |
|---|---|
| Collection Name | The target collection to query. Can be selected from a list of existing collections or specified by name. |
| Filter | JSON object defining conditions to filter points before calculating offsets. Only points satisfying these conditions are considered. |
| Sample | Number of points to randomly select from the filtered set to calculate offsets within. |
| Limit | Maximum number of results (pairs with offsets) to return. Minimum value is 1. |
| Using | Name of the vector field to use for querying. If empty, the default vector is used. |
| Shard Key | JSON 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 if set. Default is 100 seconds. |
Output
The output contains a JSON array where each element represents a pair of points along with their calculated offset (distance). The structure typically includes identifiers of the two points and the numeric offset value representing their distance in vector space.
If binary data were involved, it would represent raw vector or payload data, but this operation focuses on numeric distance results only.
Dependencies
- Requires connection to a Qdrant vector search service endpoint.
- Needs an API key or authentication token configured in n8n credentials to authorize requests.
- The node uses REST API calls to interact with the Qdrant service.
Troubleshooting
- Timeout errors: If the operation takes too long, consider increasing the timeout property or reducing the sample size.
- No results returned: Check that the collection name is correct and that the filter condition matches existing points.
- Invalid filter JSON: Ensure the filter property is valid JSON and correctly structured according to Qdrant's filtering syntax.
- Shard key issues: Specifying incorrect shard keys may result in no data found; verify shard keys correspond to existing shards.
- Authentication failures: Confirm that the API key credential is properly set up and has sufficient permissions.