RuntimeError: cuda runtime error (59) : device-side assert triggered

最新推荐文章于 2025-05-24 16:35:47 发布

默_语

最新推荐文章于 2025-05-24 16:35:47 发布

阅读量1.7w

点赞数 13

CC 4.0 BY-SA版权

本文链接：https://blog.youkuaiyun.com/qq_22821801/article/details/90212788

Environment： Ubuntu，Pytorch1.0.1

运行程序时发现这个问题

Traceback (most recent call last):
File "train_pytorch1.py", line 217, in <module>
loss = F.cross_entropy(output, target)
File "/usr/local/python3/lib/python3.5/site-packages/torch/nn/functional.py", line 1970, in cross_entropy
return nll_loss(log_softmax(input, 1), target, weight, None, ignore_index, None, reduction)
File "/usr/local/python3/lib/python3.5/site-packages/torch/nn/functional.py", line 1790, in nll_loss
ret = torch._C._nn.nll_loss(input, target, weight, _Reduction.get_enum(reduction), ignore_index)
RuntimeError: cuda runtime error (59) : device-side assert triggered at /pytorch/aten/src/THCUNN/generic/ClassNLLCriterion.cu:111

这个异常大概是和计算损失值有关

查阅资料时发现很多道友都遇到过这种cuda runtime error(59)，大部分都是索引异常

根据