TRT for yolov3 with INT8 calibration

I run the example yolo python code,but it only had FP16 and FP32.

Does anyone have an example of TRT for yolov3 INT8 calibration in python?

Thanks a lot.

any updates on this?

I’m also trying to run YOLOv3 on TensorRT using int8 precision.

I am also interested to getting access to a python sample for this.

I’m interested to get a guide for this.

And i’m interesting too

Did any of you make progress with yolov3/INT8 precision ? I’m trying to figure out how to load a custom trained YOLOv3 model into TensorRT and generate a INT8 precision version for nVidia processors. Will appreciate any related information/links on this topic.


I shared full source code about how to build INT8 TensorRT yolov3/yolov4 models in my jkjung-avt/tensorrt_demos repository. Please check out Demo #6: Using INT8 and DLA core in particular.

1 Like