Binding¶

class dragon.vm.tensorrt.Binding( cuda_engine, execution_context, idx_or_name, device_id )[source]¶: The binding wrapper for an input or output.

init¶

Binding.__init__( cuda_engine, execution_context, idx_or_name, device_id )[source]¶

Create a Binding.

Parameters:

cuda_engine (tensorrt.ICudaEngine) – The built cuda engine.
execution_context (tensorrt.IExecutionContext) – The execution context.
idx_or_name (Union[int, str]) – The binding index or name.
device_id (int, optional, default=0) – The index of executing device.

Properties¶

device_buffer¶

Binding.device_buffer¶

Return the device buffer.

Returns:: pycuda.gpuarray.GPUArray – The pycuda array taking the data.

device_dlpack¶

Binding.device_dlpack¶

Return the dlpack tensor wrapping device buffer.

Returns:: PyCapsule – The dlpack tensor object.

device_id¶

Binding.device_id¶

Return the index of binding device.

Returns:: int – The device index.

dtype¶

Binding.dtype¶

Return binding data type.

Returns:: str – The binding data type.

host_buffer¶

Binding.host_buffer¶

Return the host buffer.

Returns:: numpy.array – The numpy array taking the data.

host_dlpack¶

Binding.host_dlpack¶

Return the dlpack tensor wrapping host buffer.

Returns:: PyCapsule – The dlpack tensor object.

index¶

Binding.index¶

Return the binding index.

Returns:: int – The binding index.

is_input¶

Binding.is_input¶

Whether this binding is an input.

Returns:: bool – True if binding is an input.

name¶

Binding.name¶

Return the binding name.

Returns:: str – The binding name.

shape¶

Binding.shape¶

Return the binding shape.

Returns:: Tuple[int] – The binging shape.

Methods¶

get_async¶

Binding.get_async(stream)[source]¶

Copy and return the host buffer data.