一、创建yolov5环境空间

conda create -n yolov5 python=3.9

二、激活yolov5环境空间

conda activate yolov5

三、安装适用于CPU的torch环境

pip install torch==2.0.0 torchvision==0.15.1 torchaudio==2.0.1 --index-url https://download.pytorch.org/whl/cpu

四、克隆项目

git clone https://github.com/ultralytics/yolov5.git

五、进入项目目录

cd yolov5/

六、安装项目所需依赖

pip install -r requirements.txt

七、下载模型（在项目目录下的md文件里下翻可以看到）

八、将pt文件放到项目下

九、运行测试

python detect.py --weights pt/yolov5n.pt --source data/images/bus.jpg

问题汇总：

        1. A module that was compiled using NumPy 1.x cannot be run in NumPy 2.0.2 as it may crash. To support both 1.x and 2.x

        解决：降级NumPy版本

pip install "numpy<2" --upgrade

        2. ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts. opencv-python 4.13.0.92 requires numpy>=2; python_version >= "3.9", but you have numpy 1.26.4 which is incompatible.

        解决：安装兼容NumPy版本的OpenCV

# 卸载当前OpenCV
pip uninstall opencv-python opencv-python-headless -y

# 安装兼容NumPy 1.x的OpenCV版本
pip install opencv-python==4.9.0.80