0

I recently upgraded my graphics card on my Linux machine. I installed NVIDIA driver versioned 384, cudnn and manually installed CUDA-Toolkit 8 to opt out the included old drivers.

I compiled the "NVIDIA_CUDA-8.0_Samples" and ran the binary file generated named "deviceQuery" it passed and worked fine.

I get the following error when trying to run my code:

Traceback (most recent call last):
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/pywrap_tensorflow.py", line 41, in <module>
    from tensorflow.python.pywrap_tensorflow_internal import *
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 28, in <module>
    _pywrap_tensorflow_internal = swig_import_helper()
  File "/usr/local/lib/python3.6/dist-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 242, in load_module
    return load_dynamic(name, filename, file)
  File "/usr/lib/python3.6/imp.py", line 342, in load_dynamic
    return _load(spec)
ImportError: libcusolver.so.8.0: 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 "/home/ben/Documents/GitHub/Machine-Learning/polynomial_fitting/poly_fit.py", line 5, in <module>
    import tensorflow as tf
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/__init__.py", line 24, in <module>
    from tensorflow.python import *
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/__init__.py", line 49, in <module>
    from tensorflow.python import pywrap_tensorflow
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/pywrap_tensorflow.py", line 52, in <module>
    raise ImportError(msg)
ImportError: Traceback (most recent call last):
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/pywrap_tensorflow.py", line 41, in <module>
    from tensorflow.python.pywrap_tensorflow_internal import *
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 28, in <module>
    _pywrap_tensorflow_internal = swig_import_helper()
  File "/usr/local/lib/python3.6/dist-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 242, in load_module
    return load_dynamic(name, filename, file)
  File "/usr/lib/python3.6/imp.py", line 342, in load_dynamic
    return _load(spec)
ImportError: libcusolver.so.8.0: cannot open shared object file: No such file or directory


Failed to load the native TensorFlow runtime.

See https://www.tensorflow.org/install/install_sources#common_installation_problems

for some common reasons and solutions.  Include the entire stack trace
above this error message when asking for help.
Error in sys.excepthook:
Traceback (most recent call last):
  File "/usr/lib/python3/dist-packages/apport_python_hook.py", line 63, in apport_excepthook
    from apport.fileutils import likely_packaged, get_recent_crashes
  File "/usr/lib/python3/dist-packages/apport/__init__.py", line 5, in <module>
    from apport.report import Report
  File "/usr/lib/python3/dist-packages/apport/report.py", line 30, in <module>
    import apport.fileutils
  File "/usr/lib/python3/dist-packages/apport/fileutils.py", line 23, in <module>
    from apport.packaging_impl import impl as packaging
  File "/usr/lib/python3/dist-packages/apport/packaging_impl.py", line 23, in <module>
    import apt
  File "/usr/lib/python3/dist-packages/apt/__init__.py", line 23, in <module>
    import apt_pkg
ModuleNotFoundError: No module named 'apt_pkg'

Original exception was:
Traceback (most recent call last):
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/pywrap_tensorflow.py", line 41, in <module>
    from tensorflow.python.pywrap_tensorflow_internal import *
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 28, in <module>
    _pywrap_tensorflow_internal = swig_import_helper()
  File "/usr/local/lib/python3.6/dist-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 242, in load_module
    return load_dynamic(name, filename, file)
  File "/usr/lib/python3.6/imp.py", line 342, in load_dynamic
    return _load(spec)
ImportError: libcusolver.so.8.0: 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 "/home/ben/Documents/GitHub/Machine-Learning/polynomial_fitting/poly_fit.py", line 5, in <module>
    import tensorflow as tf
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/__init__.py", line 24, in <module>
    from tensorflow.python import *
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/__init__.py", line 49, in <module>
    from tensorflow.python import pywrap_tensorflow
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/pywrap_tensorflow.py", line 52, in <module>
    raise ImportError(msg)
ImportError: Traceback (most recent call last):
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/pywrap_tensorflow.py", line 41, in <module>
    from tensorflow.python.pywrap_tensorflow_internal import *
  File "/usr/local/lib/python3.6/dist-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 28, in <module>
    _pywrap_tensorflow_internal = swig_import_helper()
  File "/usr/local/lib/python3.6/dist-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 242, in load_module
    return load_dynamic(name, filename, file)
  File "/usr/lib/python3.6/imp.py", line 342, in load_dynamic
    return _load(spec)
ImportError: libcusolver.so.8.0: cannot open shared object file: No such file or directory


Failed to load the native TensorFlow runtime.

See https://www.tensorflow.org/install/install_sources#common_installation_problems

for some common reasons and solutions.  Include the entire stack trace
above this error message when asking for help.

I tried following:

  1. http://blog.nelsonliu.me/2017/04/29/installing-and-updating-gtx-1080-ti-cuda-drivers-on-ubuntu/
  2. Tensorflow: ImportError: libcusolver.so.8.0: cannot open shared object file: No such file or directory
  3. After building TensorFlow from source, seeing libcudart.so and libcudnn errors

Yet no results. Currently I have in my "~/.bashrc" at the EOF the two following lines:

export LD_LIBRARY_PATH=/usr/local/cuda/lib64/
export LD_LIBRARY_PATH=/usr/local/cuda-8.0/lib64/

I do have the file "libcusolver.so.8.0" at the sub-dir "lib64", so I guess it's a linking problem. DISCLAIMER: I am relatively a new user to Linux and perhaps and I don't possess the knowledge about "soft links, the full purpose of ".bashrc", "etc/environment" and such.

Thanks in advance, Ben.

talonmies
  • 70,661
  • 34
  • 192
  • 269
Ben
  • 152
  • 4
  • 13

2 Answers2

3

For Googlers, you could be having this problem if you installed the latest version of CUDA (9.0). If this is the case, you'll have the libcusolver.so.9.0 file or directory on your computer, but not 8.0, so python can't find it to import. Tensorflow doesn't yet support CUDA 9.0 (as of Oct 18th 2017).

To fix this, install CUDA 8.0, which can be found here. You can find all their legacy releases at the bottom of the main download page.

If you choose to download the deb (local) installer, the instructions are as follows:

  1. download the base installer from the link above
  2. cd to the directory where it downloaded to and run sudo dpkg -i cuda-repo-ubuntu1604-8-0-local-ga2_8.0.61-1_amd64.deb
  3. sudo apt-get update
  4. sudo apt-get install cuda-8-0 (The website says install cuda, but that will just install 9.0, you need to add -8-0 for it to install version 8.0)
Davidjb
  • 1,190
  • 1
  • 12
  • 19
  • Thank you! I can't tell if this directly solved the problem I had when I asked the question but everything works for me now. I have since reinstalled Ubuntu 16.04 and failed to compile Tensorflow, I think it was incompatibility between Tensorflow r1.3 and Cuda 9-0. I didn't need Tensorflow during this time so I didn't bother falling for the same issue I had if I would've installed Cuda-8-0. I have just successfully installed Tensorflow r1.4 from source. I am using Geforce GTX 1080 Ti. Ill mark this as the answer. Thanks again. – Ben Dec 01 '17 at 13:23
0

lots of things to check

  • check your env variables LD_LIBRARY_PATH and LD_LIBRARY_PATH are exactly what you want by echo $LD_LIBRARY_PATH
  • check your driver by nvidia-smi
Yuwen Yan
  • 4,777
  • 10
  • 33
  • 63
  • Hello, Thanks and sorry for the late reply. Indeed 'nvidia-smi' yields the current driver version I have, 384.59 and also running 'echo $LD_LIBRARY_PATH' gives "/usr/local/cuda-8.0/lib64/" as needed. Yet it still doesn't work. I might just try a fresh installation of Ubuntu as a clean solution since it's frustrating to solve. Ben. – Ben Jul 30 '17 at 10:03