コード例 #1
0
ファイル: runner.py プロジェクト: clayne/TensorRT
 def get_input_metadata_impl(self):
     bindings_per_profile = trt_util.get_bindings_per_profile(
         self.context.engine)
     # This function always uses binding names of the 0th profile.
     return trt_util.get_input_metadata_from_engine(
         self.context.engine,
         start_binding=0,
         end_binding=bindings_per_profile)
コード例 #2
0
ファイル: buffers.py プロジェクト: leo-XUKANG/TensorRT-1
 def from_engine(engine):
     buffers = Buffers()
     bindings_per_profile = trt_util.get_bindings_per_profile(engine)
     for idx in range(bindings_per_profile):
         binding = engine[idx]
         dtype = trt.nptype(engine.get_binding_dtype(binding))
         buffers.device_buffers[binding] = cuda.DeviceBuffer(dtype=dtype)
         if not engine.binding_is_input(binding):
             buffers.outputs[binding] = np.empty(shape=tuple(), dtype=dtype)
     G_LOGGER.extra_verbose("Created device buffers: {:}".format(
         buffers.device_buffers))
     return buffers
コード例 #3
0
ファイル: runner.py プロジェクト: phongphuhanam/TensorRT
        def make_buffers(engine):
            """
            Creates empty host and device buffers for the specified engine.
            Always uses binding names from Profile 0.
            """
            device_buffers = OrderedDict()
            host_output_buffers = OrderedDict()

            for idx in range(trt_util.get_bindings_per_profile(engine)):
                binding = engine[idx]
                dtype = trt_util.np_dtype_from_trt(engine.get_binding_dtype(binding))
                device_buffers[binding] = cuda.DeviceArray(dtype=dtype)
                if not engine.binding_is_input(binding):
                    host_output_buffers[binding] = np.empty(shape=tuple(), dtype=dtype)
            G_LOGGER.extra_verbose("Created device buffers: {:}".format(device_buffers))
            return device_buffers, host_output_buffers