如何将cudnn5.1升级到cudnn6
how to upgrade cudnn5.1 to cudnn6
我按照 Installing TensorFlow on Windows 安装了 tensorflow-gpu,当我导入 tensorflow 时它似乎运行良好,但今天当我尝试一个新程序时,它显示一些错误:
`
2017-07-12 09:33:31.301985: E c:\tf_jenkins\home\workspace\release-win\device\gpu\os\windows\tensorflow\stream_executor\cuda\cuda_dnn.cc:352] Loaded runtime CuDNN library: 6021 (compatibility version 6000) but source was compiled with 5105 (compatibility version 5100). If using a binary install, upgrade your CuDNN library to match. If building from sources, make sure the library loaded at runtime matches a compatible version specified during compile configuration.
2017-07-12 09:33:31.302851: F c:\tf_jenkins\home\workspace\release-win\device\gpu\os\windows\tensorflow\core\kernels\conv_ops.cc:659] Check failed: stream->parent()->GetConvolveAlgorithms(&algorithms)
`
我知道这意味着我应该将我的 cudnn5.1 升级到 cudnn6,我下载 cudnn6 并将文件复制到我的 cuda 安装目录。那么在cudn/v8.0/bin目录下存在两个cudnndll:cudnn64_5.dll和cudnn64_6.dll.
当我重新运行我的程序时,仍然出现同样的错误,当我删除cudnn64_5.dll,我什至不能运行我的程序时,出现了一些新的错误:
`
builtins.ImportError: Traceback (most recent call last):
File "c:\Users\chenc\Anaconda3\envs\tensorflow-gpu\lib\site-packages\tensorflow\python\pywrap_tensorflow_internal.py", line 18, in swig_import_helper
return importlib.import_module(mname)
File "c:\Users\chenc\Anaconda3\envs\tensorflow-gpu\lib\importlib\__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
`
我将我的 cudnn64_6.dll 重命名为 cudnn64_5..dll,上面的错误消失了,但仍然显示我的 cudnn 版本是 5105.It 似乎我必须做其他事情来制作cudnn 升级成功了,但我没有 know.What 我应该如何解决错误?
我不确定如何解决您当前的问题,但我确定您不应该更新到 cuDNN_6.0。 并且不会起作用。
我按照 Installing TensorFlow on Windows 安装了 tensorflow-gpu,当我导入 tensorflow 时它似乎运行良好,但今天当我尝试一个新程序时,它显示一些错误:
`
2017-07-12 09:33:31.301985: E c:\tf_jenkins\home\workspace\release-win\device\gpu\os\windows\tensorflow\stream_executor\cuda\cuda_dnn.cc:352] Loaded runtime CuDNN library: 6021 (compatibility version 6000) but source was compiled with 5105 (compatibility version 5100). If using a binary install, upgrade your CuDNN library to match. If building from sources, make sure the library loaded at runtime matches a compatible version specified during compile configuration.
2017-07-12 09:33:31.302851: F c:\tf_jenkins\home\workspace\release-win\device\gpu\os\windows\tensorflow\core\kernels\conv_ops.cc:659] Check failed: stream->parent()->GetConvolveAlgorithms(&algorithms)
`
我知道这意味着我应该将我的 cudnn5.1 升级到 cudnn6,我下载 cudnn6 并将文件复制到我的 cuda 安装目录。那么在cudn/v8.0/bin目录下存在两个cudnndll:cudnn64_5.dll和cudnn64_6.dll.
当我重新运行我的程序时,仍然出现同样的错误,当我删除cudnn64_5.dll,我什至不能运行我的程序时,出现了一些新的错误:
`
builtins.ImportError: Traceback (most recent call last):
File "c:\Users\chenc\Anaconda3\envs\tensorflow-gpu\lib\site-packages\tensorflow\python\pywrap_tensorflow_internal.py", line 18, in swig_import_helper
return importlib.import_module(mname)
File "c:\Users\chenc\Anaconda3\envs\tensorflow-gpu\lib\importlib\__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
`
我将我的 cudnn64_6.dll 重命名为 cudnn64_5..dll,上面的错误消失了,但仍然显示我的 cudnn 版本是 5105.It 似乎我必须做其他事情来制作cudnn 升级成功了,但我没有 know.What 我应该如何解决错误?
我不确定如何解决您当前的问题,但我确定您不应该更新到 cuDNN_6.0。