如何在 Pycharm 中运行 TensorFlow GPU？

Question

我想在 Linux Mint 上的 Pycharm 中运行 Tensorflow GPU。

我尝试了一些类似的指南

https://medium.com/@p.venkata.kishore/install-anaconda-tenserflow-gpu-keras-and-pycharm-on-windows-10-6bfb39630e4e

https://medium.com/@kekayan/step-by-step-guide-to-install-tensorflow-gpu-on-ubuntu-18-04-lts-6feceb0df5c0

我运行这段代码

import tensorflow as tf


sess = tf.Session(config=tf.ConfigProto(log_device_placement=True))

我遇到了这个错误

Traceback (most recent call last):
  File "/home/alex/anaconda3/envs/TfTestGPU/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow.py", line 58, in <module>
    from tensorflow.python.pywrap_tensorflow_internal import *
  File "/home/alex/anaconda3/envs/TfTestGPU/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 28, in <module>
    _pywrap_tensorflow_internal = swig_import_helper()
  File "/home/alex/anaconda3/envs/TfTestGPU/lib/python3.6/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 "/home/alex/anaconda3/envs/TfTestGPU/lib/python3.6/imp.py", line 243, in load_module
    return load_dynamic(name, filename, file)
  File "/home/alex/anaconda3/envs/TfTestGPU/lib/python3.6/imp.py", line 343, in load_dynamic
    return _load(spec)
ImportError: libcublas.so.10.0: cannot open shared object file: No such file or directory

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/alex/PycharmProjects/TfTestGPU/test.py", line 1, in <module>
    import tensorflow as tf
  File "/home/alex/anaconda3/envs/TfTestGPU/lib/python3.6/site-packages/tensorflow/__init__.py", line 24, in <module>
    from tensorflow.python import pywrap_tensorflow  # pylint: disable=unused-import
  File "/home/alex/anaconda3/envs/TfTestGPU/lib/python3.6/site-packages/tensorflow/python/__init__.py", line 49, in <module>
    from tensorflow.python import pywrap_tensorflow
  File "/home/alex/anaconda3/envs/TfTestGPU/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow.py", line 74, in <module>
    raise ImportError(msg)
ImportError: Traceback (most recent call last):
  File "/home/alex/anaconda3/envs/TfTestGPU/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow.py", line 58, in <module>
    from tensorflow.python.pywrap_tensorflow_internal import *
  File "/home/alex/anaconda3/envs/TfTestGPU/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow_internal.py", line 28, in <module>
    _pywrap_tensorflow_internal = swig_import_helper()
  File "/home/alex/anaconda3/envs/TfTestGPU/lib/python3.6/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 "/home/alex/anaconda3/envs/TfTestGPU/lib/python3.6/imp.py", line 243, in load_module
    return load_dynamic(name, filename, file)
  File "/home/alex/anaconda3/envs/TfTestGPU/lib/python3.6/imp.py", line 343, in load_dynamic
    return _load(spec)
ImportError: libcublas.so.10.0: cannot open shared object file: No such file or directory


Failed to load the native TensorFlow runtime.

See https://www.tensorflow.org/install/errors

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

Answer 1

首先确保 CUDA 和 CuDNN 已成功安装并验证配置。

CUDA 驱动程序版本对于 CUDA 运行时版本应该足够了。

完成后， 打开 PyCharm

转到文件->设置->项目解释器

选择安装了 tensorflow-gpu 的合适环境

选择运行->编辑配置->环境变量

由于代码正在搜索libcublas.so.10.0，

假设您发现“libcublas.so.10.0”的路径类似于 "/home/Alex/anaconda3/pkgs/cudatoolkit- 10.0.130-0/lib/"

在环境变量中添加lib路径为LD_LIBRARY_PATH

名称：LD_LIBRARY_PATH

值：/home/Alex/anaconda3/pkgs/cudatoolkit-10.0.130-0/lib/

保存然后尝试import tensorflow

Answer 2

你这期有点迷糊，你是在pycharm还是anaconda中安装tensorflow，因为你使用pycharm，但是anaconda有一些错误。

1. 检查环境变量配置，两者都是Linux 和pycharm。小心路径中的 cuda-x。 x 是 cuda 的版本，例如 10.0
2. 检查tensorflow、cuda、cudnn的版本，根据 到this site。
3. 确保您可以在此文件夹中找到 libcublas.so.10.0 /usr/local/cuda-10.0/lib64。如果没有，您应该重新安装 cudnn。
4. 有时会遇到这个问题，但是重启后 电脑，它可以工作。