系统优化加速工具_免费收录链接网_为什么中国禁止谷歌浏览器_营销型网站建设推荐

1. train.py

from ultralytics import YOLO
import cv2
from PIL import Image
import osif __name__ ==  '__main__':os.environ["CUDA_VISIBLE_DEVICES"]="0,1"# 加载一个模型model = YOLO('/home/8888/yolov8-test/yolov8s.pt')  # 加载预训练模型（推荐用于训练）# 训练模型results = model.train(data='/home/8888/yolov8-test/data/coco.yaml', epochs=5000, imgsz=640, batch=8, device=[0,1], workers=8, patience=50)

2. 执行命令

python -m torch.distributed.launch --nproc_per_node=2 train.py