J'ai installé GPU tensorflow avec Nvidia CUDA sur mac book pro 2012mid (Nvidia GT 650M)Raison: image not found GPU tensorflow
Mais quand je lance le programme simple tensorflow, image not found
erreur se produit.
J'ai parcouru le site et j'ai trouvé cette erreur liée à l'installation de cuda.
[[[Ce que j'ai vérifié sont]]]
libcudart.8,0.dylib est ici
/usr/local/cuda/lib/libcudart.8.0.dylib
et réglage de l'environnement semble propery.
(tensorflow) koala:tensorWorkspace daichi$ echo $LD_LIBRARY_PATH
/usr/local/cuda/lib:
(tensorflow) koala:tensorWorkspace daichi$ echo $CUDA_HOME
/usr/local/cuda
Y a-t-il un endroit où je devrais vérifier ???
Le journal des erreurs est ci-dessous.
(tensorflow) koala:tensorWorkspace whitebear$ python3 hello.py
Traceback (most recent call last):
File "/Users/whitebear/.pyenv/versions/3.5.2/lib/python3.5/site-packages/tensorflow/python/pywrap_tensorflow.py", line 41, in <module>
from tensorflow.python.pywrap_tensorflow_internal import *
File "/Users/whitebear/.pyenv/versions/3.5.2/lib/python3.5/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 28, in <module>
_pywrap_tensorflow_internal = swig_import_helper()
File "/Users/whitebear/.pyenv/versions/3.5.2/lib/python3.5/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 "/Users/whitebear/.pyenv/versions/3.5.2/lib/python3.5/imp.py", line 242, in load_module
return load_dynamic(name, filename, file)
File "/Users/whitebear/.pyenv/versions/3.5.2/lib/python3.5/imp.py", line 342, in load_dynamic
return _load(spec)
ImportError: dlopen(/Users/whitebear/.pyenv/versions/3.5.2/lib/python3.5/site-packages/tensorflow/python/_pywrap_tensorflow_internal.so, 10): Library not loaded: @rpath/libcublas.8.0.dylib
Referenced from: /Users/whitebear/.pyenv/versions/3.5.2/lib/python3.5/site-packages/tensorflow/python/_pywrap_tensorflow_internal.so
Reason: image not found
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "hello.py", line 1, in <module>
import tensorflow as tf
File "/Users/whitebear/.pyenv/versions/3.5.2/lib/python3.5/site-packages/tensorflow/__init__.py", line 24, in <module>
from tensorflow.python import *
File "/Users/whitebear/.pyenv/versions/3.5.2/lib/python3.5/site-packages/tensorflow/python/__init__.py", line 51, in <module>
from tensorflow.python import pywrap_tensorflow
File "/Users/whitebear/.pyenv/versions/3.5.2/lib/python3.5/site-packages/tensorflow/python/pywrap_tensorflow.py", line 52, in <module>
raise ImportError(msg)
ImportError: Traceback (most recent call last):
File "/Users/whitebear/.pyenv/versions/3.5.2/lib/python3.5/site-packages/tensorflow/python/pywrap_tensorflow.py", line 41, in <module>
from tensorflow.python.pywrap_tensorflow_internal import *
File "/Users/whitebear/.pyenv/versions/3.5.2/lib/python3.5/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 28, in <module>
_pywrap_tensorflow_internal = swig_import_helper()
File "/Users/whitebear/.pyenv/versions/3.5.2/lib/python3.5/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 "/Users/whitebear/.pyenv/versions/3.5.2/lib/python3.5/imp.py", line 242, in load_module
return load_dynamic(name, filename, file)
File "/Users/whitebear/.pyenv/versions/3.5.2/lib/python3.5/imp.py", line 342, in load_dynamic
return _load(spec)
ImportError: dlopen(/Users/whitebear/.pyenv/versions/3.5.2/lib/python3.5/site-packages/tensorflow/python/_pywrap_tensorflow_internal.so, 10): Library not loaded: @rpath/libcublas.8.0.dylib
Referenced from: /Users/whitebear/.pyenv/versions/3.5.2/lib/python3.5/site-packages/tensorflow/python/_pywrap_tensorflow_internal.so
Reason: image not found
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.
La bibliothèque à l'origine du problème est CUBLAS. Avez-vous vérifié qu'il est installé et dans un chemin où tensorflow peut le trouver? – talonmies
Merci pour votre réponse. CUBLAS est une boîte à outils cuda ??? J'ai installé boîte à outils cuda et mon nvcc --version affiche le message comme celui-ci 'nvcc --version nvcc: NVIDIA (R) du compilateur Cuda Copyright (c) 2005-2016 NVIDIA Corporation Construit sur Tue_Jan_10_13: 22: 46_CST_2017 Outils de compilation Cuda, version 8.0, V8.0.61' – whitebear
et chemin LD_LIBRARY_PATH, CUDA_HOME est correctement défini. – whitebear