Struct TensorCoreConfig 

pub struct TensorCoreConfig {
    pub use_volta_cores: bool,
    pub use_turing_cores: bool,
    pub use_ampere_cores: bool,
    pub use_hopper_cores: bool,
    pub wmma_tile_m: usize,
    pub wmma_tile_n: usize,
    pub wmma_tile_k: usize,
    pub auto_layout_optimization: bool,
    pub use_tf32: bool,
    pub sparsity_ratio: f32,
    pub async_execution: bool,
}

Tensor core matrix multiplication configuration

Fields

use_volta_cores: bool

Use Volta tensor cores (mixed precision GEMM)

use_turing_cores: bool

Use Turing tensor cores (INT8/INT4 support)

use_ampere_cores: bool

Use Ampere tensor cores (BF16/TF32 support)

use_hopper_cores: bool

Use Hopper tensor cores (FP8 support)

wmma_tile_m: usize

Warp matrix multiply tile size

wmma_tile_n: usize

wmma_tile_k: usize

auto_layout_optimization: bool

Enable automatic layout optimization

use_tf32: bool

Use TensorFloat-32 mode for FP32 operations

sparsity_ratio: f32

Sparsity level for structured sparse operations

async_execution: bool

Enable asynchronous execution

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.