pyinstaller 總結：No module named 'tensorflow.python._pywrap_tensorflow_internal

pyinstaller -F  -c --i favicon.ico Archives-image-processor-v2.0.py --add-data="conf;conf"  --hidden-import pywt._extensions._cwt

cpu版本的tensorflow可行

Q1:gpu版本的會出現：pyinstaller：No module named 'tensorflow.python._pywrap_tensorflow_internal'

方法一：tensorflow-gpu1.9.0:

把tensorflow全部拷貝出來打包進去，親測可行，可以GPU跑！C:\ProgramData\Anaconda3\envs\tensorflow\Lib\site-packages\tensorflow拷貝到打包目錄。

pyinstaller --path C:\ProgramData\Anaconda3\envs\tensorflowpy367\Lib\site-packages\PyQt5\Qt\bin -F  -c --i favicon.ico Archives-image-processor-v2.0.py --add-data="conf;conf" --add-data="tensorflow;tensorflow"  --hidden-import pywt._extensions._cwt

方法二：有人說需要降級tensorflow-gpu，親測只能CPU跑

所以安裝了pip install tensorflow-gpu==1.6.0

 

後來顯示沒有pyqt，所以把pyqt路徑添加進去

pyinstaller --path C:\ProgramData\Anaconda3\envs\tensorflowpy367\Lib\site-packages\PyQt5\Qt\bin -F  -c --i favicon.ico Archives-image-processor-v2.0.py --add-data="conf;conf"  --hidden-import pywt._extensions._cwt

可以運行，但是cudnn版本報錯，很奇怪

self.angle: 0.0
2019-10-23 11:17:27.423269: E C:\tf_jenkins\workspace\rel-win\M\windows-gpu\PY\36\tensorflow\stream_executor\cuda\cuda_dnn.cc:378] Loaded runtime CuDNN library: 7603 (compatibility version 7600) but source was compiled with 7003 (compatibility version 7000).  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.
2019-10-23 11:17:27.430828: F C:\tf_jenkins\workspace\rel-win\M\windows-gpu\PY\36\tensorflow\core\kernels\conv_ops.cc:717] Check failed: stream->parent()->GetConvolveAlgorithms( conv_parameters.ShouldIncludeWinogradNonfusedAlgo<T>(), &algorithms)

之前tensorflow1.9用的是7.6.5，現在tensorflow-gpu1.8.0需要CUDnn降級7.0.5：https://developer.nvidia.com/rdp/cudnn-archive

重裝CUDA9.0