前言
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 或 Ubuntu16.04
这是Github官网给出的安装步骤
https://github.com/tensorflow/tensorflow/blob/master/tensorflow/docs_src/install/install_linux.md
第一步、 安装Anaconda
从anaconda官网（https://www.continuum.io/downloads）上下载linux版本的安装文件，运行完成安装。
我这里是以Anaconda2-5.0.1-Linux-x86_64.sh为例，Anaconda3一样啦。这个很简单。
deeplearning@deeplearningsinglenode:~/SoftWare$ pwd /home/deeplearning/SoftWare deeplearning@deeplearningsinglenode:~/SoftWare$ ll total 519916 drwxrwxr-x 4 deeplearning deeplearning 4096 12月 4 09:42 ./ drwxr-xr-x 17 deeplearning deeplearning 4096 12月 3 20:46 ../ -rwxrw-r-- 1 deeplearning deeplearning 532375438 12月 4 09:42 Anaconda2-5.0.1-Linux-x86_64.sh* drwxr-xr-x 8 deeplearning deeplearning 4096 8月 5 2015 jdk1.8.0_60/ drwxrwxr-x 11 deeplearning deeplearning 4096 12月 3 20:07 pycharm-2017.3/ deeplearning@deeplearningsinglenode:~/SoftWare$ bash ./Anaconda2-5.0.1-Linux-x86_64.sh Welcome to Anaconda2 5.0.1 In order to continue the installation process, please review the license agreement. Please, press ENTER to continue >>> =================================== Anaconda End User License Agreement =================================== Copyright 2015, Anaconda, Inc. All rights reserved under the 3-clause BSD License: Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditio ns are met: * Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. * Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. * Neither the name of Continuum Analytics, Inc. (dba Anaconda, Inc.) ("Continuum") nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT N OT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL CON TINUUM BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY TH EORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. Notice of Third Party Software Licenses ======================================= Anaconda contains open source software packages from third parties. These are available on an "as is" basis and subject to their in dividual license agreements. These licenses are available in Anaconda or at https://docs.anaconda.com/anaconda/packages/pkg-docs . Any binary packages of these third party tools you obtain via Anaconda are subject to their individual licenses as well as the Anac onda license. Continuum reserves the right to change which third party tools are provided in Anaconda. In particular, Anaconda contains re-distributable, run-time, shared-library files from the Intel(TM) Math Kernel Library ("MKL bina ries"). You are specifically authorized to use the MKL binaries with your installation of Anaconda. You are also authorized to redi stribute the MKL binaries with Anaconda or in the conda package that contains them. Use and redistribution of the MKL binaries are subject to the licensing terms located at https://software.intel.com/en-us/license/intel-simplified-software-license. If needed, in structions for removing the MKL binaries after installation of Anaconda are available at http://www.anaconda.com. Anaconda also contains cuDNN software binaries from NVIDIA Corporation ("cuDNN binaries"). You are specifically authorized to use t he cuDNN binaries with your installation of Anaconda. You are also authorized to redistribute the cuDNN binaries with an Anaconda p ackage that contains them. If needed, instructions for removing the cuDNN binaries after installation of Anaconda are available at http://www.anaconda.com. Cryptography Notice =================== This distribution includes cryptographic software. The country in which you currently reside may have restrictions on the import, p ossession, use, and/or re-export to another country, of encryption software. BEFORE using any encryption software, please check you r country's laws, regulations and policies concerning the import, possession, or use, and re-export of encryption software, to see if this is permitted. See the Wassenaar Arrangement <http://www.wassenaar.org/> for more information. Continuum has self-classified this software as Export Commodity Control Number (ECCN) 5D002.C.1, which includes information securit y software using or performing cryptographic functions with asymmetric algorithms. The form and manner of this distribution makes i t eligible for export under the License Exception ENC Technology Software Unrestricted (TSU) exception (see the BIS Export Administ ration Regulations, Section 740.13) for both object code and source code. In addition, the Intel(TM) Math Kernel Library contained in Continuum's software is classified by Intel(TM) as ECCN 5D992b with no license required for export to non-embargoed countries. The following packages are included in this distribution that relate to cryptography: openssl The OpenSSL Project is a collaborative effort to develop a robust, commercial-grade, full-featured, and Open Source toolkit imp lementing the Transport Layer Security (TLS) and Secure Sockets Layer (SSL) protocols as well as a full-strength general purpose cr yptography library. pycrypto A collection of both secure hash functions (such as SHA256 and RIPEMD160), and various encryption algorithms (AES, DES, RSA, El Gamal, etc.). pyopenssl A thin Python wrapper around (a subset of) the OpenSSL library. kerberos (krb5, non-Windows platforms) A network authentication protocol designed to provide strong authentication for client/server applications by using secret-key cryptography. cryptography A Python library which exposes cryptographic recipes and primitives.
