Python 阻止 TensorFlow 访问 GPU？
声明：本页面是StackOverFlow热门问题的中英对照翻译，遵循CC BY-SA 4.0协议，如果您需要使用它，必须同样遵循CC BY-SA许可，注明原文地址和作者信息，同时你必须将它归于原作者（不是我）：StackOverFlow
原文地址: http://stackoverflow.com/questions/44552585/
Warning: these are provided under cc-by-sa 4.0 license. You are free to use/share it, But you must attribute it to the original authors (not me):
StackOverFlow
Prevent TensorFlow from accessing the GPU?
提问by jasekp
Is there a way to run TensorFlow purely on the CPU. All of the memory on my machine is hogged by a separate process running TensorFlow. I have tried setting the per_process_memory_fraction to 0, unsuccessfully.
有没有办法纯粹在 CPU 上运行 TensorFlow。我机器上的所有内存都被一个运行 TensorFlow 的单独进程占用。我曾尝试将 per_process_memory_fraction 设置为 0，但未成功。
回答by pfm
Have a look to this questionor this answer.
To summarise you can add this piece of code:
总而言之，您可以添加这段代码：
import os
os.environ["CUDA_VISIBLE_DEVICES"] = "-1"
import tensorflow as tf
Playing with the CUDA_VISIBLE_DEVICESenvironment variable is one of if not the wayto go whenever you have GPU-tensorflow installed and you don't want to use any GPUs.
与播放CUDA_VISIBLE_DEVICES环境变量是一个如果没有办法去，只要你有安装GPU-tensorflow，你不希望使用任何的GPU。
You to want either
export CUDA_VISIBLE_DEVICES=or alternatively use a virtualenv with a non-GPU installation of TensorFlow.
您想要
export CUDA_VISIBLE_DEVICES=或选择将 virtualenv 与 TensorFlow 的非 GPU 安装一起使用。
回答by MZHm
You can use only CPUs by openning a session with a GPU limit of 0:
您可以通过打开 GPU 限制为 0 的会话来仅使用 CPU：
sess = tf.Session(config=tf.ConfigProto(device_count={'GPU': 0}))
See https://www.tensorflow.org/api_docs/python/tf/ConfigProtofor more details.
有关更多详细信息，请参阅https://www.tensorflow.org/api_docs/python/tf/ConfigProto。
A proof that it works for @Nicolas:
证明它适用于@Nicolas：
In Python, write:
在 Python 中，编写：
import tensorflow as tf
sess_cpu = tf.Session(config=tf.ConfigProto(device_count={'GPU': 0}))
Then in a terminal:
然后在终端中：
nvidia-smi
You will see something like:
你会看到类似的东西：
+-----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| 0 24869 C /.../python 99MiB |
+-----------------------------------------------------------------------------+
Then repeat the process: In Python, write:
然后重复这个过程： 在 Python 中，写：
import tensorflow as tf
sess_gpu = tf.Session()
Then in a terminal:
然后在终端中：
nvidia-smi
You will see something like:
你会看到类似的东西：
+-----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| 0 25900 C /.../python 5775MiB |
+-----------------------------------------------------------------------------+
