Enum ComputePrecision 

pub enum ComputePrecision {
    Half,
    Single,
    Double,
    Mixed,
}

Computation precision

Variants

Half

Half precision (float16) - reduced memory, faster on tensor cores Suitable for approximate calculations where high precision isn’t critical

Single

Single precision (float32) - balanced precision and performance Recommended for most quantum simulations

Double

Double precision (float64) - highest precision Required for high-fidelity simulations and error-sensitive algorithms

Mixed

Mixed precision (automatic FP16/FP32 switching) Uses FP16 for matrix operations (tensor cores) and FP32 for accumulation Provides near-FP32 accuracy with FP16 speed

Implementations

impl ComputePrecision

pub fn bytes_per_amplitude(self) -> usize

Get bytes per complex amplitude for this precision

pub fn speed_factor(self) -> f64

Get relative speed multiplier (approximate) Higher values = faster computation

pub fn accuracy_factor(self) -> f64

Get relative accuracy (approximate) Higher values = more accurate

pub fn uses_tensor_cores(self) -> bool

Check if precision uses tensor cores (if available)

pub fn description(self) -> &'static str

Get human-readable description

Trait Implementations

impl Clone for ComputePrecision

fn clone(&self) -> ComputePrecision

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ComputePrecision

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

Formats the value using the given formatter.

impl PartialEq for ComputePrecision

fn eq(&self, other: &ComputePrecision) -> 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 Copy for ComputePrecision

impl Eq for ComputePrecision

impl StructuralPartialEq for ComputePrecision