参考
- https://docs.ultralytics.com/zh/quickstart/#use-ultralytics-with-python
- https://pytorch.org/get-started/locally/
- https://www.zhihu.com/question/275575243
- https://github.com/onnx/onnx/issues/5773
- https://stackoverflow.com/questions/72352528/how-to-fix-winerror-206-the-filename-or-extension-is-too-long-error/76452218#76452218
环境
| 环境 | 版本 | 说明 |
|---|---|---|
| Windows | Windows 10 专业版 22H2 19045.4170 | |
| VS Code | 1.88.1 | |
| Git客户端 | 2.43.0.windows.1 | |
| Python | 3.12 | |
| pip | 3.12 | |
| ultralytics | 8.1.47 | pip 包 |
| torch | 2.2.2 | pip 包 |
| torchaudio | 2.2.2 | pip 包 |
| torchvision | 0.17.2 | pip 包 |
| onnx | 1.16.0 | pip 包 |
正文
-
安装 python3
-
设置pip源:
pip config set global.index-url https://pypi.tuna.tsinghua.edu.cn/simple -
安装yolov:
pip install ultralytics -
安装torch，https://pytorch.org/get-started/locally/
pip3 install torch torchvision torchaudio
-
设置git命令终端代理:（需要先打开梯子，并且本地端口对应，不明白梯子或者如果没有梯子就跳过这一步）
git config --global http.proxy 127.0.0.1:7890 git config --global https.proxy 127.0.0.1:7890 -
修改注册表，win+r 打开运行，输入regedit ，进入注册表找到：
Computer\HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\FileSystem，修改LongPathsEnabled=1
-
执行；
pip install --no-cache "onnx>=1.12.0"。 -
尝试运行（所有本文章内的文件都会自动去下载，所以只需要下面这部分代码就可以尝试体验）
代码是演示了完整流程：创建模型/加载模型/训练模型/评估/检测/导出onnx模型from ultralytics import YOLO # Create a new YOLO model from scratch model = YOLO('yolov8n.yaml') # Load a pretrained YOLO model (recommended for training) model = YOLO('yolov8n.pt') # Train the model using the 'coco128.yaml' dataset for 3 epochs results = model.train(data='coco128.yaml', epochs=3) # Evaluate the model's performance on the validation set results = model.val() # Perform object detection on an image using the model results = model('https://ultralytics.com/images/bus.jpg') # Export the model to ONNX format success = model.export(format='onnx')