Please answer 'yes' or 'no':' >>> yes Anaconda2 will now be installed into this location: /home/deeplearning/anaconda2 - Press ENTER to confirm the location - Press CTRL-C to abort the installation - Or specify a different location below [/home/deeplearning/anaconda2] >>> PREFIX=/home/deeplearning/anaconda2
installing: python-2.7.14-hc2b0042_21 ... Python 2.7.14 :: Anaconda, Inc. installing: ca-certificates-2017.08.26-h1d4fec5_0 ... installing: conda-env-2.6.0-h36134e3_1 ... installing: intel-openmp-2018.0.0-h15fc484_7 ... installing: libgcc-ng-7.2.0-h7cc24e2_2 ... installing: libgfortran-ng-7.2.0-h9f7466a_2 ... installing: libstdcxx-ng-7.2.0-h7a57d05_2 ... installing: bzip2-1.0.6-h0376d23_1 ... installing: expat-2.2.4-hc00ebd1_1 ... installing: gmp-6.1.2-hb3b607b_0 ... installing: graphite2-1.3.10-hc526e54_0 ... installing: icu-58.2-h211956c_0 ... installing: jbig-2.1-hdba287a_0 ... installing: jpeg-9b-habf39ab_1 ... installing: libffi-3.2.1-h4deb6c0_3 ... installing: libsodium-1.0.13-h31c71d8_2 ... installing: libssh2-1.8.0-h8c220ad_2 ... installing: libtool-2.4.6-hd50d1a6_0 ... installing: libxcb-1.12-h84ff03f_3 ... installing: lzo-2.10-h1bfc0ba_1 ... installing: mkl-2018.0.0-hb491cac_4 ... installing: ncurses-6.0-h06874d7_1 ... installing: openssl-1.0.2l-h077ae2c_5 ... installing: patchelf-0.9-hf79760b_2 ... installing: pcre-8.41-hc71a17e_0 ... installing: pixman-0.34.0-h83dc358_2 ... installing: tk-8.6.7-h5979e9b_1 ... installing: unixodbc-2.3.4-hc36303a_1 ... installing: xz-5.2.3-h2bcbf08_1 ... installing: yaml-0.1.7-h96e3832_1 ... installing: zlib-1.2.11-hfbfcf68_1 ... installing: curl-7.55.1-hcb0b314_2 ... installing: glib-2.53.6-hc861d11_1 ... installing: hdf5-1.10.1-hb0523eb_0 ... installing: libedit-3.1-heed3624_0 ... installing: libpng-1.6.32-hda9c8bc_2 ... installing: libtiff-4.0.8-h90200ff_9 ... installing: libxml2-2.9.4-h6b072ca_5 ... installing: mpfr-3.1.5-h12ff648_1 ... installing: pandoc-1.19.2.1-hea2e7c5_1 ... installing: readline-7.0-hac23ff0_3 ... installing: zeromq-4.2.2-hb0b69da_1 ... installing: dbus-1.10.22-h3b5a359_0 ... installing: freetype-2.8-h52ed37b_0 ... installing: gstreamer-1.12.2-h4f93127_0 ... installing: libxslt-1.1.29-hcf9102b_5 ... installing: mpc-1.0.3-hf803216_4 ... installing: sqlite-3.20.1-h6d8b0f3_1 ... installing: fontconfig-2.12.4-h88586e7_1 ... installing: gst-plugins-base-1.12.2-he3457e5_0 ... installing: alabaster-0.7.10-py27he5a193a_0 ... installing: asn1crypto-0.22.0-py27h94ebe91_1 ... installing: backports-1.0-py27h63c9359_1 ... installing: backports_abc-0.5-py27h7b3c97b_0 ... installing: beautifulsoup4-4.6.0-py27h3f86ba9_1 ... installing: bitarray-0.8.1-py27h304d4c6_0 ... installing: boto-2.48.0-py27h9556ac2_1 ... installing: cairo-1.14.10-haa5651f_5 ... installing: cdecimal-2.3-py27h4e63abe_1 ... installing: certifi-2017.7.27.1-py27h9ceb091_0 ... installing: chardet-3.0.4-py27hfa10054_1 ... installing: click-6.7-py27h4225b90_0 ... installing: cloudpickle-0.4.0-py27ha64365b_0 ... installing: colorama-0.3.9-py27h5cde069_0 ... installing: configparser-3.5.0-py27h5117587_0 ... installing: contextlib2-0.5.5-py27hbf4c468_0 ... installing: dask-core-0.15.3-py27h53a7ee6_0 ... installing: decorator-4.1.2-py27h1544723_0 ... installing: docutils-0.14-py27hae222c1_0 ... installing: enum34-1.1.6-py27h99a27e9_1 ... installing: et_xmlfile-1.0.1-py27h75840f5_0 ... installing: fastcache-1.0.2-py27h4cb8e01_0 ... installing: filelock-2.0.12-py27h38fa839_0 ... installing: funcsigs-1.0.2-py27h83f16ab_0 ... installing: functools32-3.2.3.2-py27h4ead58f_1 ... installing: futures-3.1.1-py27hdbc8cbb_0 ... installing: glob2-0.5-py27hd3b7d1f_1 ... installing: gmpy2-2.0.8-py27hc856308_1 ... installing: greenlet-0.4.12-py27hac09c53_0 ... installing: grin-1.2.1-py27h54abee7_1 ... installing: heapdict-1.0.0-py27h33770af_0 ... installing: idna-2.6-py27h5722d68_1 ... installing: imagesize-0.7.1-py27hd17bf80_0 ... installing: ipaddress-1.0.18-py27h337fd85_0 ... installing: ipython_genutils-0.2.0-py27h89fb69b_0 ... installing: itsdangerous-0.24-py27hb8295c1_1 ... installing: jdcal-1.3-py27h2cc5433_0 ... installing: jedi-0.10.2-py27h8af4e35_0 ... installing: lazy-object-proxy-1.3.1-py27h682c727_0 ... installing: locket-0.2.0-py27h73929a2_1 ... installing: lxml-4.1.0-py27hb025457_0 ... installing: markupsafe-1.0-py27h97b2822_1 ... installing: mccabe-0.6.1-py27h0e7c7be_1 ... installing: mistune-0.7.4-py27h6da7e90_0 ... installing: mkl-service-1.1.2-py27hb2d42c5_4 ... installing: mpmath-0.19-py27h4bb41bd_2 ... installing: msgpack-python-0.4.8-py27hc2fa789_0 ... installing: multipledispatch-0.4.9-py27h9b5f95a_0 ... installing: numpy-1.13.3-py27hbcc08e0_0 ... installing: olefile-0.44-py27h4bd3e3c_0 ... installing: pandocfilters-1.4.2-py27h428e1e5_1 ... installing: path.py-10.3.1-py27hc258cac_0 ... installing: pep8-1.7.0-py27h444351c_0 ... installing: pkginfo-1.4.1-py27hee1a9ad_1 ... installing: ply-3.10-py27hd6d9ae5_0 ... installing: psutil-5.4.0-py27h7da3062_0 ... installing: ptyprocess-0.5.2-py27h4ccb14c_0 ... installing: py-1.4.34-py27he5894e4_1 ... installing: pycodestyle-2.3.1-py27h904819d_0 ... installing: pycosat-0.6.2-py27h1cf261c_1 ... installing: pycparser-2.18-py27hefa08c5_1 ... installing: pycrypto-2.6.1-py27h9abbf5c_1 ... installing: pycurl-7.43.0-py27hcf8ebea_3 ... installing: pyodbc-4.0.17-py27h7f7627d_0 ... installing: pyparsing-2.2.0-py27hf1513f8_1 ... installing: pysocks-1.6.7-py27he2db6d2_1 ... installing: pytz-2017.2-py27hcac29fa_1 ... installing: pyyaml-3.12-py27h2d70dd7_1 ... installing: pyzmq-16.0.2-py27h297844f_2 ... installing: qt-5.6.2-h974d657_12 ... installing: qtpy-1.3.1-py27h63d3751_0 ... installing: rope-0.10.5-py27hcb0a616_0 ... installing: ruamel_yaml-0.11.14-py27h672d447_2 ... installing: scandir-1.6-py27hf7388dc_0 ... installing: simplegeneric-0.8.1-py27h19e43cd_0 ... installing: sip-4.18.1-py27he9ba0ab_2 ... installing: six-1.11.0-py27h5f960f1_1 ... installing: snowballstemmer-1.2.1-py27h44e2768_0 ... installing: sortedcontainers-1.5.7-py27he59936f_0 ... installing: