YOLOX-ONNXRuntime in Python

This doc introduces how to convert your pytorch model into onnx, and how to run an onnxruntime demo to verify your convertion.

Download ONNX models.

Model	Parameters	GFLOPs	Test Size	mAP	Weights
YOLOX-Nano	0.91M	1.08	416x416	25.8	github
YOLOX-Tiny	5.06M	6.45	416x416	32.8	github
YOLOX-S	9.0M	26.8	640x640	40.5	github
YOLOX-M	25.3M	73.8	640x640	47.2	github
YOLOX-L	54.2M	155.6	640x640	50.1	github
YOLOX-Darknet53	63.72M	185.3	640x640	48.0	github
YOLOX-X	99.1M	281.9	640x640	51.5	github

Convert Your Model to ONNX

First, you should move to <YOLOX_HOME> by:

cd <YOLOX_HOME>

Then, you can:

1. Convert a standard YOLOX model by -n:

python3 tools/export_onnx.py --output-name yolox_s.onnx -n yolox-s -c yolox_s.pth

Notes:

• -n: specify a model name. The model name must be one of the [yolox-s,m,l,x and yolox-nane, 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.py:

  dummy_input = torch.randn(1, 3, exp.test_size[0], exp.test_size[1])

1. Convert a standard YOLOX model by -f. When using -f, the above command is equivalent to:

python3 tools/export_onnx.py --output-name yolox_s.onnx -f exps/default/yolox_s.py -c yolox_s.pth

1. To convert your customized model, please use -f:

python3 tools/export_onnx.py --output-name your_yolox.onnx -f exps/your_dir/your_yolox.py -c your_yolox.pth

ONNXRuntime Demo

Step1.

cd <YOLOX_HOME>/demo/ONNXRuntime

Step2.

python3 onnx_inference.py -m <ONNX_MODEL_PATH> -i <IMAGE_PATH> -o <OUTPUT_DIR> -s 0.3 --input_shape 640,640

Notes:

• -m: your converted onnx model
• -i: input_image
• -s: score threshold for visualization.
• --input_shape: should be consistent with the shape you used for onnx convertion.