Installing Steps:

TensorFlow is one among the leading frameworks for deep learning and machine learning, and as of 2017. It is now integrating for Intel® Xeon Phi™ processors. 

Here are the steps provided to provide a quick overview of the installation process in distinct operating systems using Python dependencies.

Installing TensorFlow on Windows:

Step1: A 64-bit Python 3.5.x window is the only version that TensorFlow supports.
Head over to Python 3.5.x from python.org
Step 2:  Select the 3.5.2 download from downloads
Step 3:  Select either the x86-64 or amd64 installer like Windows x86-64 executable installer
Step 4:  Choose Add Python 3.5 to PATH
Step 5:  Now a message appears as “Setup was successful.”

Open Command Prompt and check the version to confirm that the installation is successful.

Open Command Prompt

Installing TensorFlow on Mac

Start with installing Homebrew (brew), which makes it easy to install a large number of different packages. Installing brew is simple:

Now, install Python 2

brew install python

Now install pip using the following command:

<span style="color: #0000ff;">sudo apt-get install python-setuptools</span><br><span style="color: #0000ff;">sudo easy_install pip</span>

Subscribe to our youtube channel to get new updates..!

Go to finder and create a folder to install everything. As soon as you create your virtual environment, a few folders should appear inside your folder.

<span style="color: #0000ff;">virtualenv –system-site-packages SOME_PATH/SOME_FOLDER </span>

Instead of entering the whole folder path every time, macOS provides a current directory with the “cd” command. Simply type “cd” with space at the end and hit enter.

After that go to finder, and then drag the folder to your terminal window that enables you to paste the absolute folder path to the end of your input.

<span style="color: #0000ff;">cd /SOME_REALLY_LONG_PATH/SOME_FOLDER</span>

 Now activate the virtual environment that you have installed on the folder

<span style="color: #0000ff;">source bin/activate</span>

Now download the binary files from Google to install TensorFlow

<span style="color: #0000ff;">Export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/tensorflow-0.9.0rc0-py2-none-any.whl</span>

Install them

<span style="color: #0000ff;">sudo pip install –upgrade $TF_BINARY_URL</span>

With this, installing Tensorflow on macOs is successful.

Installing TensorFlow on Ubuntu

Before installing TensorFlow on Ubuntu, some prerequisites are required as follows:

An Ubuntu 16.04 server with 1GB of RAM, following Ubuntu 16.04 initial server setup guide, a sudo non-root user and at the firewall. 

  • Install Python 3.3 or higher version is required.

Install Git

Step 1: Installing TensorFlow,

Create a project-directory tf-demo

<span style="color: #0000ff;">$ mkdir ~/tf-demo</span>

Navigate to the newly created directory

<span style="color: #0000ff;">$ cd ~/tf-demo</span>

With this, a new tensorflow-dev directory that contains the packages you install while the environment is activated, including a standalone version of Python and pip.

<span style="color: #0000ff;">$source tensorflow-dev/bin/activate</span>
<span style="color: #0000ff;">(tensorflow-dev)username@hostname:~/tf-demo $</span>

Now, install TensorFlow in your virtual environment.

To install and upgrade TensorFlow newest version in PyPi, run the following command

<span style="color: #0000ff;">(tensorflow-dev) $pip3 install --upgrade tensorflow</span>

After successful installation, the output will be:

Collecting tensorflow Downloading tensorflow-1.4.0-cp36-cp36m-macosx_10_11_x86_64.whl (39.3MB)  100% |????????????????????????????????| 39.3MB 35kB/s
Successfully installed bleach-1.5.0 enum34-1.1.6 html5lib-0.9999999 markdown-2.6.9 numpy-1.13.3 protobuf-3.5.0.post1 setuptools-38.2.3 six-1.11.0 tensorflow-1.4.0 tensorflow-tensorboard-0.4.0rc3 werkzeug-0.12.2 wheel-0.30.0

With this, the TensorFlow installation is completed. Make sure that it is working.

Installing TensorFlow, TensorFlow Installing, steps of TensorFlow, TensorFlow Installing steps,