about tensorrt How do I use my own model

iwatake2222 / play_with_tensorrt

Sample projects for TensorRT in C++

Apache License 2.0

191 stars 34 forks source link

about tensorrt How do I use my own model #23

Closed zhanghongyong123456 closed 2 years ago

zhanghongyong123456 commented 2 years ago

Environment (Hardware)

Hardware: Device, CPU, GPU, etc.
Software: VS 2019 (Please include version information)

Project Name

pj_tensorrt_seg_robust_video_matting

Issue Details

1660553167985 1660553328098 What should be done?

How to Reproduce

Steps to reproduce the behavior. Please include your cmake command.

Error Log

error log

Additional Information

Add any other context about the problem here.

iwatake2222 commented 2 years ago

Please modify https://github.com/iwatake2222/play_with_tensorrt/blob/master/pj_tensorrt_seg_robust_video_matting/image_processor/segmentation_engine.cpp , and please read source code to find how to modify it.

zhanghongyong123456 commented 2 years ago

Please modify https://github.com/iwatake2222/play_with_tensorrt/blob/master/pj_tensorrt_seg_robust_video_matting/image_processor/segmentation_engine.cpp , and please read source code to find how to modify it.

Thank you very much for your reply ，I am looking at this .cpp source file，hello i want to ask：1>I have my own trained model（about MobilenetV3 model）, how do I convert it to the format .onnx or .trt required by the project ；2> Can our project achieve this speed, or 4K real time on nvidia 3090, if not, should improve in those areas