当设备设置为CPU时,为什么TensorFlow使用我的GPU

时间:2018-02-09 08:46:38

标签: c++ tensorflow memory-management gpu cpu

TensorFlow正在使用分配我的所有GPU内存并忽略我的命令来使用CPU,我该如何解决这个问题呢?

这是我的testprog

的代码摘录
Session *session;
SessionOptions opts = SessionOptions();

//force to allocate 0 memory on gpu
opts.config.mutable_gpu_options()->set_per_process_gpu_memory_fraction(0);
opts.config.mutable_gpu_options()->set_allow_growth(false);

//create session with these settings
TF_CHECK_OK(NewSession(opts, &session));
TF_CHECK_OK(session->Create(graph_def));

//set device to cpu
graph::SetDefaultDevice("/cpu:0", &graph_def);

//run arbitrary model
Status status = session->Run(classifierInput, {output_layer},{},&outputs);

TF_CHECK_OK(session->Close());

致电nvidi-smi告诉我:

+-----------------------------------------------------------------------------+
| NVIDIA-SMI 375.66                 Driver Version: 375.66                    |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  Quadro P4000        Off  | 0000:01:00.0     Off |                  N/A |
| N/A   50C    P0    28W /  N/A |   7756MiB /  8114MiB |     42%      Default |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID  Type  Process name                               Usage      |
|=============================================================================|
|    0      1784    G   /usr/bin/X                                     139MiB |
|    0      3828    G   qtcreator                                       28MiB |
|    0      7721    C   ...testprog/build/testprog                    7585MiB |
+-----------------------------------------------------------------------------+

为什么会这样?

2 个答案:

答案 0 :(得分:6)

因为这个问题用C ++标记。解决方案是

tensorflow::Session *sess;
tensorflow::SessionOptions options;

tensorflow::ConfigProto* config = &options.config;
// disabled GPU entirely
(*config->mutable_device_count())["GPU"] = 0;
// place nodes somewhere
config->set_allow_soft_placement(true);

请参阅example here。 我的另一篇文章how TensorFlow places the nodes

修改:有GitHub issue。你可以尝试:

#include <stdlib.h>
setenv("CUDA_VISIBLE_DEVICES", "", 1);

auto gpu_options = config->gpu_options();
gpu_options.set_visible_device_list("");

但这可能会给你failed call to cuInit: CUDA_ERROR_NO_DEVICE

答案 1 :(得分:3)

当您将参数设置为cpu:1时,它不会阻止tensorflow初始化GPU设备。

session_conf = tf.ConfigProto(
    device_count={'CPU' : 1, 'GPU' : 0},
    allow_soft_placement=True,
    log_device_placement=False
)

另外......不得已:

  

alias nogpu='export CUDA_VISIBLE_DEVICES=-1;'

     

nogpu python disable_GPU_tensorflow.py

  

setenv("CUDA_VISIBLE_DEVICES", "", 1);