Python version of YOLOX object detection base on MegEngine.


Step1: install requirements

python3 -m pip install megengine -f https://megengine.org.cn/whl/mge.html

Step2: convert checkpoint weights from torch’s path file

python3 convert_weights.py -w yolox_s.pth -o yolox_s_mge.pkl

Step3: run demo

This part is the same as torch’s python demo, but no need to specify device.

python3 demo.py image -n yolox-s -c yolox_s_mge.pkl --path ../../../assets/dog.jpg --conf 0.25 --nms 0.45 --tsize 640 --save_result

[Optional]Step4: dump model for cpp inference

Note: result model is dumped with optimize_for_inference and enable_fuse_conv_bias_nonlinearity.

python3 dump.py -n yolox-s -c yolox_s_mge.pkl --dump_path yolox_s.mge