Struct TensorRtConfig 

pub struct TensorRtConfig {
    pub plan_path: PathBuf,
    pub max_batch_size: u32,
    pub precision: TrtPrecision,
    pub device_id: u32,
}

Engine-loading configuration.

The plan_path is a serialised TensorRT plan (output of IBuilder::buildSerializedNetwork or atomr-accel-tensorrt::TrtMsg::Build). Builds are out-of-scope for this runner — operators either hand-build a plan with the upstream actor or import an ONNX file via the tensorrt-onnx feature on atomr-accel-tensorrt.

Fields

plan_path: PathBuf

Path to a serialised TensorRT plan.

max_batch_size: u32

Maximum batch size the engine was built for. Used by the adapter layer (when wired) to chunk requests.

precision: TrtPrecision

Precision the engine was built for. Reported via telemetry; the engine itself encodes the constraints.

device_id: u32

CUDA device ordinal. Defaults to 0.

Trait Implementations

impl Clone for TensorRtConfig

fn clone(&self) -> TensorRtConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for TensorRtConfig

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for TensorRtConfig

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for TensorRtConfig

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

where __S: Serializer,

Serialize this value into the given Serde serializer.