Tensorflow-GPU unable to import

Question

I am trying to use Hypergan. When i Import tensorflow i get the following error message: I am using CUDA V8 on Windows 10 with CudNN V5.1 . I moved the the Cudnn files to the acording Cuda Folders and added the Path Variables. I installed the Visual C++ Redistuable 2017 aswell.

Any Clue where the error is? Greetings Tobias

import tensorflow
Traceback (most recent call last):

  File "<ipython-input-1-a649b509054f>", line 1, in <module>
    import tensorflow

  File "D:\Programme\Anaconda\envs\Deep\lib\site-packages\tensorflow\__init__.py", line 24, in <module>
    from tensorflow.python import *

  File "D:\Programme\Anaconda\envs\Deep\lib\site-packages\tensorflow\python\__init__.py", line 49, in <module>
    from tensorflow.python import pywrap_tensorflow

  File "D:\Programme\Anaconda\envs\Deep\lib\site-packages\tensorflow\python\pywrap_tensorflow.py", line 52, in <module>
    raise ImportError(msg)

ImportError: Traceback (most recent call last):
  File "D:\Programme\Anaconda\envs\Deep\lib\site-packages\tensorflow\python\pywrap_tensorflow_internal.py", line 18, in swig_import_helper
    return importlib.import_module(mname)
  File "D:\Programme\Anaconda\envs\Deep\lib\importlib\__init__.py", line 126, in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
  File "<frozen importlib._bootstrap>", line 985, in _gcd_import
  File "<frozen importlib._bootstrap>", line 968, in _find_and_load
  File "<frozen importlib._bootstrap>", line 957, in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 666, in _load_unlocked
  File "<frozen importlib._bootstrap>", line 577, in module_from_spec
  File "<frozen importlib._bootstrap_external>", line 938, in create_module
  File "<frozen importlib._bootstrap>", line 222, in _call_with_frames_removed
ImportError: DLL load failed: Das angegebene Modul wurde nicht gefunden.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "D:\Programme\Anaconda\envs\Deep\lib\site-packages\tensorflow\python\pywrap_tensorflow.py", line 41, in <module>
    from tensorflow.python.pywrap_tensorflow_internal import *
  File "D:\Programme\Anaconda\envs\Deep\lib\site-packages\tensorflow\python\pywrap_tensorflow_internal.py", line 21, in <module>
    _pywrap_tensorflow_internal = swig_import_helper()
  File "D:\Programme\Anaconda\envs\Deep\lib\site-packages\tensorflow\python\pywrap_tensorflow_internal.py", line 20, in swig_import_helper
    return importlib.import_module('_pywrap_tensorflow_internal')
  File "D:\Programme\Anaconda\envs\Deep\lib\importlib\__init__.py", line 126, in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
ImportError: No module named '_pywrap_tensorflow_internal'


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.

Answer 1

TL;DR : With Tensorflow 1.3.0 pre-built use cuDNN 6, not cuDNN 5.1

I assume you are using the latest binaries released for the 1.3.0 setup. our problem is that you're using cuDNN 5.1 when you should use cuDNN 6, because since TF 1.3.0 they build against that one instead. As stated in this GitHub issue discussion , the websites were out of date and the installation guide told you to use the wrong cuDNN version. The issue should be fixed now, though.