Struct MBAT

pub struct MBAT {
    pub data: Vec<i8>,
}

MBAT (Matrix Binding of Additive Terms) implementation using bipolar vectors.

Internally, an MBAT hypervector is stored as a Vec<i8> where each element is either -1 or +1.

Fields

data: Vec<i8>

The underlying bipolar vector.

Trait Implementations

impl Clone for MBAT

fn clone(&self) -> MBAT

Returns a copy of the value.

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

Performs copy-assignment from source.

impl Debug for MBAT

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for MBAT

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl PartialEq for MBAT

fn eq(&self, other: &MBAT) -> 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 MBAT

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

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl VSA for MBAT

fn generate(dim: usize, rng: &mut impl Rng) -> Self

Generate a random MBAT hypervector of the specified dimension.

Each element is randomly set to either +1 or -1 with equal probability.

Arguments

• dim - The dimensionality of the hypervector.
• rng - A mutable reference to a random number generator.

fn bundle( &self, other: &Self, tie_breaker: TieBreaker, rng: &mut impl Rng, ) -> Self

Bundle (superpose) two MBAT hypervectors.

Bundling is performed element-wise by computing the sum of corresponding elements and then applying the sign function. In the event of a tie (i.e. when the sum is zero), the provided tie-breaker is used to determine the output.

Arguments

• other - The hypervector to bundle with.
• tie_breaker - The rule to resolve ties.
• rng - A mutable reference to a random number generator.

fn bind(&self, other: &Self) -> Self

Bind two MBAT hypervectors.

Binding is implemented as the element-wise product.

Arguments

• other - The hypervector to bind with.

fn cosine_similarity(&self, other: &Self) -> f32

Compute the cosine similarity between two MBAT hypervectors.

For bipolar vectors, cosine similarity is computed as the dot product of the vectors divided by the dimension.

Arguments

• other - The hypervector to compare with.

fn hamming_distance(&self, other: &Self) -> f32

Compute the normalized Hamming distance between two MBAT hypervectors.

The normalized Hamming distance is defined as the fraction of positions where the two vectors differ.

Arguments

• other - The hypervector to compare with.

fn to_vec(&self) -> Vec<f32>

Convert the MBAT hypervector into a plain vector of f32 values.

Each element is converted from i8 to f32.

type Elem = i8

The type used to represent each element in the hypervector.

fn bundle_many( vectors: &[Self], tie_breaker: TieBreaker, rng: &mut impl Rng, ) -> Self

Bundles many hypervectors (folding a slice using the bundling operation).

fn bind_many(vectors: &[Self]) -> Self

Binds many hypervectors (folding a slice using the binding operation).

impl StructuralPartialEq for MBAT