Apart from my univesity curriculum, I do spend time learning stuff online in my free time. However, there are countless resources which you can get online, be it online courses, videos, podcasts, blogs etc which’d help you to learn stuff. In this article, I’d mention the resources which I used along with their concise description. This article is indeed a live one, and would get timely updates after completing new milestones :). As of now, the list mainly consists of the data science related stuff.
Online Courses:-
-
Elements of AI: This was my first introduction to this new world where I came to know about Artificial Intelligence, Machine Learning, Deep Learning, Alan Turing, etc. It is a completely theoretical course and the main purpose is to get you acquainted with these buzz words. There are mentions about how this field evolved with time and contributions of some prominent researchers as well. There is a high level desciptions about Neural Networks and also about their cool applications. Do go for it if you don’t have in-depth idea about what these buzzwords mean and are curios to know them.
-
Machine Learning: Arguably one of the most famous course on Machine Learning out there on the internet. This can be attributed to the popularity of the instructor Andrew Ng, a world class AI researcher. This course is a subtle introduction to Machine Learning, and introduces about the algorithms which facilitate predictions. This course has assignments as well as quizzes which in my opinion are on the easier side, but nevertheless they are worth your time. This course is for those who aren’t familiar with Machine Learning algorithms as such. If you aren’t in this category than you may choose not to do it and proceed for the next one. Another reason for the popularity of this course is the pre-requisites is Introductory Linear Algebra and some knowledge about algorithms. However, this course uses Octave/Matlab as languages for Programming Assignments contrary to Python and R which are the languages which are mostly used.
-
CS109-Data Science: One of the not-so-popular online course but, believe me it is one of the best. The reasons why I am saying this is that it covers everything, you should have knowledge about if you want to become a data scientist someday. You’d learn Data Scraping, Data Processing, Data Visualization, Machine Learning, and Storytelling! The homework assignments are awesome projects on real datasets. However, the prereqs for this is Probability and Statistics which’d be covered in any intro-level course in your college. All the assignments use Python 2, and you’d have to figure out the Python3 equivalent, but it shouldn’t be too much of an issue. I highly recommend going for this if you satisfy the pre-reqs.
-
Introduction to Deep Learning: It’s an okayish course and it briefly introduces the various neural net architectures. The framework used for assignments is Tensorflow, and Keras. I found their explanation of backpropogation very cool. Their explanation of TensorFlow as a framework is also very lucid and intuitive. In rest of the course, they discuss CNNs, LSTMs, RNNs, Autoencoders, etc. Some of the assignments can be difficult for those who are using TensorFlow for the first time or using Neural Nets for the first time.
-
How to Win a Data Science Competition: This is an amazing course, targeted for those who know how to use Machine Learning in Python, but they want to hone their skills. What I mean is simply using model.fit() followed by model.predict() isn’t sufficient. There are several other aspects such as Feature Engineering, Hyperparameter tuning, Model Selection, Validation Strategies, etc which one needs to take care of while developing models which generalize well. This course covers all these other aspects. To be honest, this course is mainly for people who want to compete in data science competitions and get good scores but it’d certainly help you become a better data scientist at your workplace.
Blogs to Follow:
- Towards Data Science: If you’re a newbie in this field and don’t understand any particular concept related to probability, statistic, machine learning, etc. this is the place where you’d most probably find a blog post related to that. The writers also write about new cool applications and what it feels to work in the industry.
- Machine Learning Mastery - By Jason Brownlie Ph.D: Here you’d find many relevant blog posts for Machine Learning.
- Kaggle Discussion: Kaggle organizes data science competitions but apart from that it has an excellent forum for discussing the advances in this area. You’d be able to learn new tricks and hacks from the experts and also interact with them.