Home / docs / ImportError: libcublas.so.9.0: cannot open shared object file: No such file or directory

ImportError: libcublas.so.9.0: cannot open shared object file: No such file or directory

ImportError: libcublas.so.9.0: cannot open shared object file: No such file or directory

When trying to import tensorflow-gpu, I was getting error:  ImportError: libcublas.so.9.0: cannot open shared object file: No such file or directory .

 

Diagnosis:

This error is due to the version of tensorflow-gpu installed that is built with a different version of cuda and cudnn and that compatible version of CUDA and CUDNN is not installed in your machine.

The tensorflow website has given a chart mentioning compatible versions of CUDA and CUDNN with tensorflow-gpu.

Hence when installing tensorflow-gpu, first step is to check the GPU model that is installed in your computer and then to check which latest version of CUDA and CUDNN is compatible with your hardware GPU.

How to check Cuda Version compatible with installed GPU

 

Once, you know the CUDA version and CUDNN version, you can choose which version of tensorflow-gpu is compatible with your hardware.

The tensorflow-gpu compatibility with different versions of CUDA and CUDNN are given as:

Linux

VersionPython versionCompilerBuild toolscuDNNCUDA
tensorflow_gpu-1.12.02.7, 3.3-3.6GCC 4.8Bazel 0.15.079
tensorflow_gpu-1.11.02.7, 3.3-3.6GCC 4.8Bazel 0.15.079
tensorflow_gpu-1.10.02.7, 3.3-3.6GCC 4.8Bazel 0.15.079
tensorflow_gpu-1.9.02.7, 3.3-3.6GCC 4.8Bazel 0.11.079
tensorflow_gpu-1.8.02.7, 3.3-3.6GCC 4.8Bazel 0.10.079
tensorflow_gpu-1.7.02.7, 3.3-3.6GCC 4.8Bazel 0.9.079
tensorflow_gpu-1.6.02.7, 3.3-3.6GCC 4.8Bazel 0.9.079
tensorflow_gpu-1.5.02.7, 3.3-3.6GCC 4.8Bazel 0.8.079
tensorflow_gpu-1.4.02.7, 3.3-3.6GCC 4.8Bazel 0.5.468
tensorflow_gpu-1.3.02.7, 3.3-3.6GCC 4.8Bazel 0.4.568
tensorflow_gpu-1.2.02.7, 3.3-3.6GCC 4.8Bazel 0.4.55.18
tensorflow_gpu-1.1.02.7, 3.3-3.6GCC 4.8Bazel 0.4.25.18
tensorflow_gpu-1.0.02.7, 3.3-3.6GCC 4.8Bazel 0.4.25.18

Updated: 12-18-2018 [source]

Windows:

VersionPython versionCompilerBuild toolscuDNNCUDA
tensorflow_gpu-1.12.03.5-3.6MSVC 2015 update 3Bazel 0.15.079
tensorflow_gpu-1.11.03.5-3.6MSVC 2015 update 3Bazel 0.15.079
tensorflow_gpu-1.10.03.5-3.6MSVC 2015 update 3Cmake v3.6.379
tensorflow_gpu-1.9.03.5-3.6MSVC 2015 update 3Cmake v3.6.379
tensorflow_gpu-1.8.03.5-3.6MSVC 2015 update 3Cmake v3.6.379
tensorflow_gpu-1.7.03.5-3.6MSVC 2015 update 3Cmake v3.6.379
tensorflow_gpu-1.6.03.5-3.6MSVC 2015 update 3Cmake v3.6.379
tensorflow_gpu-1.5.03.5-3.6MSVC 2015 update 3Cmake v3.6.379
tensorflow_gpu-1.4.03.5-3.6MSVC 2015 update 3Cmake v3.6.368
tensorflow_gpu-1.3.03.5-3.6MSVC 2015 update 3Cmake v3.6.368
tensorflow_gpu-1.2.03.5-3.6MSVC 2015 update 3Cmake v3.6.35.18
tensorflow_gpu-1.1.03.5MSVC 2015 update 3Cmake v3.6.35.18
tensorflow_gpu-1.0.03.5MSVC 2015 update 3Cmake v3.6.35.18

 

MacOS:

VersionPython versionCompilerBuild toolscuDNNCUDA
tensorflow_gpu-1.1.02.7, 3.3-3.6Clang from xcodeBazel 0.4.25.18
tensorflow_gpu-1.0.02.7, 3.3-3.6Clang from xcodeBazel 0.4.25.18

 

If you have installed CUDA 8 and CUDNN 5.1, then the table above for Linux shows: tensorflow_gpu-1.2

To install tensorflow-gpu 1.2:

 

Related Links:

Check Also

Smart Messenger with Self Reminder app doesnt show notifications on time

Avoid apps from being removed from Recent Apps List

To save the RAM, apps are usually being removed from Recent Apps List, but with …

Leave a Reply

Your email address will not be published. Required fields are marked *

%d bloggers like this: