Tensoflow-GPU 安装问题



我已经使用在CPU上运行的tensorflow几天了,但我决定开始使用GPU。因此,我通过pip安装了CUDA和tensorflow-gpu,但它不起作用,所以我卸载了两个版本的tensorflow,并再次安装了tensorflow-gpu。出现以下消息:

Traceback (most recent call last):
File "C:Python35libsite-
packagestensorflowpythonpywrap_tensorflow_internal.py", line 18, in 
swig_import_helper
return importlib.import_module(mname)
File "C:Python35libimportlib__init__.py", line 126, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
File "<frozen importlib._bootstrap>", line 986, in _gcd_import
File "<frozen importlib._bootstrap>", line 969, in _find_and_load
File "<frozen importlib._bootstrap>", line 958, 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 906, in create_module
File "<frozen importlib._bootstrap>", line 222, in _call_with_frames_removed
ImportError: DLL load failed: The specified module could not be found.
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
  File "C:Python35libsite-
packagestensorflowpythonpywrap_tensorflow.py", line 41, in <module>
from tensorflow.python.pywrap_tensorflow_internal import *
File "C:Python35libsite-
packagestensorflowpythonpywrap_tensorflow_internal.py", line 21, in 
<module>
_pywrap_tensorflow_internal = swig_import_helper()
File "C:Python35libsite-
packagestensorflowpythonpywrap_tensorflow_internal.py", line 20, in 
swig_import_helper
return importlib.import_module('_pywrap_tensorflow_internal')
File "C:Python35libimportlib__init__.py", line 126, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
ImportError: No module named '_pywrap_tensorflow_internal'
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 "C:Python35libsite-packagestensorflow__init__.py", line 24, in 
<module>
from tensorflow.python import *
File "C:Python35libsite-packagestensorflowpython__init__.py", line 51, 
in <module>
from tensorflow.python import pywrap_tensorflow
File "C:Python35libsite-packagestensorflowpythonpywrap_tensorflow.py", 
line 52, in <module>
raise ImportError(msg)
ImportError: Traceback (most recent call last):
File "C:Python35libsite-
packagestensorflowpythonpywrap_tensorflow_internal.py", line 18, in 
swig_import_helper
return importlib.import_module(mname)
File "C:Python35libimportlib__init__.py", line 126, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
File "<frozen importlib._bootstrap>", line 986, in _gcd_import
File "<frozen importlib._bootstrap>", line 969, in _find_and_load
File "<frozen importlib._bootstrap>", line 958, 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 906, in create_module
File "<frozen importlib._bootstrap>", line 222, in _call_with_frames_removed
ImportError: DLL load failed: The specified module could not be found.
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "C:Python35libsite-packagestensorflowpythonpywrap_tensorflow.py", 
line 41, in <module>
from tensorflow.python.pywrap_tensorflow_internal import *
File "C:Python35libsite-
packagestensorflowpythonpywrap_tensorflow_internal.py", line 21, in 
<module>
_pywrap_tensorflow_internal = swig_import_helper()
File "C:Python35libsite-
packagestensorflowpythonpywrap_tensorflow_internal.py", line 20, in 
swig_import_helper
return importlib.import_module('_pywrap_tensorflow_internal')
File "C:Python35libimportlib__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.

以前有人遇到过类似的错误吗?

  1. cuDNN v6.0 对很多人不起作用。尝试使用 cuDNN v5.1。您正在使用的 cuDNN 版本是什么?

  2. 从 GIT 文件夹内部导入也是原因之一。

我建议您在单独的虚拟环境中同时安装 CPU 和 GPU 版本。你可以使用python的内置虚拟环境,或者我的偏好是anaconda。保持两个版本彼此独立应该可以解决问题。

最新更新