Struct EncodedVector 

pub struct EncodedVector {
    pub codec: Codec,
    pub dim: u16,
    pub bytes: Vec<u8>,
    pub params: Vec<f32>,
}

Encoded vector ready to persist or hand to a distance routine.

The byte buffer in Self::bytes is opaque to callers other than the matching Codec. The remaining fields are inspectable so an operator can reason about the row without decoding it.

Fields

codec: Codec

Codec used to produce Self::bytes.

dim: u16

Number of dimensions in the original f32 vector.

bytes: Vec<u8>

Codec-private payload.

params: Vec<f32>

Codec-private parameters. For Int8Quantized this is [min, scale]; for Fp16 it is empty. Stored as f32 so an inspect tool can render them without parsing the payload.

Implementations

impl EncodedVector

pub fn l2_norm(&self) -> f32

L2 norm of the decoded vector, computed by re-decoding.

Used by the inspect tooling and by some distance metrics that need to factor out vector magnitude.

Trait Implementations

impl Clone for EncodedVector

fn clone(&self) -> EncodedVector

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for EncodedVector

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for EncodedVector

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl PartialEq for EncodedVector

fn eq(&self, other: &EncodedVector) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Serialize for EncodedVector

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

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl StructuralPartialEq for EncodedVector