个人学习YOLOV8下载（cpu）（mobaxterm）笔记

1.创建虚拟环境,激活虚拟环境

conda create -n yolov8 python=3.9
conda activate yolov8

2.安装cpu版本的pytorch

conda install pytorch torchvision torchaudio cpuonly -c pytorch

3.安装 ultralytics库

pip install ultralytics

4.官网下载ultralytics(即yolov8)

GitHub - ultralytics/ultralytics: NEW - YOLOv8 🚀 in PyTorch > ONNX > OpenVINO > CoreML > TFLite

5.测试环境是否建立成功

yolo predict model=yolov8n.pt source='https://ultralytics.com/images/bus.jpg'

6.在ultralytics-main文件下创立datasets文件夹

---

datasets------images----train

                                -----val

              ------labels------train

                                ------val

---

放入数据图片及标签，train是训练集，val是验证集

7.ultralytics-main文件夹下增添mydata.yaml文件

对mydata.yaml中的路径进行修改，和类别也可进行修改

path: ../datasets/CMD  # dataset root dir
train: images/train
val: images/val

7.然后可以对程序进行训练，命令是

yolo detect train data=mydata.yaml model=yolov8n.yaml epochs=100 imgsz=640

8.测试集命令

yolo detect val data=mydata.yaml model=runs/detect/train4/weights/best.pt batch=4