RuntimeError: Cannot re-initialize CUDA in forked subprocess解决方法之一

在程序最开始的地方加上：

torch.multiprocessing.set_start_method(‘spawn’)

即可解决，如下第二行所示：

if __name__=='__main__':torch.multiprocessing.set_start_method('spawn')args = config_parser().parse_args()data_loader = CreateDataLoader(args)train()

其他详细的解释和解决方法：

https://www.pythonheidong.com/blog/article/505157/8b5b28765cba2958ac02/
https://www.pythonf.cn/read/65459

RuntimeError: Cannot re-initialize CUDA in forked subprocess解决方法之一相关推荐

Cannot re-initialize CUDA in forked subprocess. To use CUDA with multiprocessing, you
RuntimeError: Cannot re-initialize CUDA in forked subprocess. To use CUDA with multiprocessing, you ...
cannot re-initialize CUDA in forked subprocess.To use CUDA with multiprocessing,you must use the ...
cannot re-initialize CUDA in forked subprocess.To use CUDA with multiprocessing,you must use the 'sp ...
DeepFaceLab报错，CUDA driver is insufficient 解决方法!
DeepFaceLab报错,CUDA driver is insufficient 解决方法! 参考文章: (1)DeepFaceLab报错,CUDA driver is insufficient 解 ...
Python 使用numpy报错：runtimeError: package fails to pass a sanity check解决方法
Python中使用numpy和pandas时报错:RuntimeError: The current Numpy installation ('...\venv\lib\site-packages\n ...
hibernate中报错could not initialize proxy - no Session的解决方法
hibernate中报错:could not initialize proxy - no Session,怎么解决? 解决方法: 在该hbm文件中的 many-to-one 的最后加上:lazy=&q ...
安装pytorch报错torch.cuda.is_available()=false的解决方法
问题介绍在安装torch之后,命令行(Anaconda Powershell Prompt)运行这三行代码: python # python import torch torch.cuda.is_a ...
RuntimeError: Cannot re-initialize CUDA in forked subprocess
问题就在于训练所使用的num_workers数量问题,直接将 num_workers=0 就不会报错了
IPVS: Can't initialize ipvs: Protocol not available解决方法
经排查发现本地的测试环境磁盘空间满了,造成数据库不能写入数据. 最终的罪魁祸首是日志文件. 由于我们使用的是lvs负载均衡,前端的Keepalived启动之后不能初始化ipvs协议,所以日志在不停的刷 ...
runtimeError: package fails to pass a sanity check解决方法
目录 import pandas运行时报错解决办法 import pandas运行时报错 raise RuntimeError(msg.format(__file__)) from None Run ...
CUDA安装失败的解决方法（Win10）
取消下面的勾选取消勾选CUDA里Visual Studio Integration 重启即可

RuntimeError: Cannot re-initialize CUDA in forked subprocess解决方法之一

RuntimeError: Cannot re-initialize CUDA in forked subprocess解决方法之一相关推荐

最新文章

热门文章