[Previous line repeated 993 more times] RecursionError: maximum recursion depth exceeded

[LiChen-98]

我按照如下运行
python main.py configs/clrnet/clr_resnet18_culane.py --gpus 0

有报错，但程序未停止
[Previous line repeated 993 more times] RecursionError: maximum recursion depth exceeded

我在main.py文件中，按照如下修改
import sys
sys.setrecursionlimit(2000)

程序未报错，但在打印如下内容后，程序无反应
2022-09-29 16:37:07,902 - clrnet.utils.recorder - INFO - Build train loader...
2022-09-29 16:37:07,947 - clrnet.datasets.base_dataset - INFO - Loading CULane annotations...
2022-09-29 16:37:09,947 - clrnet.utils.recorder - INFO - Start training...
此时，电脑内存RAM占用，但显存和CPU未占用

[kkarnatak]

Hi, I have the same issue. I also modified it to 2000 but then everything is hung after "Start training" log. The GPU/CPU and RAM resources are unchanged. Did you find any solution?

[keeplearning-again]

I have the same problem on Windows system, and my GPU is single card with cuda 11.3. Hopefully anyone can solve this problem.

[Huapyw]

have you sloved the issue?

[fenghuohuo2001]

I also met the same problem. Have you solved it

[CrazyMustard-404]

have you sloved the issue?

楼主你好，我也遇到同样的错误请问解决了吗？

[fenghuohuo2001]

have you sloved the issue?

楼主你好，我也遇到同样的错误请问解决了吗？

I solved this problem by changing the "workers=0"

[CrazyMustard-404]

have you sloved the issue?

楼主你好，我也遇到同样的错误请问解决了吗？

I solved this problem by changing the "workers=0"

Thank you very much for your suggestions, but I don't know where to make the modification. I couldn't find it in the main.py file. Could you please tell me where to make the modification? Thank you once again.

[fenghuohuo2001]

For example, in path ./configs/clrnet/clr_resnet18_culane.py, the 'workers' is on line 120.

[CrazyMustard-404]

For example, in path ./configs/clrnet/clr_resnet18_culane.py, the 'workers' is on line 120.

Thank you very much for your help! It was able to train using your method.