A 10-minute guide to setup Machine Learning and Deep Learning on your laptop

Tanya Dixit
Sep 9, 2018 · 3 min read

I know it is quite cumbersome to set up your laptop before starting projects. But luckily for ML/DL projects, it doesn’t have to be this way.

I have heard people saying that it took them 2–3 hours to setup everything on their laptops when they were starting ML projects. That might have been true some years back. But you are definitely doing something wrong now if you face something like this today.

I will take you through a 10-minute guide to setup your machine. Please note that this doesn’t include the installation time. This just includes the time that you will be kept engaged. Also please note that this is a Windows guide.

Step 1: Install Python

You can download the latest version of Python from here : https://www.python.org/downloads/

Press the download button. It will download an exe file that you can run. Keep pressing next, and after the installation, don’t forget to update the system and environment variables.

In environment variables you will see a variable called “Path”. Just update it with C:\Python{version}, C:\Python{version}\Lib\site-packages, C:\Python{version}\Scripts

So I have a python 2.7 version. So my Path variable looks something like in the picture above.

Step 2: Download Anaconda

You can download Anaconda from the link above. It will take about 15–20 minutes to install.

After anaconda is installed, open the anaconda prompt.

Anaconda has something called an environment. So suppose you have python 2.7 on your laptop (as I have), but I need to use libraries that work only on python 3.6 let’s say. You can create an environment which is like an isolated virtual container where you can install python 3.6 and install your libraries and code away.

conda list envs — Lists all the environments created in Anaconda

conda create -n env_name — Creates a new environment

conda activate env_name — To activate and start working in a particular environment

Now the great thing about Anaconda is that it’s really great for installing python libraries and dependenies.

So basically if I want my next project to be in Keras, I can directly give the command : conda install keras

and Anaconda will take care of the correct version of python, update it in your environment, and install tensorflow as well if not installed.

With this, you can start your python projects and start coding. Your environment is ready with all the libraries you’ll need.

For actual coding, you can use iPython Notebooks. They are a great tool for prototyping projects. More about jupyter notebooks in my next post.


Thanks for reading!!!

Coinmonks

Coinmonks is a non-profit Crypto educational publication. Follow us on Twitter @coinmonks Our other project — https://coincodecap.com

Tanya Dixit

Written by

I live to write and I write to live

Coinmonks

Coinmonks

Coinmonks is a non-profit Crypto educational publication. Follow us on Twitter @coinmonks Our other project — https://coincodecap.com