Struct GreedyMethod 

pub struct GreedyMethod {
    pub alpha: f64,
    pub temperature: f64,
}

Configuration for the greedy optimizer.

Fields

alpha: f64

Weight balancing output size vs input size reduction.

• α = 0.0: Minimize output tensor size (default)
• α = 1.0: Maximize input tensor size reduction

temperature: f64

Temperature for stochastic selection.

• temperature = 0.0: Deterministic greedy (default)
• temperature > 0.0: Boltzmann sampling

Implementations

impl GreedyMethod

pub fn new(alpha: f64, temperature: f64) -> Self

Create a new greedy method with custom parameters.

pub fn stochastic(temperature: f64) -> Self

Create a stochastic greedy method with given temperature.

Trait Implementations

impl Clone for GreedyMethod

fn clone(&self) -> GreedyMethod

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl CodeOptimizer for GreedyMethod

fn optimize<L: Label>( &self, code: &EinCode<L>, size_dict: &HashMap<L, usize>, ) -> Option<NestedEinsum<L>>

Optimize the contraction order for an EinCode.

impl Debug for GreedyMethod

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

Formats the value using the given formatter.

impl Default for GreedyMethod

fn default() -> Self

Returns the “default value” for a type.