深度学习框架TensorFlow实例——在windows中安装tensorflow-gpu开发环境

1、安装Anaconda

Anaconda对于学过python的人来说应该不陌生，它集成了python环境下的很多包，省去自己去下载各类依赖的麻烦，相当于一个工具包，下载地址为https://www.anaconda.com/download/，选择64位版本。

下载完成后安装，一步步点击ok就好，我将Anaconda安装在C:\Users\yanghang\Anaconda3 下。

注意将安装路径添加到系统环境变量中，右键我的电脑--属性--高级系统设置-环境变量，在系统变量path中加入Anaconda的安装路径即可，如下图所示：

此时在命令行下查看Python的版本，可以发现电脑中已经成功安装了python3.6的开发环境：

2、安装tensorflow-gpu版本

目前tensorflow-gpu版本只支持python3.5.2，所以我们使用如下命令建立一个tensorflow的计算环境：

conda create -n tensorflow python=3.5.2

随后激活此环境：

activate tensorflow

在此环境下执行安装命令：

pip install --upgrade https://storage.googleapis.com/tensorflow/windows/gpu/tensorflow_gpu-0.12.0-cp35-cp35m-win_amd64.whl

安装成功！

3、安装cuda和cudnn

只安装了tensorflow是不够的，想要配置gpu的开发环境，还需要安装cuda和cudnn。

cuda下载地址：https://developer.nvidia.com/cuda-downloads

下载之后安装即可，一路点下一步！

cudnn下载地址：https://developer.nvidia.com/rdp/cudnn-download

cudnn下载后直接解压即可。

需要将bin目录和libnvvp目录添加到环境变量path中,如下图所示：

4、使用开发工具进行测试（这里使用eclipse pydev）

首先我们需要将解释器选择为我们刚刚创建的tensorflow环境，然后编写代码如下：

import tensorflow as tf


a = tf.constant(5, name="input_a")
b = tf.constant(3, name="input_b")
c = tf.multiply(a, b, name="mul_c")
d = tf.add(a, b, name="add_d")
# d = tf.divide(a, b, name="div_e")
e = tf.subtract(c, d, name="sub_e")

ses = tf.Session()
foo = ses.run(e)
print(foo)

如果顺利的话，运行结果如下，输出了结果为7：

2018-04-19 20:49:41.284578: W c:\tf_jenkins\home\workspace\release-win\m\windows-gpu\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.
2018-04-19 20:49:41.285009: W c:\tf_jenkins\home\workspace\release-win\m\windows-gpu\py\35\tensorflow\core\platform\cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE2 instructions, but these are available on your machine and could speed up CPU computations.
2018-04-19 20:49:41.285336: W c:\tf_jenkins\home\workspace\release-win\m\windows-gpu\py\35\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.
2018-04-19 20:49:41.285673: W c:\tf_jenkins\home\workspace\release-win\m\windows-gpu\py\35\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.
2018-04-19 20:49:41.286012: W c:\tf_jenkins\home\workspace\release-win\m\windows-gpu\py\35\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.
2018-04-19 20:49:41.286343: W c:\tf_jenkins\home\workspace\release-win\m\windows-gpu\py\35\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.
2018-04-19 20:49:41.286667: W c:\tf_jenkins\home\workspace\release-win\m\windows-gpu\py\35\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.
2018-04-19 20:49:41.287723: W c:\tf_jenkins\home\workspace\release-win\m\windows-gpu\py\35\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.
2018-04-19 20:49:41.863562: I c:\tf_jenkins\home\workspace\release-win\m\windows-gpu\py\35\tensorflow\core\common_runtime\gpu\gpu_device.cc:940] Found device 0 with properties: 
name: GeForce 940M
major: 5 minor: 0 memoryClockRate (GHz) 0.98
pciBusID 0000:04:00.0
Total memory: 1.00GiB
Free memory: 831.82MiB
2018-04-19 20:49:41.864253: I c:\tf_jenkins\home\workspace\release-win\m\windows-gpu\py\35\tensorflow\core\common_runtime\gpu\gpu_device.cc:961] DMA: 0 
2018-04-19 20:49:41.864672: I c:\tf_jenkins\home\workspace\release-win\m\windows-gpu\py\35\tensorflow\core\common_runtime\gpu\gpu_device.cc:971] 0:   Y 
2018-04-19 20:49:41.864962: I c:\tf_jenkins\home\workspace\release-win\m\windows-gpu\py\35\tensorflow\core\common_runtime\gpu\gpu_device.cc:1030] Creating TensorFlow device (/gpu:0) -> (device: 0, name: GeForce 940M, pci bus id: 0000:04:00.0)
7

原文链接：http://www.sohu.com/a/146708563_572440