1:.pth ------.onnx --------.engine
利用的库是
2: float32-float16-int8
3：遇见显性，隐性batch可参考
【TensorRT】execute_async VS execute_async_v2_context.execute_async_v2_昌山小屋的博客-CSDN博客
Developer Guide :: NVIDIA Deep Learning TensorRT Documentation
IExecutionContext --- NVIDIA TensorRT Standard Python API Documentation 8.6.1 documentation