PyCharm 远程解释器和 Tensorflow -> 无法导入 Cudart.so

Question

我正在使用 PyCharm 进行远程调试，客户端 Win10，服务器 Ubuntu 16。在 Ubuntu 机器上，我正确安装了 Cuda 8.0，以及 python 2.7 和 tensorflow 1.0.0。 TensorFlow 在服务器上完美运行（ssh + 调用 python、导入 tensorflow 等）。

在 PyCharm 中，我为我的服务器机器设置了一个远程调试器，但现在导入 tensorflow 会导致错误：

Traceback (most recent call last):
  File "/home/*****/*****/********/*******.py", line 24, in <module>
    import tensorflow as tf
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/__init__.py", line 24, in <module>
    from tensorflow.python import *
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/__init__.py", line 60, in <module>
    raise ImportError(msg)
ImportError: Traceback (most recent call last):
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/__init__.py", line 49, in <module>
    from tensorflow.python import pywrap_tensorflow
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/pywrap_tensorflow.py", line 28, in <module>
    _pywrap_tensorflow = swig_import_helper()
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/pywrap_tensorflow.py", line 24, in swig_import_helper
    _mod = imp.load_module('_pywrap_tensorflow', fp, pathname, description)
ImportError: libcudart.so.8.0: cannot open shared object file: No such file or directory


Error importing tensorflow.  Unless you are using bazel,
you should not try to import tensorflow from its source directory;
please exit the tensorflow source tree, and relaunch your python interpreter
from there.

我试过了：

为 sys.path 变量添加正确的路径

将 LD_LIBRARY_PATH 添加到 os.environ 和/或 PyCharm 环境变量

没有成功！

我记得前段时间我能够毫无问题地进行调试，但我不知道出了什么问题。有什么建议吗？

更新 1 重启后，错误变成了这个

ssh://****@192.168.***.***:22/usr/bin/python -u /home/*****/.pycharm_helpers/pydev/pydevd.py --multiproc --qt-support --client '0.0.0.0' --port 35523 --file /home/*/*/*/*/test.py
warning: Debugger speedups using cython not found. Run '"/usr/bin/python" "/home/*/.pycharm_helpers/pydev/setup_cython.py" build_ext --inplace' to build.
pydev debugger: process 1913 is connecting

Connected to pydev debugger (build 163.10154.50)
Traceback (most recent call last):
  File "/home/*/.pycharm_helpers/pydev/pydevd.py", line 1596, in <module>
    globals = debugger.run(setup['file'], None, None, is_module)
  File "/home/*/.pycharm_helpers/pydev/pydevd.py", line 974, in run
    pydev_imports.execfile(file, globals, locals)  # execute the script
  File "/home/*/*/*/*/test.py", line 1, in <module>
    import tensorflow
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/__init__.py", line 24, in <module>
    from tensorflow.python import *
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/__init__.py", line 72, in <module>
    raise ImportError(msg)
ImportError: Traceback (most recent call last):
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/__init__.py", line 61, in <module>
    from tensorflow.python import pywrap_tensorflow
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/pywrap_tensorflow.py", line 28, in <module>
    _pywrap_tensorflow = swig_import_helper()
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/pywrap_tensorflow.py", line 24, in swig_import_helper
    _mod = imp.load_module('_pywrap_tensorflow', fp, pathname, description)
ImportError: libcudart.so.8.0: cannot open shared object file: No such file or directory


Failed to load the native TensorFlow runtime.

See https://github.com/tensorflow/tensorflow/blob/master/tensorflow/g3doc/get_started/os_setup.md#import_error

for some common reasons and solutions.  Include the entire stack trace
above this error message when asking for help.

Process finished with exit code 1

更新 2 / 已解决 正如我之前尝试过的，解决方案是在 pycharm 上将 LD_LIBRARY_PATH 添加到配置面板上的环境变量中。我认为在名称（-> 'LD_LIBRARY_PATH'）之后的空格不起作用之前。这里的解决方案： solution

Answer 1

解决方法是在Pycharm环境面板中添加正确的路径：

LD_LIBRARY_PATH='/usr/local/cuda/lib64'

或适合您配置的那个。

Answer 2

您收到的错误消息是“请退出 tensorflow 源代码树，然后从那里重新启动您的 python 解释器。”。

如果您下载 tensorflow 源，手动编译等安装它，然后启动 python 并导入 tensorflow，但仍在源文件夹中，通常会收到此错误消息。

您可以尝试阻止此错误的方法： - 检查 tensorflow 源是否在您运行远程应用程序的文件夹中（您是否按照类似的方法使其工作？http://www.pinchofintelligence.com/faster-writing-and-testing-machine-learning-applications/） - 尝试查找并删除 tensorflow 源文件夹 - 重新启动您的电脑（这会清除您可能正在其中运行您的东西的 /tmp 文件夹）

希望其中一项能奏效，让我知道！

Answer 3

远程 nvidia-docker 案例

我花了几个小时来解决类似的问题。实际上，我多次阅读 stackoverflow 并毫不费力地添加了所有这些环境变量。

因此，如果您有远程 nvidia-docker 并打开了 ssh，请阅读此线程： https://github.com/bethgelab/docker/issues/5

当您使用 ssh 连接时，nvidia-docker 无法正确启动终端。

Pycharm实际上是使用ssh连接进行远程调试，所以也无法启动tensorflow-gpu