Struct TensorCoreConfig 

pub struct TensorCoreConfig {
    pub enabled: bool,
    pub precision: PrecisionMode,
    pub tile_size: (usize, usize, usize),
    pub use_sparsity: bool,
    pub sparsity_ratio: f64,
    pub auto_scale: bool,
    pub loss_scale: f64,
}

Tensor core configuration

Fields

enabled: bool

Enable tensor cores if available

precision: PrecisionMode

Precision mode

tile_size: (usize, usize, usize)

Tile size for tensor core operations (M, N, K)

use_sparsity: bool

Use structured sparsity if available (NVIDIA Ampere+)

sparsity_ratio: f64

Sparsity ratio (e.g., 0.5 for 2:4 sparsity)

auto_scale: bool

Enable automatic precision scaling

loss_scale: f64

Loss scaling factor for mixed precision training

Trait Implementations

impl Clone for TensorCoreConfig

fn clone(&self) -> TensorCoreConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for TensorCoreConfig

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

Formats the value using the given formatter.

impl Default for TensorCoreConfig

fn default() -> Self

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for TensorCoreConfig

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for TensorCoreConfig

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

where __S: Serializer,

Serialize this value into the given Serde serializer.