Struct EmbeddingSpace 

pub struct EmbeddingSpace {
    pub vector: Vec<f32>,
    pub dimensions: u32,
    pub model: String,
    pub generated_at: u64,
}

An embedding vector in a named semantic space.

Knowledge Objects can carry embeddings in multiple spaces simultaneously:

• "semantic" — general-purpose sentence embedding (e.g., text-embedding-3-small)
• "code" — code-specific embedding (e.g., CodeBERT)
• "temporal" — time-series embedding for temporal similarity
• "visual" — image/diagram embedding (e.g., CLIP)

Each space can have a different dimensionality and distance metric.

Fields

vector: Vec<f32>

The embedding vector (f32 components).

dimensions: u32

Dimensionality of this embedding.

model: String

The model that generated this embedding. Enables re-embedding when models are upgraded.

generated_at: u64

When this embedding was generated (HLC-encoded microseconds). Enables staleness detection and re-embedding triggers.

Implementations

impl EmbeddingSpace

pub fn new( vector: Vec<f32>, model: impl Into<String>, generated_at: u64, ) -> Self

Create a new embedding in a given space.

pub fn norm(&self) -> f32

The L2 norm of this embedding vector.

pub fn normalize(&mut self)

Normalize this embedding to unit length (for cosine similarity as dot product).

Trait Implementations

impl Clone for EmbeddingSpace

fn clone(&self) -> EmbeddingSpace

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for EmbeddingSpace

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for EmbeddingSpace

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for EmbeddingSpace

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

where __S: Serializer,

Serialize this value into the given Serde serializer.