Python tensorflow 的 libcublas.so.8.0 错误
声明：本页面是StackOverFlow热门问题的中英对照翻译，遵循CC BY-SA 4.0协议，如果您需要使用它，必须同样遵循CC BY-SA许可，注明原文地址和作者信息，同时你必须将它归于原作者（不是我）：StackOverFlow
原文地址: http://stackoverflow.com/questions/44865253/
Warning: these are provided under cc-by-sa 4.0 license. You are free to use/share it, But you must attribute it to the original authors (not me):
StackOverFlow
libcublas.so.8.0 error with tensorflow
提问by Shimron F
I am currently trying to run a project from GitHub: https://github.com/nilboy/pixel-recursive-super-resolution
我目前正在尝试从 GitHub 运行一个项目：https: //github.com/nilboy/pixel-recursive-super-resolution
After installing ubuntu on VM, I installed python with tensorflow 2.7 using the pip command.
在 VM 上安装 ubuntu 后，我使用 pip 命令安装了带有 tensorflow 2.7 的 python。
When I try to run the training session with python I get this error:
当我尝试使用 python 运行培训课程时，出现此错误：
ImportError: libcublas.so.8.0: cannot open shared object file: No such file or directory
导入错误：libcublas.so.8.0：无法打开共享对象文件：没有这样的文件或目录
I could't find anything exactly the same, only this error: ImportError: libcudart.so.8.0: cannot open shared object file
我找不到完全相同的东西，只有这个错误： ImportError: libcudart.so.8.0: cannot open shared object file
I am using Yoga 13, so I don't have an NVIDIA GPU and I am trying to run this project on the CPU. Can you help identify the tensorflow problem?
我正在使用 Yoga 13，所以我没有 NVIDIA GPU，我正在尝试在 CPU 上运行这个项目。你能帮助识别张量流问题吗？
回答by u7772176
You need to install Cuda 8.0 and configure the environment as below:
您需要安装 Cuda 8.0 并配置如下环境：
export PATH="$PATH:/usr/local/cuda-8.0/bin"
export LD_LIBRARY_PATH="/usr/local/cuda-8.0/lib64"
回答by nexoma
may be preinstalled tensorflow-gpu, so
可能预装了tensorflow-gpu，所以
pip3 install tensorflow --ignore-installed
pip3 安装 tensorflow --ignore-installed
as this helped me from stackoverflow
因为这帮助我摆脱了 stackoverflow
回答by Simbarashe Timothy Motsi
It may be an issue with the CUDA version installed. try
$ sudo apt-get install cuda-8-0
可能是安装的 CUDA 版本有问题。试试
$ sudo apt-get install cuda-8-0
回答by James L.
In my opinion the tricky steps are installing Nvidia drivers and Cuda-8. If you follow the instructions you're gonna have a bad time. Instead install it thru apt:
在我看来，棘手的步骤是安装 Nvidia 驱动程序和 Cuda-8。如果你按照说明你会过得很糟糕。而是通过apt以下方式安装它：
sudo apt-get install nvidia-384 #can type nvidia then hit "tab" to view all available options
sudo apt-get install cuda-8-0
If that doesn't work...good luck. You can try this:
如果那不起作用……祝你好运。你可以试试这个：
Check that you do not have cuda-9: ls /usr/local
检查您是否没有 cuda-9： ls /usr/local
Reboot
重启
Run export LD_LIBRARY_PATH=/usr/local/cuda-8.0/extras/CUPTI/lib64:$LD_LIBRARY_PATHbut replace /usr/local/cuda-8.0/extras/CUPTI/lib64with the relevant output from echo $LD_LIBRARY_PATH
运行export LD_LIBRARY_PATH=/usr/local/cuda-8.0/extras/CUPTI/lib64:$LD_LIBRARY_PATH但替换/usr/local/cuda-8.0/extras/CUPTI/lib64为相关输出echo $LD_LIBRARY_PATH
Check that it worked nvcc --versionprints out CUDA version
检查它是否有效nvcc --version打印出 CUDA 版本
回答by Umair Qadir
I got the same error while running Tensorflow with GPU support in Python and solved it like this:
我在 Python 中运行具有 GPU 支持的 Tensorflow 时遇到了同样的错误，并像这样解决了它：
$ export PATH="$PATH:/usr/local/cuda-8.0/bin"
$ export LD_LIBRARY_PATH="/usr/local/cuda-8.0/lib64"
For enabling all users you may have to enter these export statements in etc/environments in Ubuntu
为了启用所有用户，您可能必须在 Ubuntu 的 etc/environments 中输入这些导出语句
回答by Estello
On my computer I had :
在我的电脑上，我有：
cudnn5.1,
cuda8.0
When importing tensorflow, I had the same error ImportError:
导入tensorflow时，出现同样的错误ImportError：
libcublas.so.8.0: cannot open shared object file: No such file or directory
libcublas.so.8.0：无法打开共享对象文件：没有这样的文件或目录
and even
乃至
libcublas.so.9.0: cannot open shared object file: No such file or directory
libcublas.so.9.0：无法打开共享对象文件：没有这样的文件或目录
To overcome this issue I installed tensorflow 1.4instead of 1.6
and did what Umair Qadir suggested.
