Enum Backend 

pub enum Backend {
    TensorRt,
    Cuda,
    Cpu,
    Custom {
        name: String,
        provider: ExecutionProviderDispatch,
    },
}

An installed ONNX Runtime execution provider.

Variants

TensorRt

NVIDIA TensorRT execution provider.

Cuda

NVIDIA CUDA execution provider.

Cpu

ONNX Runtime’s CPU execution provider.

Custom

Configure an additional ORT provider without adding its native dependencies. CUDA/TensorRT dispatches are rejected: use the built-in variants and typed options so the engine can own streams and coordinate graph capture. Custom implementations must not register CUDA/TensorRT indirectly.

Fields

name: String

Display name used in backend reports and fallback events.

provider: ExecutionProviderDispatch

Configured provider to register with ONNX Runtime.

Implementations

impl Backend

pub fn name(&self) -> &str

Display name used in backend reports and fallback events.

Trait Implementations

impl Clone for Backend

fn clone(&self) -> Self

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for Backend

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

Formats the value using the given formatter.