Specify IDs for distance matrix API
Nobody has claimed this yet.
- Dominant language
- Rust
- Stars
- 34.7k
- Forks
- 2.7k
- Avg merge
- 1d 18h
- Merged PRs (30d)
- 187
Description
Is your feature request related to a problem? Please describe.
I would like to cluster/visualize retrieved vectors from the search or scroll APIs. This can be done by setting with_vectors=True but adds significant latency especially for a large number of high dim vectors. Common clustering and dimensionality reduction algorithms like UMAP or DBSCAN can be applied on precomputed distance matrices, instead of the raw vectors.
`
Describe the solution you'd like
The recent distance matrix API looks relevant: https://qdrant.tech/documentation/concepts/explore/#distance-matrix. But it samples points from a collection, and does not allow for specifying the ids.
Related issue: https://github.com/qdrant/qdrant/issues/2334
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start with the distance matrix API described in the linked Qdrant documentation and review related issue #2334 for existing context. The work is complete when the API accepts caller-specified point IDs instead of only sampling points from a collection.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- rust
- Domain
- api, databases
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100