人工智能开发环境 使用Anaconda安装TensorFlow

红太狼 2022-06-09 04:11 276阅读 0赞

文章来源 http://blog.csdn.net/nxcxl88/article/details/52704877?locationNum=13

目录(?)[+]

建议参照最新的tensorflow安装步骤(Linux,官方网站经常访问不是很稳定,所以给了一个github的地址):https://github.com/tensorflow/tensorflow/blob/master/tensorflow/docs\_src/install/install\_linux.md

最近,tensorflow网站上给出了新的使用Anaconda配置和安装Tensorflow的步骤,经过测试,在国内可以无障碍的访问。Anaconda 是一个基于Python的科学计算包集合,目前支持python 2.7,3.4,3.5,3.6。

注意:在安装过程中如果出现很长的报错,观察错误信息的末尾,如果是网络链接相关,就重新运行一遍语句即可(如出现进度条不动的情况,也可重新运行语句),Anaconda自身约500M,tensorflow所需软件包约几十M。

操作系统: Ubuntu 14.04

1. 安装Anaconda

从anaconda官网(https://www.continuum.io/downloads)上下载[linux][Linux]版本的安装文件(推荐Python 2.7版本),运行sh完成安装。

2. 建立一个tensorflow的运行环境

[plain] view plain copy

  1. # Python 2.7
  2. $ conda create -n tensorflow python=2.7
  3. # Python 3.4
  4. $ conda create -n tensorflow python=3.4
  5. # Python 3.5
  6. $ conda create -n tensorflow python=3.5

3.在conda环境中安装tensorflow

在conda环境中安装tensorflow的好处是可以便捷的管理tensorflow的依赖包。分为两个步骤:激活上一步建立的名为tensorflow的conda环境;用conda或者pip工具安装Tensorflow,作者选择的是pip方式。

3.1 pip方式

pip方式需要首先激活conda环境

[plain] view plain copy

  1. $ source activate tensorflow

然后根据要安装的不同tensorflow版本选择对应的 一条环境变量设置export语句(操作系统,Python版本,CPU版本还是CPU+GPU版本)

[plain] view plain copy

  1. # Ubuntu/Linux 64-bit, CPU only, Python 2.7
  2. (tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.10.0-cp27-none-linux\_x86\_64.whl
  3. # Ubuntu/Linux 64-bit, GPU enabled, Python 2.7
  4. # Requires CUDA toolkit 7.5 and CuDNN v5. For other versions, see “Install from sources” below.
  5. (tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow-0.10.0-cp27-none-linux\_x86\_64.whl
  6. # Mac OS X, CPU only, Python 2.7:
  7. (tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.10.0-py2-none-any.whl
  8. # Mac OS X, GPU enabled, Python 2.7:
  9. (tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow-0.10.0-py2-none-any.whl
  10. # Ubuntu/Linux 64-bit, CPU only, Python 3.4
  11. (tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.10.0-cp34-cp34m-linux\_x86\_64.whl
  12. # Ubuntu/Linux 64-bit, GPU enabled, Python 3.4
  13. # Requires CUDA toolkit 7.5 and CuDNN v5. For other versions, see “Install from sources” below.
  14. (tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow-0.10.0-cp34-cp34m-linux\_x86\_64.whl
  15. # Ubuntu/Linux 64-bit, CPU only, Python 3.5
  16. (tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.10.0-cp35-cp35m-linux\_x86\_64.whl
  17. # Ubuntu/Linux 64-bit, GPU enabled, Python 3.5
  18. # Requires CUDA toolkit 7.5 and CuDNN v5. For other versions, see “Install from sources” below.
  19. (tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow-0.10.0-cp35-cp35m-linux\_x86\_64.whl
  20. # Mac OS X, CPU only, Python 3.4 or 3.5:
  21. (tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.10.0-py3-none-any.whl
  22. # Mac OS X, GPU enabled, Python 3.4 or 3.5:
  23. (tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow-0.10.0-py3-none-any.whl

最后根据是python 2还是3版本选择一句进行安装。

[plain] view plain copy

  1. # Python 2
  2. (tensorflow)$ pip install —ignore-installed —upgrade $TF_BINARY_URL
  3. # Python 3
  4. (tensorflow)$ pip3 install —ignore-installed —upgrade $TF_BINARY_URL

3.2 conda方式

conda上面目前有人已经做好了tensorflow的pkg,但是版本不一定最新,且只有CPU版本,不支持GPU。

步骤也是首先激活conda环境,然后调用conda install 语句安装.

[plain] view plain copy

  1. $ source activate tensorflow
  2. (tensorflow)$ # Your prompt should change
  3. # Linux/Mac OS X, Python 2.7/3.4/3.5, CPU only:
  4. (tensorflow)$ conda install -c conda-forge tensorflow

上面的步骤完成后,从conda环境中退出:

[plain] view plain copy

  1. (tensorflow)$ source deactivate

4. 测试安装

[plain] view plain copy

  1. $ source activate tensorflow
  2. (tensorflow)$ # Your prompt should change.
  3. # Run Python programs that use TensorFlow.
  4. # When you are done using TensorFlow, deactivate the environment.
  5. (tensorflow)$ source deactivate

发表评论

表情:
评论列表 (有 0 条评论,276人围观)

还没有评论,来说两句吧...

相关阅读