Pytorch RuntimeError: couldn‘t find libcudart. Make sure CUDA libraries are installed in a default

最新推荐文章于 2024-07-10 09:57:43 发布

AI剑客

最新推荐文章于 2024-07-10 09:57:43 发布

阅读量804

点赞数

分类专栏： AI 文章标签： 1024程序员节

本文链接：https://blog.csdn.net/qq_43258953/article/details/109256730

版权

PyTorch GPU CUDA 错误解决 Anaconda

关键词由CSDN通过智能技术生成

AI 专栏收录该内容

145 篇文章 11 订阅

订阅专栏

# confirm PyTorch sees the GPU
from torch import cuda
assert cuda.is_available()
assert cuda.device_count() > 0
print(cuda.get_device_name(cuda.current_device()))

pytorch gpu无法运行，老是报这个error，重新安装CUDA，pytorch还是无法解决，keras与tensorflow gpu可以使用，最后重新安装Anaconda后成功解决。

>>>
>>> from torch import cuda
>>>
>>> assert cuda.is_available()
>>>
>>> assert cuda.device_count() > 0
>>>
>>> print(cuda.get_device_name(cuda.current_device()))
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "C:\Users\admin\Anaconda3\lib\site-packages\torch\cuda\__init__.py", line 367, in current_device
    _lazy_init()
  File "C:\Users\admin\Anaconda3\lib\site-packages\torch\cuda\__init__.py", line 180, in _lazy_init
    _cudart = _load_cudart()
  File "C:\Users\admin\Anaconda3\lib\site-packages\torch\cuda\__init__.py", line 87, in _load_cudart
    'LD_LIBRARY_PATH'))
RuntimeError: couldn't find libcudart. Make sure CUDA libraries are installed in a default location, or that they're in LD_LIBRARY_PATH.
>>>