sphinxcontrib-1.0-py27h1512b58_1 ... installing: sqlalchemy-1.1.13-py27hb0a01da_0 ... installing: subprocess32-3.2.7-py27h373dbce_0 ... installing: tblib-1.3.2-py27h51fe5ba_0 ... installing: toolz-0.8.2-py27hd3b1e7e_0 ... installing: typing-3.6.2-py27h66f49e2_0 ... installing: unicodecsv-0.14.1-py27h5062da9_0 ... installing: wcwidth-0.1.7-py27h9e3e1ab_0 ... installing: webencodings-0.5.1-py27hff10b21_1 ... installing: werkzeug-0.12.2-py27hbf75dff_0 ... installing: wrapt-1.10.11-py27h04f6869_0 ... installing: xlrd-1.1.0-py27ha77178f_1 ... installing: xlsxwriter-1.0.2-py27h12cbc6b_0 ... installing: xlwt-1.3.0-py27h3d85d97_0 ... installing: babel-2.5.0-py27h20693cd_0 ... installing: backports.shutil_get_terminal_size-1.0.0-py27h5bc021e_2 ... installing: bottleneck-1.2.1-py27h21b16a3_0 ... installing: cffi-1.10.0-py27hf1aaaf4_1 ... installing: conda-verify-2.0.0-py27hf052a9d_0 ... installing: cycler-0.10.0-py27hc7354d3_0 ... installing: cytoolz-0.8.2-py27hf14aec9_0 ... installing: entrypoints-0.2.3-py27h502b47d_2 ... installing: h5py-2.7.0-py27h71d1790_1 ... installing: harfbuzz-1.5.0-h2545bd6_0 ... installing: html5lib-0.999999999-py27hdf15f34_0 ... installing: llvmlite-0.20.0-py27_0 ... installing: networkx-2.0-py27hfc23926_0 ... installing: nltk-3.2.4-py27h41293c3_0 ... installing: numexpr-2.6.2-py27he5efce1_1 ... installing: openpyxl-2.4.8-py27h9f0c937_1 ... installing: packaging-16.8-py27h5e07c7c_1 ... installing: partd-0.3.8-py27h4e55004_0 ... installing: pathlib2-2.3.0-py27h6e9d198_0 ... installing: pexpect-4.2.1-py27hcf82287_0 ... installing: pillow-4.2.1-py27h7cd2321_0 ... installing: pycairo-1.13.3-py27hea6d626_0 ... installing: pyqt-5.6.0-py27h4b1e83c_5 ... installing: python-dateutil-2.6.1-py27h4ca5741_1 ... installing: pywavelets-0.5.2-py27hecda097_0 ... installing: qtawesome-0.4.4-py27hd7914c3_0 ... installing: scipy-0.19.1-py27h1edc525_3 ... installing: setuptools-36.5.0-py27h68b189e_0 ... installing: singledispatch-3.4.0.3-py27h9bcb476_0 ... installing: sortedcollections-0.5.3-py27h135218e_0 ... installing: sphinxcontrib-websupport-1.0.1-py27hf906f22_1 ... installing: ssl_match_hostname-3.5.0.1-py27h4ec10b9_2 ... installing: sympy-1.1.1-py27hc28188a_0 ... installing: traitlets-4.3.2-py27hd6ce930_0 ... installing: zict-0.1.3-py27h12c336c_0 ... installing: backports.functools_lru_cache-1.4-py27he8db605_1 ... installing: bleach-2.0.0-py27h3a0dcc8_0 ... installing: clyent-1.2.2-py27h7276e6c_1 ... installing: cryptography-2.0.3-py27hea39389_1 ... installing: cython-0.26.1-py27hdbcff32_0 ... installing: datashape-0.5.4-py27hf507385_0 ... installing: get_terminal_size-1.0.0-haa9412d_0 ... installing: gevent-1.2.2-py27h475ea6a_0 ... installing: imageio-2.2.0-py27hf108a7f_0 ... installing: isort-4.2.15-py27hcfa4749_0 ... installing: jinja2-2.9.6-py27h82327ae_1 ... installing: jsonschema-2.6.0-py27h7ed5aa4_0 ... installing: jupyter_core-4.3.0-py27hcd9ae3a_0 ... installing: navigator-updater-0.1.0-py27h0f9cd39_0 ... installing: nose-1.3.7-py27heec2199_2 ... installing: numba-0.35.0-np113py27_10 ... installing: pandas-0.20.3-py27h820b67f_2 ... installing: pango-1.40.11-h8191d47_0 ... installing: patsy-0.4.1-py27hd1cf8c0_0 ... installing: pickleshare-0.7.4-py27h09770e1_0 ... installing: pyflakes-1.6.0-py27h904a57d_0 ... installing: pygments-2.2.0-py27h4a8b6f5_0 ... installing: pytables-3.4.2-py27h1f7bffc_2 ... installing: pytest-3.2.1-py27h98000ae_1 ... installing: scikit-learn-0.19.1-py27h445a80a_0 ... installing: testpath-0.3.1-py27hc38d2c4_0 ... installing: tornado-4.5.2-py27h97b179f_0 ... installing: wheel-0.29.0-py27h411dd7b_1 ... installing: astroid-1.5.3-py27h8f8f47c_0 ... installing: astropy-2.0.2-py27h57072c0_4 ... installing: bkcharts-0.2-py27h241ae91_0 ... installing: bokeh-0.12.10-py27he46cc6b_0 ... installing: distributed-1.19.1-py27h38c4a05_0 ... installing: flask-0.12.2-py27h6d5c1cd_0 ... installing: jupyter_client-5.1.0-py27hbee1118_0 ... installing: matplotlib-2.1.0-py27h09aba24_0 ... installing: nbformat-4.4.0-py27hed7f2b2_0 ... installing: pip-9.0.1-py27hbf658b2_3 ... installing: prompt_toolkit-1.0.15-py27h1b593e1_0 ... installing: pyopenssl-17.2.0-py27h189ff3b_0 ... installing: statsmodels-0.8.0-py27hc87d62d_0 ... installing: terminado-0.6-py27h4be8df9_0 ... installing: dask-0.15.3-py27hb94b45f_0 ... installing: flask-cors-3.0.3-py27h1a8a27f_0 ... installing: ipython-5.4.1-py27h36c99b6_1 ... installing: nbconvert-5.3.1-py27he041f76_0 ... installing: pylint-1.7.4-py27h6bc7935_0 ... installing: seaborn-0.8.0-py27h9d2aaa1_0 ... installing: urllib3-1.22-py27ha55213b_0 ... installing: ipykernel-4.6.1-py27hc93e584_0 ... installing: odo-0.5.1-py27h9170de3_0 ... installing: requests-2.18.4-py27hc5b0589_1 ... installing: scikit-image-0.13.0-py27h06cb35d_1 ... installing: anaconda-client-1.6.5-py27hc8169bf_0 ... installing: blaze-0.11.3-py27h5f341da_0 ... installing: conda-4.3.30-py27h6ae6dc7_0 ... installing: jupyter_console-5.2.0-py27hc6bee7e_1 ... installing: notebook-5.0.0-py27h3661c2b_2 ... installing: qtconsole-4.3.1-py27hc444b0d_0 ... installing: sphinx-1.6.3-py27hf9b1778_0 ... installing: anaconda-project-0.8.0-py27hd7a9a97_0 ... installing: conda-build-3.0.27-py27hff9f855_0 ... installing: jupyterlab_launcher-0.4.0-py27h0e16d15_0 ... installing: numpydoc-0.7.0-py27h9647a75_0 ... installing: widgetsnbextension-3.0.2-py27hcb77dec_1 ... installing: anaconda-navigator-1.6.9-py27hfbc306d_0 ... installing: ipywidgets-7.0.0-py27h4fda95d_0 ... installing: jupyterlab-0.27.0-py27h42ebfef_2 ... installing: spyder-3.2.4-py27h04a3490_0 ... installing: _ipyw_jlab_nb_ext_conf-0.1.0-py27h08a7f0c_0 ... installing: jupyter-1.0.0-py27h505fd4b_0 ... installing: anaconda-5.0.1-py27hd9359a7_1 ... installation finished. Do you wish the installer to prepend the Anaconda2 install location to PATH in your /home/deeplearning/.bashrc ? [yes|no] [no] >>> You may wish to edit your .bashrc to prepend the Anaconda2 install location to PATH: export PATH=/home/deeplearning/anaconda2/bin:$PATH Thank you for installing Anaconda2!
