【问题解决】2024-1-9 用gpu跑yolov8模型遇到的问题 : An attempt has been made to start a new process before the curre

最新推荐文章于 2024-12-24 08:31:23 发布

原创

最新推荐文章于 2024-12-24 08:31:23 发布 · 2.9k 阅读

48 ·

CC 4.0 BY-SA版权

文章标签：

#YOLO #python #深度学习

本文围绕使用GPU训练YOLO模型时运行mytrain.py文件报错展开。报错信息显示试图在当前进程完成引导阶段前启动新进程，分析指出是Windows系统使用多进程模块未正确设置条件所致。给出解决方案，需在脚本中添加if __name__ == '__main__'条件判断。

问题描述：An attempt has been made to start a new process before the current process has finished its bootstrapping phase.

想用gpu训练模型，运行mytrain.py

mytrain.py文件：
from ultralytics import YOLO
import multiprocessing
# 加载一个模型
    model = YOLO('yolov8n.yaml') # 从yaml构建一个新模型
    model = YOLO('weights/yolov8n.pt') # 加载预先训练的模型（建议进行训练）
    model = YOLO('yolov8n.yaml').load('weights/yolov8n.pt') # 根据YAML构建并传递权重
    results = model.train(data='datas/my_data.yaml', epochs=100, imgsz=640)

报错：

canning D:\myDesktop\ultralytics-main2\datas\labels\train.cache... 1720 images, 0 backgrounds, 0 corrupt: 100%|██████████| 1720/1720 [00:00<?, ?it/s] Traceback (most recent call last): File "<string>", line 1, in <module> File "D:\MyTools\mylab\lib\multiprocessing\spawn.py", line 116, in spawn_main exitcode = _main(fd, parent_sentinel) File "D:\MyTools\mylab\lib\multiprocessing\spawn.py", line 125, in _main prepare