Tensorflow未在GPU上运行

Question

我已经花了相当多的时间在堆栈溢出上挖掘,否则寻找答案,但找不到任何东西

大家好,

我正在使用Keras运行Tensorflow.我90%肯定我安装了Tensorflow GPU,有没有办法检查我做了哪个安装？

我试图从Jupyter笔记本运行一些CNN模型,我注意到Keras在CPU上运行模型(检查任务管理器,CPU为100%).

我尝试从tensorflow网站运行此代码:

# Creates a graph.
a = tf.constant([1.0, 2.0, 3.0, 4.0, 5.0, 6.0], shape=[2, 3], name='a')
b = tf.constant([1.0, 2.0, 3.0, 4.0, 5.0, 6.0], shape=[3, 2], name='b')
c = tf.matmul(a, b)
# Creates a session with log_device_placement set to True.
sess = tf.Session(config=tf.ConfigProto(log_device_placement=True))
# Runs the op.
print(sess.run(c))

这就是我得到的:

MatMul: (MatMul): /job:localhost/replica:0/task:0/cpu:0
2017-06-29 17:09:38.783183: I c:\tf_jenkins\home\workspace\release-win\m\windows\py\35\tensorflow\core\common_runtime\simple_placer.cc:847] MatMul: (MatMul)/job:localhost/replica:0/task:0/cpu:0
b: (Const): /job:localhost/replica:0/task:0/cpu:0
2017-06-29 17:09:38.784779: I c:\tf_jenkins\home\workspace\release-win\m\windows\py\35\tensorflow\core\common_runtime\simple_placer.cc:847] b: (Const)/job:localhost/replica:0/task:0/cpu:0
a: (Const): /job:localhost/replica:0/task:0/cpu:0
2017-06-29 17:09:38.786128: I c:\tf_jenkins\home\workspace\release-win\m\windows\py\35\tensorflow\core\common_runtime\simple_placer.cc:847] a: (Const)/job:localhost/replica:0/task:0/cpu:0
[[ 22.  28.]
 [ 49.  64.]]

对我来说,显示我正在运行我的CPU,出于某种原因.

我有一个GTX1050(驱动程序版本382.53),我安装了CUDA和Cudnn,并且没有任何问题安装了tensorflow.我也安装了Visual Studio 2015,因为它被列为兼容版本.

我记得CUDA提到了一些关于安装了不兼容驱动程序的问题,但是如果我没记错的话,CUDA应该已经安装了自己的驱动程序.

编辑: 我运行了这些命令来列出可用的设备

from tensorflow.python.client import device_lib
print(device_lib.list_local_devices())

这就是我得到的

[name: "/cpu:0"
device_type: "CPU"
memory_limit: 268435456
locality {
}
incarnation: 14922788031522107450
]

还有很多像这样的警告

2017-06-29 17:32:45.401429: W c:\tf_jenkins\home\workspace\release-win\m\windows\py\35\tensorflow\core\platform\cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE instructions, but these are available on your machine and could speed up CPU computations.

编辑2

试过跑

pip3 install --upgrade tensorflow-gpu

我明白了

Requirement already up-to-date: tensorflow-gpu in c:\users\goofynose\appdata\local\programs\python\python35\lib\site-packages
Requirement already up-to-date: markdown==2.2.0 in c:\users\goofynose\appdata\local\programs\python\python35\lib\site-packages (from tensorflow-gpu)
Requirement already up-to-date: html5lib==0.9999999 in c:\users\goofynose\appdata\local\programs\python\python35\lib\site-packages (from tensorflow-gpu)
Requirement already up-to-date: werkzeug>=0.11.10 in c:\users\goofynose\appdata\local\programs\python\python35\lib\site-packages (from tensorflow-gpu)
Requirement already up-to-date: wheel>=0.26 in c:\users\goofynose\appdata\local\programs\python\python35\lib\site-packages (from tensorflow-gpu)
Requirement already up-to-date: bleach==1.5.0 in c:\users\goofynose\appdata\local\programs\python\python35\lib\site-packages (from tensorflow-gpu)
Requirement already up-to-date: six>=1.10.0 in c:\users\goofynose\appdata\local\programs\python\python35\lib\site-packages (from tensorflow-gpu)
Requirement already up-to-date: protobuf>=3.2.0 in c:\users\goofynose\appdata\local\programs\python\python35\lib\site-packages (from tensorflow-gpu)
Requirement already up-to-date: backports.weakref==1.0rc1 in c:\users\goofynose\appdata\local\programs\python\python35\lib\site-packages (from tensorflow-gpu)
Requirement already up-to-date: numpy>=1.11.0 in c:\users\goofynose\appdata\local\programs\python\python35\lib\site-packages (from tensorflow-gpu)
Requirement already up-to-date: setuptools in c:\users\goofynose\appdata\local\programs\python\python35\lib\site-packages (from protobuf>=3.2.0->tensorflow-gpu)

解决: 检查注释的解决方案.感谢所有帮助过的人!

我是新手,所以任何帮助都非常感谢!谢谢.

Answer 1

要检查TensorFlow可用的设备,您可以使用它并查看GPU卡是否可用:

from tensorflow.python.client import device_lib
print(device_lib.list_local_devices())

编辑此外,如果您使用TensorFlow Cuda版本,您应该看到这种日志:

I tensorflow/stream_executor/dso_loader.cc:128] successfully opened CUDA library libcublas.so.*.* locally
I tensorflow/stream_executor/dso_loader.cc:128] successfully opened CUDA library libcudnn.so.*.*  locally
I tensorflow/stream_executor/dso_loader.cc:128] successfully opened CUDA library libcufft.so.*.*  locally

Answer 2

这可能听起来很愚蠢,但尝试重启.它帮助我和GitHub中的其他人.

Answer 3

即使在通过pip正确安装tensorflow-gpu之后,我仍然无法获得GPU支持.我的问题是我安装了tensorflow 1.5和CUDA 9.1(默认版本Nvidia指示你),而预编译的tensorflow 1.5适用于CUDA版本<= 9.0.这是nvidia网站上的下载页面,以获得正确的CUDA 9.0:

https://developer.nvidia.com/cuda-90-download-archive

另外,请确保将您的cuDNN更新为与CUDA 9.0兼容的版本 https://developer.nvidia.com/cudnn https://developer.nvidia.com/rdp/cudnn-download