Error installing TensorFlow GPU on Ubuntu

Question

So I've spent my day trying to install TF GPU, without luck.

I've installed CUDA Toolkit 9.0 and tried CuDNN v7.0.3 (for CUDA 8 and 9) and also the v6.0.

After trying to execute my demo program, the terminal throws this message:

Traceback (most recent call last):
File "/home/gemboly/anaconda3/envs/tensorflow/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow.py", line 58, in <module>
from tensorflow.python.pywrap_tensorflow_internal import *
File "/home/gemboly/anaconda3/envs/tensorflow/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 28, in <module>
_pywrap_tensorflow_internal = swig_import_helper()
File "/home/gemboly/anaconda3/envs/tensorflow/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 "/home/gemboly/anaconda3/envs/tensorflow/lib/python3.6/imp.py", line 243, in load_module
return load_dynamic(name, filename, file)
File "/home/gemboly/anaconda3/envs/tensorflow/lib/python3.6/imp.py", line 343, in load_dynamic
return _load(spec)
ImportError: libcublas.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 "twicetwo.py", line 5, in <module>
import tensorflow
File "/home/gemboly/anaconda3/envs/tensorflow/lib/python3.6/site-packages/tensorflow/__init__.py", line 24, in <module>
from tensorflow.python import *
File "/home/gemboly/anaconda3/envs/tensorflow/lib/python3.6/site-packages/tensorflow/python/__init__.py", line 49, in <module>
from tensorflow.python import pywrap_tensorflow
File "/home/gemboly/anaconda3/envs/tensorflow/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow.py", line 72, in <module>
raise ImportError(msg)
ImportError: Traceback (most recent call last):
File "/home/gemboly/anaconda3/envs/tensorflow/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow.py", line 58, in <module>
from tensorflow.python.pywrap_tensorflow_internal import *
File "/home/gemboly/anaconda3/envs/tensorflow/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 28, in <module>
_pywrap_tensorflow_internal = swig_import_helper()
File "/home/gemboly/anaconda3/envs/tensorflow/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 "/home/gemboly/anaconda3/envs/tensorflow/lib/python3.6/imp.py", line 243, in load_module
return load_dynamic(name, filename, file)
File "/home/gemboly/anaconda3/envs/tensorflow/lib/python3.6/imp.py", line 343, in load_dynamic
return _load(spec)
ImportError: libcublas.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've updated the LD_LIBRARY_PATH variable and also the PATH .

I'm using Anaconda, created a source with Python 3.6, since python --version threw this version.

Can anybody help me with this?

Answer 1

I've managed to run it. If you are here because you are stuck like me do the following:

• use CUDA 8.0 ( https://developer.nvidia.com/cuda-toolkit-archive ) for me only the runfile install worked!

• I have Python 3.6 installed, but when creating your environment you should use python="3.5" instead