基于Ana conda的环境的Tensorflow安装

Tensorflow 安装分为两种情况，一种是只使用CPU；另一种是使用GPU，这时还要安装CUDA和cuDNN，情况相对复杂。以上两类都推荐使用Anaconda作为Python环境，python的基础版本是3.5版本。本文则是给予conda环境配置安装的Tensorflow，可以不用独立安装anaconda环境外的python。

前置安装条件

安装python 以及环境配置参照博客文章:
ubuntu 环境下python安装http://blog.youkuaiyun.com/cs_leebo/article/details/64442156
安装python之后还需要安装 python-pip和python-dev
输入命令：$ sudo apt-get install python-pip python-dev

pip可以理解成一个比较高级的软件安装器，安装Tensorflow要用到。 dev则是一个额外的类库，也是Tensorflow的安装和运行需要的。

对于python3.X 系列的执行环境。则需要安装pyhon3-pip
$pip –version查看pip版本
方式一：
sudo aptitude install python3-pip
方式二：
安装sudo apt-get install python3-pip
卸载sudo apt-get remove python3-pip

对于python-pip的升级问题，按照相关提示，正常可保持默认。

1、安装Anaconda

anaconda 是python的一个科学计算发行版。默认含有python3.X系列的环境。Anaconda 使用的包管理器是 “conda”，conda environments独立于其他python项目。 Anaconda environment环境下安装的tensorflow，不会覆盖之前python环境下的安装的tensorflow。但是还是要卸载之外安装的tensorflow

Anaconda 安装步骤：
● Install Anaconda.
● Create a conda environment.
● Activate the conda environment and install TensorFlow in it.
● After the install you will activate the conda environment each time you want to use TensorFlow.
● Optionally install ipython and other packages into the conda environment.

Note:卸载之前通过pip 安装的tensorflow,若没有安装，则跳过此处。
If tensorflow has been installed via pip outside the Anaconda environment previously, then one should uninstall it if one wants to use the tensorflow installed within an Anaconda environment, because Anaconda searches system site-packages from .local with higher priority. “`bash
Python 2
pip uninstall tensorflow
Python 3
pip3 uninstall tensorflow

清华软件源 https://mirrors.tuna.tsinghua.edu.cn/anaconda/archive/
下载后，执行#bash Anaconda3-4.3.0-Linux-x86_64.sh
执行后显示的是License文档，按q键跳过，输入yes确认。安装完后在～/.bashrc文件下配置环境变量：
export PATH=”/home/leebo/anaconda3/bin:$PATH”重启终端配置完成。

1.1 Conda 安装
Conda 是一个开源的软件包管理系统和环境管理系统，用于安装多个版本的软件包及其依赖关系，并在它们之间轻松切换。
常用命令：conda list 列出当前 conda 环境所链接的软件包
conda create 创建一个 conda 环境，名称为 tf
conda create -n 环境名 -c 镜像源

1.2 创建conda执行环境
Create a conda environment called tensorflow:

• Python 2.7
  $ conda create -n tensorflow python=2.7

• Python 3.4
  $ conda create -n tensorflow python=3.4

• Python 3.5
  $ conda create -n tensorflow python=3.5

安装完成后，使用如下命令激活\关闭，conda执行环境。
To activate this environment, use:

source activate tensorflow
To deactivate this environment, use:
source deactivate tensorflow

激活conda环境后，可使用conda或者pip两种办法安装tensorfolw.

2、安装tensorflow

2.1 使用conda命令安装tensorflow
Using conda参照如下网址
A community maintained conda package is available from conda-forge.
https://github.com/conda-forge/tensorflow-feedstock

Only the CPU version of TensorFlow is available at the moment and can be installed in the conda environment for Python 2 or Python 3.

