“In eight months” too, was supposed to be in the title. But I spared it for two reasons. One, that would make the title too long. Two, in this big Internet world which is full of blogs, there are already thousands of those which are about Machine Learning and Data Science. And people(like me) who call themselves Machine Learning aspirants and Data Science enthusiasts would rather not like to read a blog which tells them how to go from zero to a beginner in EIGHT months. But that’s my journey so far, in my quest for getting into this fast-growing and all-happening field and I’m not shy to say it out loud for the whole Internet.
One day when I was on my way to office(I have a full time job), sharing my ride in a cab, I made an acquaintance who in the same week became my ML-friend. It didn’t take more than few minutes for the conversation to be led into yet another discussion on how to get started with Machine Learning. By then, we both were 2–3 months into Machine Learning, researching on finding more helpful resources, trying to find a course or a book or a blog which will neither be too much mathy nor too naive. We both had something in common, which was Andrew Ng’s Machine Learning course from Coursera. We both started it, loved it, finished the first few weeks and then shelved it for that moment, to go collect the necessary Math wisdom and then return back to it. There is now a new course on Deep Learning from Andrew and I will be getting back to his courses soon.
One of the main problems that everyone who tries to learn and know Machine Learning on their own faces is the availability of so many resources, free resources. MOOCs, Youtube tutorials, blogs, posts in social networking sites. Having so many of them at hand, we as newbies take it easy and granted when one of the resource we chose doesn’t meet our taste or expectation. “This course has too much math, let me try another.” . “This course is in R, I should find some other which is in Python”. “Too much theory, how can I get into Machine Learning without any hands on. I will try the hands on course from EDX instead”
Having said that, there is no definite path for Machine Learning and Data Science which would work the same for everyone. Programming experience, and prior knowledge in Statistics and Mathematics are two major deciding factors which will surely make a difference in how easy or hard it’s going to be for someone to get into this field. But that doesn’t mean you should start with Calculus, then Probability and Statistics before you get started with Machine Learning. Though Mathematics can be a great tool if you can have it in your inventory, it doesn’t mean you can’t proceed without it. Specially if you are like my friend who says, “I don’t care about how the algorithm is built. Tell me what it does and how can I use it to get the task done”.
There are a lot of blogs which teach and explain many concepts of Machine Learning in an intuitive and understandable way and also blogs that contain list of so many resources that help in learning and practising Machine Learning. I owe a lot to such blogs for many of the things that I know and learnt about Machine Learning and Data Science. In the next part of this blog(or Story as it says here in Medium) instead of sharing an exhaustive list of tens and hundreds of resources that are available, I will be sharing only the specific resources which include Youtube series, MOOCs, blogs, research papers that really helped me in knowing about Machine Learning. Will be posting the next part soon.
Updated : It doesn’t really take months to get started with Machine Learning. More about it in my next blog here.
Information is Bliss. Share what you know!