aXeleRate/example_scripts/nvidia_jetson/classifier/engine.py

34 lines
1.1 KiB
Python

import tensorrt as trt
TRT_LOGGER = trt.Logger(trt.Logger.WARNING)
trt_runtime = trt.Runtime(TRT_LOGGER)
def build_engine(onnx_path, shape = [1,224,224,3], precision='FP32'):
"""
This is the function to create the TensorRT engine
Args:
onnx_path : Path to onnx_file.
shape : Shape of the input of the ONNX file.
"""
with trt.Builder(TRT_LOGGER) as builder, builder.create_network(1) as network, trt.OnnxParser(network, TRT_LOGGER) as parser:
if precision=='FP16':
builder.fp16_mode = True
builder.strict_type_constraints = True
builder.max_workspace_size = (256 << 20)
with open(onnx_path, 'rb') as model:
parser.parse(model.read())
network.get_input(0).shape = shape
engine = builder.build_cuda_engine(network)
return engine
def save_engine(engine, file_name):
buf = engine.serialize()
with open(file_name, 'wb') as f:
f.write(buf)
def load_engine(trt_runtime, plan_path):
with open(plan_path, 'rb') as f:
engine_data = f.read()
engine = trt_runtime.deserialize_cuda_engine(engine_data)
return engine