因为这是一个坑，是安装时最后一步添加环境变量的时候没有选择yes导致运行 conda info 时出错，很好解决，根据错误提示：
然后，紧接着去配置Anaconda2的环境变量。怎么做呢？很简单。
在命令行输入就可以了。
$ export PATH=/home/deeplearning/anaconda2/bin:$PATH
第二步、建立一个tensorflow的运行环境
# Python 2.7 （选好自己的） $ conda create -n tensorflow python=2.7 # Python 3.4 （选好自己的） $ conda create -n tensorflow python=3.4 # Python 3.5 （选好自己的） $ conda create -n tensorflow python=3.5
注意：在这一步，你也许会遇到conda: command not found
遇到这个问题的时候，
解决方法是：
export PATH="/home/[your_name]/anaconda/bin:$PATH"
比如我这里是
export PATH=/home/deeplearning/anaconda2/bin:$PATH
但是下一次重启之后，还是会出现这个问题，所以我们要激活下 ~/.bash_profile
. ~/.bash_profile
#或者
source ~/.bash_profile
或者source /etc/profile
那是因为我的环境变量是如下：
#Anaconda2 ANACONDA2_HOME=/home/deeplearning/anaconda2 ANACONDA2_BIN=/home/deeplearning/anaconda2/bin PATH=$PATH:$ANACONDA2_BIN export ANACONDA2_HOME ANACONDA2_BIN PATH
所以，
deeplearning@deeplearningsinglenode:~$ conda create -n tensorflow python=2.7 Fetching package metadata ........... Solving package specifications: . Package plan for installation in environment /home/deeplearning/anaconda2/envs/tensorflow: The following NEW packages will be INSTALLED: ca-certificates: 2017.08.26-h1d4fec5_0 certifi: 2017.11.5-py27h71e7faf_0 libedit: 3.1-heed3624_0 libffi: 3.2.1-hd88cf55_4 libgcc-ng: 7.2.0-h7cc24e2_2 libstdcxx-ng: 7.2.0-h7a57d05_2 ncurses: 6.0-h9df7e31_2 openssl: 1.0.2m-h26d622b_1 pip: 9.0.1-py27ha730c48_4 python: 2.7.14-hdd48546_24 readline: 7.0-ha6073c6_4 setuptools: 36.5.0-py27h68b189e_0 sqlite: 3.20.1-hb898158_2 tk: 8.6.7-hc745277_3 wheel: 0.30.0-py27h2bc6bb2_1 zlib: 1.2.11-ha838bed_2 Proceed ([y]/n)? y
第三步、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，我选择的是pip方式。
3.1 pip方式（可以这种方式来安装）
pip方式需要首先激活conda环境
deeplearning@deeplearningsinglenode:~$ source activate tensorflow
(tensorflow) deeplearning@deeplearningsinglenode:~$
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）
# Ubuntu/Linux 64-bit, CPU only, Python 2.7 (tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.10.0-cp27-none-linux_x86_64.whl # Ubuntu/Linux 64-bit, GPU enabled, Python 2.7 # Requires CUDA toolkit 7.5 and CuDNN v5. For other versions, see "Install from sources" below. (tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow-0.10.0-cp27-none-linux_x86_64.whl # Mac OS X, CPU only, Python 2.7: (tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.10.0-py2-none-any.whl # Mac OS X, GPU enabled, Python 2.7: (tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow-0.10.0-py2-none-any.whl # Ubuntu/Linux 64-bit, CPU only, Python 3.4 (tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.10.0-cp34-cp34m-linux_x86_64.whl # Ubuntu/Linux 64-bit, GPU enabled, Python 3.4 # Requires CUDA toolkit 7.5 and CuDNN v5. For other versions, see "Install from sources" below. (tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow-0.10.0-cp34-cp34m-linux_x86_64.whl # Ubuntu/Linux 64-bit, CPU only, Python 3.5 (tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.10.0-cp35-cp35m-linux_x86_64.whl # Ubuntu/Linux 64-bit, GPU enabled, Python 3.5 # Requires CUDA toolkit 7.5 and CuDNN v5. For other versions, see "Install from sources" below. (tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow-0.10.0-cp35-cp35m-linux_x86_64.whl # Mac OS X, CPU only, Python 3.4 or 3.5: (tensorflow)$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.10.0-py3-none-any.whl # Mac OS X, GPU enabled, Python 3.4 or 3.5: (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。
