Question

所以我安装了tensorflow并且CPU版本工作正常，但我似乎无法让GPU工作。我从Nvidia下载.deb安装了Cuda。我复制了cudNN内容

cat /usr/local/cuda/include/cudnn.h | grep CUDNN_MAJOR -A 2
#define CUDNN_MAJOR      6
#define CUDNN_MINOR      0
#define CUDNN_PATCHLEVEL 20
--
#define CUDNN_VERSION    (CUDNN_MAJOR * 1000 + CUDNN_MINOR * 100 +     CUDNN_PATCHLEVEL)

#include "driver_types.h"

我在〜/ .profile

中输入了路径

export CUDA_HOME=/usr/local/cuda
export CUDA_ROOT=/usr/local/cuda
export PATH=$PATH:$CUDA_ROOT/bin
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$CUDA_ROOT/lib64

哦，Nvidia-smi表示：

| NVIDIA-SMI 375.39                 Driver Version: 375.39                                            |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  GeForce GTX 1080    Off  | 0000:01:00.0      On |                  N/A |
| 10%   54C    P0    42W / 200W |    591MiB /  8105MiB |      4%      Default |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID  Type  Process name                               Usage      |
|=============================================================================|
|    0      1368    G   /usr/bin/X                                     344MiB |
|    0      3078    G   cinnamon                                       129MiB |
|    0      6549    G   /usr/lib/virtualbox/VirtualBox                  20MiB |
|    0     15491    G   ...bleH2AndQuicRequests/Enabled/*NetworkTime    96MiB |

然而我仍然使用Tensorflow：

>>> python
>>> import tensorflow as tf
>>> sess = tf.Session(config=tf.ConfigProto(log_device_placement=True))

W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE3 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.1 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.2 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX2 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use FMA instructions, but these are available on your machine and could speed up CPU computations.
Device mapping: no known devices.
I tensorflow/core/common_runtime/direct_session.cc:257] Device mapping:

我希望你能告诉我还能做些什么。最好的问候和提前感谢

Answer 1

我认为你安装的只是一个简单（CPU）版本的tensorflow。您必须安装 gpu版本的tensorflow 。最简单的方法是使用Python的anaconda发行版和tensorflow。

kmario23 ❯ conda install -c anaconda tensorflow-gpu                         
Fetching package metadata .........
Solving package specifications: ..........

Package plan for installation in environment /home/kmario23/anaconda3:

The following packages will be downloaded:

    package                    |            build
    ---------------------------|-----------------
    conda-env-2.6.0            |                0          502 B  anaconda
    cudatoolkit-7.5            |                0       217.2 MB  anaconda
    cudnn-5.1                  |                0        77.2 MB  anaconda
    tensorflow-gpu-1.0.1       |           py35_4        77.4 MB  anaconda
    conda-4.3.16               |           py35_0         510 KB  anaconda
    ------------------------------------------------------------
                                           Total:       372.4 MB

The following NEW packages will be INSTALLED:

    cudatoolkit:    7.5-0         anaconda   
    cudnn:          5.1-0         anaconda   
    tensorflow-gpu: 1.0.1-py35_4  anaconda   

The following packages will be UPDATED:

    conda:          4.2.13-py35_0 conda-forge --> 4.3.16-py35_0 anaconda

Proceed ([y]/n)?

立即安装，立即安装。

另外，看看简单的包搜索会为GPU列出单独的tensorflow版本。

kmario23 ❯ anaconda search -t conda tensorflow                                                                                                          
Using Anaconda API: https://api.anaconda.org
Run 'anaconda show <USER/PACKAGE>' to get more details:
Packages:
     Name                      |  Version | Package Types   | Platforms      
     ------------------------- |   ------ | --------------- | ---------------
     HCC/tensorflow            |    1.0.0 | conda           | linux-64       
     HCC/tensorflow-cpucompat  |    1.0.0 | conda           | linux-64       
     HCC/tensorflow-fma        |    1.0.0 | conda           | linux-64       
     SentientPrime/tensorflow  |    0.6.0 | conda           | osx-64         
                                          : TensorFlow helps the tensors flow
     acellera/tensorflow-cuda  |   0.12.1 | conda           | linux-64       
     anaconda/tensorflow       |    1.0.1 | conda           | linux-64       
     anaconda/tensorflow-gpu   |    1.0.1 | conda           | linux-64       
     conda-forge/tensorflow    |    1.0.0 | conda           | linux-64, win-64, osx-64
                                          : TensorFlow helps the tensors flow

我建议您从anaconda频道tensorflow-gpu安装anaconda/tensorflow-gpu。

Linux Mint / Ubuntu 14.04上Tensorflow的GPU支持

1 个答案: