170819 Anaconda两行命令安装tensorflow-gpu+keras-gpu及Gpu vs Cpu验证

参考文献：
清华大学开源软件镜像站
Using GPUs
Linux下Anaconda的安装使用与卸载-注：安装Anaconda最后一步要选yes

Step1 添加清华镜像，加快下载速度, 创建tensorflow-gpu环境

conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/
conda config --set show_channel_urls yes

conda create -n tensorflow-gpu python=3.6
source activate tensorflow-gpu #（linux下+source， windows下无需+source）

Step2 安装tensorflow-gpu

conda install tensorflow-gpu

Step3 安装keras-gpu

conda install keras-gpu

注意：一定要加上-gpu，否则系统会默认成cpu

Step4 验证是gpu还是cpu

• 默认gpu

import tensorflow as tf
# 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))

• 手动设置gpu与cpu

# Creates a graph.
import tensorflow as tf
with tf.device('/cpu:0'):
  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))

若成功运行Gpu则在终端会有相应的gpu提示提示，例如：/gpu: 0 如下图：