# Python 2 (tensorflow)$ pip install --ignore-installed --upgrade $TF_BINARY_URL # Python 3 (tensorflow)$ pip3 install --ignore-installed --upgrade $TF_BINARY_URL
(tensorflow) deeplearning@deeplearningsinglenode:~$ pip install --ignore-installed --upgrade $TF_BINARY_URL Collecting tensorflow==0.10.0 from https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.10.0-cp27-none-linux_x86_64.whl Downloading https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.10.0-cp27-none-linux_x86_64.whl (36.6MB) 12% |████ | 4.5MB 14.0MB/s eta 0:00:03^[^A^[^AException: Traceback (most recent call last): File "/home/deeplearning/anaconda2/envs/tensorflow/lib/python2.7/site-packages/pip/basecommand.py", line 215, in main status = self.run(options, args) File "/home/deeplearning/anaconda2/envs/tensorflow/lib/python2.7/site-packages/pip/commands/install.py", line 335, in run wb.build(autobuilding=True) File "/home/deeplearning/anaconda2/envs/tensorflow/lib/python2.7/site-packages/pip/wheel.py", line 749, in build self.requirement_set.prepare_files(self.finder) File "/home/deeplearning/anaconda2/envs/tensorflow/lib/python2.7/site-packages/pip/req/req_set.py", line 380, in prepare_files ignore_dependencies=self.ignore_dependencies)) File "/home/deeplearning/anaconda2/envs/tensorflow/lib/python2.7/site-packages/pip/req/req_set.py", line 620, in _prepare_file session=self.session, hashes=hashes) File "/home/deeplearning/anaconda2/envs/tensorflow/lib/python2.7/site-packages/pip/download.py", line 821, in unpack_url hashes=hashes File "/home/deeplearning/anaconda2/envs/tensorflow/lib/python2.7/site-packages/pip/download.py", line 659, in unpack_http_url hashes) File "/home/deeplearning/anaconda2/envs/tensorflow/lib/python2.7/site-packages/pip/download.py", line 882, in _download_http_url _download_url(resp, link, content_file, hashes) File "/home/deeplearning/anaconda2/envs/tensorflow/lib/python2.7/site-packages/pip/download.py", line 605, in _download_url consume(downloaded_chunks) File "/home/deeplearning/anaconda2/envs/tensorflow/lib/python2.7/site-packages/pip/utils/__init__.py", line 852, in consume deque(iterator, maxlen=0) File "/home/deeplearning/anaconda2/envs/tensorflow/lib/python2.7/site-packages/pip/download.py", line 571, in written_chunks for chunk in chunks: File "/home/deeplearning/anaconda2/envs/tensorflow/lib/python2.7/site-packages/pip/utils/ui.py", line 139, in iter for x in it: File "/home/deeplearning/anaconda2/envs/tensorflow/lib/python2.7/site-packages/pip/download.py", line 560, in resp_read decode_content=False): File "/home/deeplearning/anaconda2/envs/tensorflow/lib/python2.7/site-packages/pip/_vendor/requests/packages/urllib3/response.py", line 357, in stream data = self.read(amt=amt, decode_content=decode_content) File "/home/deeplearning/anaconda2/envs/tensorflow/lib/python2.7/site-packages/pip/_vendor/requests/packages/urllib3/response.py", line 324, in read flush_decoder = True File "/home/deeplearning/anaconda2/envs/tensorflow/lib/python2.7/contextlib.py", line 35, in __exit__ self.gen.throw(type, value, traceback) File "/home/deeplearning/anaconda2/envs/tensorflow/lib/python2.7/site-packages/pip/_vendor/requests/packages/urllib3/response.py", line 246, in _error_catcher raise ReadTimeoutError(self._pool, None, 'Read timed out.') ReadTimeoutError: HTTPSConnectionPool(host='storage.googleapis.com', port=443): Read timed out. (tensorflow) deeplearning@deeplearningsinglenode:~$
注意：这是在安装tensorflow的时候创建tensorflow环境失败，这是个坑，因为有些版本地址失效了。
换其他版本试试。比如如下我现在是2017年12月份，采用conda方式安装tensorflow，版本已经是1.4.0-py27_0
3.2 conda方式（或者也可以这种方式来安装）
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.
