Learn AI

**** This page is still under preparation ****

Are you interested in learning more about big data, machine learning and deep neural networks? This page lists some of the free online resources for learning and programming machine learning.

Prerequisites

Since the field of machine learning is relatively new and very advanced, you will need some skills before you can actually dive into these online courses.

1. Decent programming skills: More specifically python or c++. You don't need to learn all the object-oriented programming stuff since you only need to write a few lines of code if you are using a machine learning toolkit. But you should know the basic coding principles of python or C++.
2. Understanding the math behind the models you will be using: With the toolkits, you can essentially try anything and sometimes end up with promising results! But if you don't know what you are doing or how/why it worked, then it is pointless, no? It is always better to do some courses (which will be listed below) before shooting in the dark! 
3. A keen observation and patience: Yes! You need a lot of patience since machine learning is not magic. There are a lot of settings to tweak and knobs you can turn! You will sometimes need to try all possibilities before you actually get to what you need. Aaaaaand, observation helps you to arrive their faster. You are dealing with big data and there is no single recipe that works on all datasets.
4. Some high-end computing: You are going to face two monsters: 1) big data and 2) deeeeep neural nets. Both will need a lot of processing power and you will need good computational resources (GPU cards are beasts that can fight the monsters). But don't worry, for the demos and courses, you only need your laptop since they use simple examples.

Online Courses

There are several online courses to choose from. Will categorize them based on the subtopics.

Python

Machine Learning and Data Science

Machine Learning and Python

Hands-on Courses for Toolkits

Machine Learning Toolkits

• Google Tensorflow : Python or C++: Originally developed to solve machine learning problems internally at Google, they eventually decided to release it as open-source. It has a big online community and many people (including me) are using it.
• Microsoft Computational Network toolkit (CNTK) : Much like Tensorflow, this is also a popular toolkit released by Microsoft.