$ source activate tensorflow
(tensorflow)$ # Your prompt should change

Linux/Mac OS X, Python 2.7/3.4/3.5, CPU only:

(tensorflow)$ conda install -c conda-forge tensorflow

2.2、使用pip命令安装tensorflow
If using pip make sure to use the –ignore-installed flag to prevent errors about easy_install.
$ source activate tensorflow
(tensorflow)$ # Your prompt should change

2.2.1、Tensorflow CPU版本安装
export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-1.0.0rc0-cp35-cp35m-linux_x86_64.whl

Finally install TensorFlow:
Python 2
(tensorflow)$$ pip install –ignore-installed –upgrade $TF_BINARY_URL

Python 3
(tensorflow)$$ pip3 install –ignore-installed –upgrade $TF_BINARY_URL

• 错误信息提示
  1) tensorflow-1.0.0rc0-cp35-cp35m-linux_x86_64.whl is not a supported wheel on this platform.

  则需要更改TF_BINARY_URL的源地址。
  $ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.12.0rc1-cp34-cp34m-linux_x86_64.whl

  2)出现PermissionError: [Errno 13] 权限不够: /usr/local/lib/python3.5/dist-packages/numpy-1.12.1.dist-info错误信息，执行下面安装命令。
  $pip install packageName –user

2.2.2、Tensorflow GPU版本的安装
对于Nvidia 显卡，需要安装显卡驱动、CUDA 、cuDNN
CUDA是NVIDIA 推出的使用GPU 资源进行计算的SDK ，CUDA 里面集成了显卡驱动。
https://developer.nvidia.com/cuda-toolkit

Ubuntu/Linux 64-bit, GPU enabled. Requires CUDA toolkit 7.5 and CuDNN v4. For
other versions, see “Install from sources” below.
$ sudo pip3 install –upgrade https://sto

3、tensorflow 安装之后

With the conda environment activated, you can now test your installation.
When you are done using TensorFlow, deactivate the environment.
(tensorflow)$ source deactivate

$ # Your prompt should change back

To use TensorFlow later you will have to activate the conda environment again:
$ source activate tensorflow
(tensorflow)$# Your prompt should change. Run Python programs that use TensorFlow. … When you are done using TensorFlow, deactivate the environment. (tensorflow)$ source deactivate

4、实例

(tensorflow) android@local:~$ python
Python 3.4.5 |Continuum Analytics, Inc.| (default, Jul 2 2016, 17:47:47)
[GCC 4.4.7 20120313 (Red Hat 4.4.7-1)] on linux
Type “help”, “copyright”, “credits” or “license” for more information.

import tensorflow as tf
hello = tf.constant(‘hello,tensorflow!’)
sess = tf.Session()
print(sess.run(hello))
b’hello,tensorflow!’
a=tf.constant(10)
b=tf.constant(32)
print(sess.run(a+b))

5、其它工具包的安装，

IPython工具包
IPython 是一个 python 的交互式 shell，比默认的python shell 好用得多，支持变量自动补全，自动缩进，支持 bash shell 命令，内置了许多很有用的功能和函数。
IPython 是基于BSD 开源的。
IPython 为交互式计算提供了一个丰富的架构，包含：
● 强大的交互式 shell
● Jupyter 内核
● 交互式的数据可视化工具
● 灵活、可嵌入的解释器
● 易于使用，高性能的并行计算工具[1]
安装
在ubuntu 下只要 sudo apt-get install ipython 就装好了，通过 ipython 启动。

To use tensorflow with IPython it may be necessary to install IPython into the tensorflow environment:
$source activate tensorflow (tensorflow)$ conda install ipython

Similarly, other Python packages like pandas may need to get installed into the tensorflow environment before they can be used together with tensorflow.

6、常见问题以及错误

6.1 安装指导网站
https://www.tensorflow.org/versions/master/get_started/os_setup.html#anaconda-installation

6.2 安装后，运行实例提示ModuleNotFoundError: No module named ‘tensorflow’
import tensorflow as tf
Traceback (most recent call last):
File “”, line 1, in
ModuleNotFoundError: No module named ‘tensorflow’
解决办法：不要安装太新的版本