Open Issues Need Help
View All on GitHub Discussion: Help us figure out the best way to evaluate vchordrq recall about 2 months ago
AI Summary: The task is to design and implement a PostgreSQL function, `vchordrq_evaluate_query_recall`, that efficiently estimates the recall of vector similarity searches using the `vchordrq` index. The function's input syntax needs to be finalized, considering options for specifying the query and parameters. The goal is to provide a reliable metric for determining when to reindex the `vchordrq` index based on recall degradation, improving upon the current heuristic of reindexing after a 4x data increase.
Complexity:
4/5
help wanted type/discussion 🧵
Scalable, fast, and disk-friendly vector search in Postgres, the successor of pgvecto.rs.
Rust
#artificial-intelligence#llmops#postgresql#vector-database#vector-search