Machine Learning and Deep Learning Environment Setup

In this post, I will go through the python environment setup. I will be doing the steps in Windows, but will work for Linux as well.

Target of this Post - We will have a python working environment to start working on Machine Learning and Deep Learning.

Lets start the installation -

I will divide the installation process into below steps -

1. Downlaod Anaconda
2. Install Anaconda
3. Install CUDA Toolkit
4. Install Deep Learning libraries

Step 1: Download Anaconda - 
Follow below link and download Anaconda for you Platform.
 Download



Step 1: Install Anaconda 
With the downloaded installer, install Anaconda. Steps to install are very easy and straight forward.



 Installation will take more than 5 mins.

After the installation is done, execute below commands to update anaconda

conda update conda
conda update --all

Step 3: Install Cuda ToolKit(Optinal For Nvidia GPU Users)

Download and Install CUDA ToolKit from Download

Select version based on your Nvidia Card.

2. Download cuDNN Download

Choose your version depending on your Operating System. 

Unzipp the folder in C drive as follows

Put your unzipped folder in C drive as follows:

C:\cudnn-9.0-windows10-x64-v7

Add cuDNN to environment variables

C:\cudnn-9.0-windows10-x64-v7\cuda\bin

Create Anaconda Environment

create an environment specific to anaconda which doesn't impact normal enviroment.

Open Anaconda Prompt to type the following commands.

1. Create a conda environment named “tensorflow” (you can change the name) by invoking the following command:

conda create -n tensorflow pip python=3.5

2. Activate the conda environment by issuing the following command:

activate tensorflow
 (tensorflow)C:>  # Your prompt should change

Step 4: Install Deep Learning Libraries

In this step, we will install Python libraries used for deep learning, specifically: TensorFlow

TesorFlow :

 TensorFlow is mainly designed for deep neural network models.

Open Anaconda Prompt and type the following commands.

To install the CPU-only version of TensorFlow:

C:\> pip3 install --upgrade tensorflow

If your machine or system is only CPU supported you can install CPU version for basic learning and practice.

To install the GPU version of TensorFlow:

C:\> pip3 install --upgrade tensorflow-gpu

To test the installation, run this program on shell:

>>> import tensorflow as tf
>>> hello = tf.constant('Hello, TensorFlow!')
>>> sess = tf.Session()
>>> print(sess.run(hello))

For documentation, visit TensorFlow website.