feat: support to sepcify ef search param (#1844)

Signed-off-by: BubbleCal <bubble-cal@outlook.com>

rust/lancedb/src/query.rs

@@ -704,6 +704,9 @@ pub struct VectorQuery {
     // IVF PQ - ANN search.
     pub(crate) query_vector: Vec<Arc<dyn Array>>,
     pub(crate) nprobes: usize,
+    // The number of candidates to return during the refine step for HNSW,
+    // defaults to 1.5 * limit.
+    pub(crate) ef: Option<usize>,
     pub(crate) refine_factor: Option<u32>,
     pub(crate) distance_type: Option<DistanceType>,
     /// Default is true. Set to false to enforce a brute force search.
@@ -717,6 +720,7 @@ impl VectorQuery {
             column: None,
             query_vector: Vec::new(),
             nprobes: 20,
+            ef: None,
             refine_factor: None,
             distance_type: None,
             use_index: true,
@@ -776,6 +780,18 @@ impl VectorQuery {
         self
     }
 
+    /// Set the number of candidates to return during the refine step for HNSW
+    ///
+    /// This argument is only used when the vector column has an HNSW index.
+    /// If there is no index then this value is ignored.
+    ///
+    /// Increasing this value will increase the recall of your query but will
+    /// also increase the latency of your query.  The default value is 1.5*limit.
+    pub fn ef(mut self, ef: usize) -> Self {
+        self.ef = Some(ef);
+        self
+    }
+
     /// A multiplier to control how many additional rows are taken during the refine step
     ///
     /// This argument is only used when the vector column has an IVF PQ index.

rust/lancedb/src/remote/table.rs

@@ -196,6 +196,7 @@ impl<S: HttpSend> RemoteTable<S> {
         body["prefilter"] = query.base.prefilter.into();
         body["distance_type"] = serde_json::json!(query.distance_type.unwrap_or_default());
         body["nprobes"] = query.nprobes.into();
+        body["ef"] = query.ef.into();
         body["refine_factor"] = query.refine_factor.into();
         if let Some(vector_column) = query.column.as_ref() {
             body["vector_column"] = serde_json::Value::String(vector_column.clone());
@@ -1121,6 +1122,7 @@ mod tests {
                 "prefilter": true,
                 "distance_type": "l2",
                 "nprobes": 20,
+                "ef": Option::<usize>::None,
                 "refine_factor": null,
             });
             // Pass vector separately to make sure it matches f32 precision.
@@ -1166,6 +1168,7 @@ mod tests {
                 "bypass_vector_index": true,
                 "columns": ["a", "b"],
                 "nprobes": 12,
+                "ef": Option::<usize>::None,
                 "refine_factor": 2,
             });
             // Pass vector separately to make sure it matches f32 precision.

rust/lancedb/src/table.rs

@@ -1904,6 +1904,9 @@ impl TableInternal for NativeTable {
             query.base.offset.map(|offset| offset as i64),
         )?;
         scanner.nprobs(query.nprobes);
+        if let Some(ef) = query.ef {
+            scanner.ef(ef);
+        }
         scanner.use_index(query.use_index);
         scanner.prefilter(query.base.prefilter);
         match query.base.select {