Struct Runner

pub struct Runner<B: BackendIr> { /* private fields */ }

A runner is responsible for executing tensor operations for a given intermediate backend.

Implementations

impl<B: BackendIr> Runner<B>

pub fn new(device: B::Device) -> Self

Create a new runner.

pub fn register_tensor_data_id(&self, id: TensorId, data: TensorData)

Register a tensor from its data and id.

pub fn register_tensor_data_desc(&self, data: TensorData) -> TensorIr

Register a tensor and returns its intermediate representation.

pub fn register_empty_tensor_desc( &self, shape: Vec<usize>, dtype: DType, ) -> TensorIr

Register an empty tensor and returns its intermediate representation.

Trait Implementations

impl<B: Clone + BackendIr> Clone for Runner<B>

where B::Device: Clone,

fn clone(&self) -> Runner<B>

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<B: BackendIr> RunnerClient for Runner<B>

fn register(&self, op: OperationIr)

Execute a tensor operation.

type Device = <B as Backend>::Device

Device type.

fn read_tensor(&self, tensor: TensorIr) -> DynFut<TensorData>

Read the values contained by a tensor.

fn register_tensor_data(&self, data: TensorData) -> RouterTensor<Self>

Create a new RouterTensor from the tensor data.

fn register_empty_tensor( &self, shape: Vec<usize>, dtype: DType, ) -> RouterTensor<Self>

Create a new RouterTensor with no resources associated.

fn register_float_tensor( &self, shape: Vec<usize>, dtype: FloatDType, ) -> RouterTensor<Self>

Create a new float RouterTensor with no resources associated.

fn device(&self) -> Self::Device

Get the current device used by all operations handled by this client.

fn sync(&self)

Sync the runner, ensure that all computations are finished.

fn seed(&self, seed: u64)

Seed the runner.