Actions27
- Collection Actions
- Payload Actions
- Point Actions
- Search Actions
- Vector Actions
Search → Matrix Offsets
AI-generatedOverview
This node operation performs a 'Matrix Offsets' search on a specified collection within Qdrant, a vector search engine. It allows users to query points in the collection based on vector similarity and other filtering criteria, returning results with matrix offsets that indicate the relative positions of points in the vector space. This operation is useful for applications involving similarity search, recommendation systems, and spatial data analysis where understanding the relative positioning of data points is important.
Use Case Examples
- Finding similar items in a product catalog by querying a vector collection with specific filters and limits.
- Analyzing spatial relationships between data points in a geographic dataset by retrieving matrix offsets.
- Performing advanced vector searches with custom consistency and timeout settings to optimize query performance.
Properties
| Name | Meaning |
|---|---|
| Collection Name | Specifies the name of the collection to search within. This is required to identify the target dataset for the matrix offsets query. |
| Filter | A JSON object defining conditions to filter which points are selected for the search. Only points satisfying these conditions are considered. |
| Sample | Defines how many points to select and search within from the filtered set. |
| Limit | Maximum number of results to return from the query. |
| Using | Specifies which vector name to use for querying. If left empty, the default vector is used. |
| Shard Key | A JSON object specifying which shards to search within. If not specified, the search is performed across all shards. |
| Consistency | Defines the read consistency guarantees for the operation, affecting how the query reads data from replicas. |
| Timeout | Overrides the global timeout for this request in seconds, controlling how long the query waits before aborting. |
| Request Options | Additional settings for the request such as batching, SSL certificate validation, proxy configuration, and HTTP timeout. |
Output
JSON
result- The main output containing the matrix offsets results from the query, representing relative positions of points in the vector space.
Dependencies
- Requires an API key credential for Qdrant REST API access, including the base URL and authentication headers.
Troubleshooting
- Common issues include invalid collection names or filters leading to empty or failed queries.
- Timeout errors may occur if the specified timeout is too low; increasing the timeout can help.
- SSL certificate validation errors can be bypassed by enabling the 'Ignore SSL Issues' option, but this reduces security.
- Incorrect shard key specifications may result in incomplete search results.
Links
- Qdrant Documentation - Official documentation for Qdrant, providing detailed information on collections, points, and search operations including matrix offsets.