YOLOv8 模型训练

模型训练

yolo task=detect mode=train data=yolo-qie.yaml model=yolov8n.pt epochs=50 workers=1 batch=16

训练的三种方式

命令行方式

cd /Users/wangzhiwei/study/ai/anaconda-study/ultralytics

yolo task=detect mode=train data=yolo-qie.yaml model=yolov8n.pt epochs=50 workers=1 batch=16

Results saved to /Users/wangzhiwei/study/ai/anaconda-study/ultralytics/runs/detect/train8

脚本方式

├── datasets (ultralytics 项目外)
│   └── qie
│   ├── images
│   │   ├── train
│   │   └── val
│   ├── labels
│   │   ├── classes.txt
│   │   ├── train
│   │   ├── train.cache
│   │   ├── val
│   │   └── val.cache
│   └── yolo11n.pt
└── ultralytics
└── yolov11-train.py

yolo-qie.yaml

path: qie # dataset root dir
train: images/train # train images (relative to 'path') 4 images
val: images/val # val images (relative to 'path') 4 images
test: # test images (optional)

# Classes
names:
  0: qie

yolov11-train.py

from ultralytics import YOLO
model = YOLO("yolov8n.pt") # yolo11 的仓库也可以用 yolov8n.pt 训练
model.train(data="yolo-qie.yaml", epochs=50, batch=16, workers=0)


训练结果
Results saved to /Users/wangzhiwei/study/ai/anaconda-study/ultralytics/runs/detect/train7

默认配置方式

验证训练结果
cd /Users/wangzhiwei/study/ai/anaconda-study/ultralytics

yolo detect predict model=runs/detect/train8/weights/best.pt source=./video.mp4 show=True conf=0.03

参考：
https://www.bilibili.com/video/BV1CfQqYUEL5?spm_id_from=333.788.player.switch&vd_source=ffda878df0ed45bee1ade91d8f451048&p=5