First, you should move to by:
cd <RT-ODLab>
cd tools/Then, you can:
- Convert a standard YOLO model by:
python3 export_onnx.py -m yolov1 --weight ../weight/coco/yolov1/yolov1_coco.pth -nc 80 --img_size 640Notes:
-
-n: specify a model name. The model name must be one of the [yolox-s,m,l,x and yolox-nano, yolox-tiny, yolov3]
-
-c: the model you have trained
-
-o: opset version, default 11. However, if you will further convert your onnx model to OpenVINO, please specify the opset version to 10.
-
--no-onnxsim: disable onnxsim
-
To customize an input shape for onnx model, modify the following code in tools/export_onnx.py:
dummy_input = torch.randn(args.batch_size, 3, args.img_size, args.img_size)
Step1.
cd <YOLOX_HOME>/deployment/ONNXRuntimeStep2.
python3 onnx_inference.py --model ../../weights/onnx/11/yolov1.onnx -i ../test_image.jpg -s 0.3 --img_size 640Notes:
- --model: your converted onnx model
- -i: input_image
- -s: score threshold for visualization.
- --img_size: should be consistent with the shape you used for onnx convertion.