Struct EfTrainingPoint 

pub struct EfTrainingPoint {
    pub query_norm: f64,
    pub query_variance: f64,
    pub ef_used: usize,
    pub result_count: usize,
    pub search_time_us: u64,
    pub recall: Option<f64>,
}

Training point for the ef (beam width) model.

Records search characteristics so the model can learn the optimal ef_search for a given query profile. Carries both the ef that was used AND the recall observed (when available), so the 2-head model can learn the ef → recall tradeoff jointly.

Fields

query_norm: f64

L2 norm of the query vector.

query_variance: f64

Variance across query dimensions.

ef_used: usize

ef_search value used for this query.

result_count: usize

Number of results returned.

search_time_us: u64

Wall-clock search time in microseconds.

recall: Option<f64>

Observed recall for this query (None if not measured).

Trait Implementations

impl Clone for EfTrainingPoint

fn clone(&self) -> Self

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for EfTrainingPoint

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for EfTrainingPoint

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for EfTrainingPoint

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.