为了解决这个问题，我安装tensorflow 1.4了 1.6 而不是 1.6 并按照 Umair Qadir 的建议做了。
But then I had another issue :
但后来我有另一个问题：
libcudnn.so.6: cannot open shared object file: No such file or directory
libcudnn.so.6：无法打开共享对象文件：没有这样的文件或目录
This was because I had cudnn5.1 as required but actually it needs cudnn6
这是因为我需要 cudnn5.1 但实际上它需要 cudnn6
Here are the steps to uninstall cudnn5.1 and install cudnn6:
以下是卸载cudnn5.1和安装cudnn6的步骤：
1- Uninstall cudnn 5.1 rm -f /usr/include/cudnn.h rm -f /usr/lib/x86_64-linux-gnu/libcudnn rm -f /usr/local/cuda-/lib64/libcudnn
1-卸载cudnn 5.1 rm -f /usr/include/cudnn.h rm -f /usr/lib/x86_64-linux-gnu/libcudnn rm -f /usr/local/cuda-/lib64/libcudnn
2- Install cudnn6
2- 安装 cudnn6
After having subscribed to nvdia, go to here https://developer.nvidia.com/rdp/cudnn-downloadand download cudnn6 for cuda8 and then go to the folder where you downloaded the cudnn and do :
订阅 nvdia 后，请转到此处https://developer.nvidia.com/rdp/cudnn-download并下载 cuda8 的 cudnn6，然后转到下载 cudnn 的文件夹并执行以下操作：
$ tar xvzf cudnn-8.0-linux-x64-v5.1-ga.tgz
$ sudo cp -P cuda/include/cudnn.h /usr/local/cuda/include
$ sudo cp -P cuda/lib64/libcudnn /usr/local/cuda/lib64
$ sudo chmod a+r /usr/local/cuda/include/cudnn.h
/usr/local/cuda/lib64/libcudnn*
Now you should have tensorflow
现在你应该有 tensorflow
Try it on typing in the console:
尝试在控制台中输入：
$python
import tensorflow
If you want to work in anaconda and the error persists , try :
如果您想在 anaconda 中工作并且错误仍然存在，请尝试：
$jupyter notebook --generate-config
then you can find the name of the directory where you have your config file (Ill call it
然后你可以找到你的配置文件所在目录的名称（我称之为
) and open /jupyter_notebook_config.py and add at the top :
) 并打开 /jupyter_notebook_config.py 并在顶部添加：
import os
c = get_config()
os.environ['LD_LIBRARY_PATH'] = '/usr/local/cuda-8.0/lib64:usr/local/cuda-8.0/lib64/libcudart.so.8.0'
c.Spawner.env.update('LD_LIBRARY_PATH')
Now it should work...
现在它应该工作...
Here are the websites that helped me:
以下是对我有帮助的网站：
https://askubuntu.com/questions/952075/how-to-upgrade-tensorflow-to-v1-3-cudnn-cuda-upgradehttps://developer.nvidia.com/rdp/cudnn-downloadhttps://medium.com/@ikekramer/installing-cuda-8-0-and-cudnn-5-1-on-ubuntu-16-04-6b9f284f6e77tensorflow gpu can not be called from jupyterhub/jupyter notebook, why?
https://askubuntu.com/questions/952075/how-to-upgrade-tensorflow-to-v1-3-cudnn-cuda-upgrade https://developer.nvidia.com/rdp/cudnn-download https:// medium.com/@ikekramer/installing-cuda-8-0-and-cudnn-5-1-on-ubuntu-16-04-6b9f284f6e77无法从 jupyterhub/jupyter notebook 调用 tensorflow gpu，为什么？