$ 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
(tensorflow) deeplearning@deeplearningsinglenode:~$ conda install -c conda-forge tensorflow Fetching package metadata ............. Solving package specifications: . Package plan for installation in environment /home/deeplearning/anaconda2/envs/tensorflow: The following NEW packages will be INSTALLED: bleach: 1.5.0-py27_0 conda-forge enum34: 1.1.6-py27_1 conda-forge funcsigs: 1.0.2-py_2 conda-forge futures: 3.2.0-py27_0 conda-forge html5lib: 0.9999999-py27_0 conda-forge intel-openmp: 2018.0.0-hc7b2577_8 markdown: 2.6.9-py27_0 conda-forge mkl: 2018.0.1-h19d6760_4 mock: 2.0.0-py27_0 conda-forge numpy: 1.13.3-py27hbcc08e0_0 pbr: 3.1.1-py27_0 conda-forge protobuf: 3.5.0-py27_0 conda-forge six: 1.11.0-py27_1 conda-forge tensorboard: 0.4.0rc3-py27_0 conda-forge tensorflow: 1.4.0-py27_0 conda-forge webencodings: 0.5-py27_0 conda-forge werkzeug: 0.12.2-py_1 conda-forge Proceed ([y]/n)? y intel-openmp-2 100% |#################################| Time: 0:00:01 478.61 kB/s mkl-2018.0.1-h 100% |#################################| Time: 0:01:08 2.84 MB/s enum34-1.1.6-p 100% |#################################| Time: 0:00:01 32.00 kB/s funcsigs-1.0.2 100% |#################################| Time: 0:00:00 38.56 kB/s futures-3.2.0- 100% |#################################| Time: 0:00:00 74.10 kB/s markdown-2.6.9 100% |#################################| Time: 0:00:01 73.17 kB/s six-1.11.0-py2 100% |#################################| Time: 0:00:00 62.19 kB/s webencodings-0 100% |#################################| Time: 0:00:00 25.65 kB/s werkzeug-0.12. 100% |#################################| Time: 0:00:14 17.24 kB/s html5lib-0.999 100% |#################################| Time: 0:00:04 39.10 kB/s bleach-1.5.0-p 100% |#################################| Time: 0:00:00 66.33 kB/s protobuf-3.5.0 100% |#################################| Time: 0:00:47 128.41 kB/s tensorboard-0. 100% |#################################| Time: 0:00:22 77.40 kB/s pbr-3.1.1-py27 100% |#################################| Time: 0:00:02 41.01 kB/s mock-2.0.0-py2 100% |#################################| Time: 0:00:03 30.23 kB/s tensorflow-1.4 100% |#################################| Time: 0:03:53 153.09 kB/s (tensorflow) deeplearning@deeplearningsinglenode:~$ (tensorflow) deeplearning@deeplearningsinglenode:~$
上面的步骤完成后，从conda环境中退出:
(tensorflow)$ source deactivate
第四步、测试安装是否成功
首先激活 tensorflow 环境，然后进入 python，最后导入 tensorflow 库。如果导入成功则表明安装成功。
(tensorflow) deeplearning@deeplearningsinglenode:~$ source deactivate deeplearning@deeplearningsinglenode:~$ deeplearning@deeplearningsinglenode:~$ deeplearning@deeplearningsinglenode:~$ source activate tensorflow (tensorflow) deeplearning@deeplearningsinglenode:~$ (tensorflow) deeplearning@deeplearningsinglenode:~$ python Python 2.7.14 |Anaconda, Inc.| (default, Nov 20 2017, 18:04:19) [GCC 7.2.0] on linux2 Type "help", "copyright", "credits" or "license" for more information. >>> import tensorflow as tf >>> hello = tf.constant('Hi,TensorFlow!') >>> sess = tf.Session() 2017-12-04 19:18:08.790862: I tensorflow/core/platform/cpu_feature_guard.cc:137] Your CPU supports instructions that this TensorFlow binary was not compiled to use: SSE4.1 SSE4.2 AVX AVX2 FMA >>> print sess.run(hello) Hi,TensorFlow! >>>
第五步、需要使用 TensorFlow 的时候必须重新激活
当使用完毕后，关闭 tensorflow 环境。
Use exit() or Ctrl-D (i.e. EOF) to exit >>> exit() (tensorflow) deeplearning@deeplearningsinglenode:~$ (tensorflow) deeplearning@deeplearningsinglenode:~$ (tensorflow) deeplearning@deeplearningsinglenode:~$ (tensorflow) deeplearning@deeplearningsinglenode:~$ source deactivate deeplearning@deeplearningsinglenode:~$
然后你的终端提示符就会变会原的样子。
当你需要再次使用的时候就必须再次激活 tensorflow 环境。
source activate tensorflow
..........
......
关闭 tensorflow 环境，并重新激活
第五步、 Finally
至此，你已经拥有了一个可以玩耍机器学习的 tensorflow 环境，好好玩耍吧：）
你可以参照官方文档快速的运行一个手写数字识别的示例。友情提示：仅 CPU 版本你需要有足够的耐心。。。。。。
本文转自大数据躺过的坑博客园博客，原文链接：http://www.cnblogs.com/zlslch/p/7975791.html，如需转载请自行联系原作者
