[英]While Import python. ImportError: libcublas.so.9.0: cannot open shared object file: No such file or directory
I have tried many solutions like installing from different sources official google link Google.api...
, pypi
and also building from git repo. 我尝试了许多解决方案,例如从不同来源安装官方谷歌链接
Google.api...
, pypi
以及从git repo构建。
But every time I face the same problem ImportError: libcublas.so.9.0:
但每次我遇到同样的问题
ImportError: libcublas.so.9.0:
OS: Linux Arch
tensorflow: tensorflow-gpu
version 1.5
操作系统:
Linux Arch
tensorflow: tensorflow-gpu
1.5
版
Nvidia: Cuda 9.1 and Cudnn 7.0.5
Nvidia:
Cuda 9.1 and Cudnn 7.0.5
Note: tensorflow cpu is working fine
注意:tensorflow cpu工作正常
Traceback (most recent call last):
File "/usr/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow.py", line 58, in <module>
from tensorflow.python.pywrap_tensorflow_internal import *
File "/usr/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 28, in <module>
_pywrap_tensorflow_internal = swig_import_helper()
File "/usr/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 24, in swig_import_helper
_mod = imp.load_module('_pywrap_tensorflow_internal', fp, pathname, description)
File "/usr/lib/python3.6/imp.py", line 243, in load_module
return load_dynamic(name, filename, file)
File "/usr/lib/python3.6/imp.py", line 343, in load_dynamic
return _load(spec)
ImportError: libcublas.so.9.0: cannot open shared object file: No such file or directory
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/usr/lib/python3.6/site-packages/tensorflow/__init__.py", line 24, in <module>
from tensorflow.python import *
File "/usr/lib/python3.6/site-packages/tensorflow/python/__init__.py", line 49, in <module>
from tensorflow.python import pywrap_tensorflow
File "/usr/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow.py", line 74, in <module>
raise ImportError(msg)
ImportError: Traceback (most recent call last):
File "/usr/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow.py", line 58, in <module>
from tensorflow.python.pywrap_tensorflow_internal import *
File "/usr/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 28, in <module>
_pywrap_tensorflow_internal = swig_import_helper()
File "/usr/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 24, in swig_import_helper
_mod = imp.load_module('_pywrap_tensorflow_internal', fp, pathname, description)
File "/usr/lib/python3.6/imp.py", line 243, in load_module
return load_dynamic(name, filename, file)
File "/usr/lib/python3.6/imp.py", line 343, in load_dynamic
return _load(spec)
ImportError: libcublas.so.9.0: cannot open shared object file: No such file or directory
Failed to load the native TensorFlow runtime.
for some common reasons and solutions. Include the entire stack trace
above this error message when asking for help.
Your error message indicates that Tensorflow is looking for CUDA 9.0, while the default download is CUDA 9.1. 您的错误消息表明Tensorflow正在寻找CUDA 9.0,而默认下载是CUDA 9.1。 I suggest down-reviving to CUDA 9.0.
我建议下降到CUDA 9.0。 I just installed TF prebuilt binaries with CUDA 9.0 and the corresponding cudnn 7.05 and everything ran fine.
我刚刚用CUDA 9.0和相应的cudnn 7.05安装了TF预构建的二进制文件,一切运行正常。 From here and here it seems there are some issues with CUDA 9.1 that are still be to be worked out.
从这里到这里 ,似乎CUDA 9.1仍有一些问题需要解决。
Note also that currently the TF 1.5 install guide seems to be incorrect since it specifies CUDA 8.0 and cudnn 6.0 for the prebuilt TF while the release notes specify cuda 9 and cudnn 7 另请注意,目前TF 1.5安装指南似乎不正确,因为它为预构建的TF指定了CUDA 8.0和cudnn 6.0,而发行说明指定了cuda 9和cudnn 7
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.