How to do quantization to int8 model of img_stage_lt_d.onnx and bev_stage_lt_d.onnx?

LCH1238 / bevdet-tensorrt-cpp

BEVDet implemented by TensorRT, C++； Achieving real-time performance on Orin

238 stars 40 forks source link

Open avBuffer opened 1 year ago

avBuffer commented 1 year ago

Hey, How to do quantization to int8 model of img_stage_lt_d.onnx and bev_stage_lt_d.onnx? Thanks!

LCH1238 commented 1 year ago

Later I will publish a unified engine model using explicit PTQ quantization, and I am cleaning up the code.

taoxunqiang commented 8 months ago

Is there a plan to release the code？

Thanks！