I have been doing machine learning professionally for almost a year(doing deep learning for healthcare startup currently) and I'm completely self-taught.
A quick story on my ML journey. Back in 2011, in my sophomore year, I knew I wanted to be involved in ML professionally; but I didn't know how. My path to becoming proficient in ML had lots of ups and downs. With so much trial and error and almost no guidance. Initially, I wasted a lot of time in dabbling with lots of math heavy books and not knowing where I'm heading. Fast forward 3 years with a concrete plan in mind, I spent close to 2 years rigorously practicing ML(reading books, MOOCs, ML competitions, etc) after work hours and in the weekends.
Moreover, I started off this work as a way to clean up and re-organise the ML notes I have taken in the past 4-5 years. Which I also referred while giving plenty of my ML interviews. And thought about why not make it useful for others as well as a way to benefit their ml journey.
Good luck, may the force be with you.
Inspired from Coding Interview University, machine-learning-for-software-engineers.
This is a multi-month study plan for becoming a professional ML engineer.
The main goal I'm planning to achieve with this repo is to get anyone level up their ML skills in a quick timeframe. Especially, to get anyone make rapid progress in becoming ML professional and not repeat the mistakes that I made in my early years with ML. I have organised this guide in a way it’s open to anyone to get started in their ML journey.
ML is a really broad field, it takes years if not decades to master ML. But it should not be the case for becoming a ML hacker or professional ML engineer.
Typical books and university-level courses are bottom-up. They teach or require the mathematics before grinding through a few key algorithms and theories before finishing up. This can be a good approach if you have the time, patience and appropriate background. Not everyone has so much free time or the desire to move through so much low-level material before getting to the meat and potatoes of applied machine learning.
Taking courses, reading books can be done once you get a grasp of how all the components fit into the big picture. While in the beginning; end of the day one needs to be in a position to know what are the things they know in ML and things they don’t know in ML.
Top-down approach for getting started in applied machine learning actually works. They should feel familiar because it’s probably the same top-down approach that you used to learn how to program. Namely, get the basics, practice a lot and dive into the details later after you’re hooked.
A "hacker", here, is "someone who likes to solve problems and experiment with new technologies", i.e who is interested less on theory but more on solving problems while getting things done. Moreover, this is ideal for programmers from any background.
- First principles
- Machine learning glossary
- ML in Action
From zero(any background, not necessarily CS) to ML professional. I hope this serves as a reference for people who know ML and want to brush up few areas.
- Theory & concepts
- Algorithms
- Interview
- Building Machine learning systems & projects
- Applied ML in practice
From ML professional - ML rockstar
- ML from scratch
- Math of ML
- Scalable ML
- Building ML Products
PS: I'm currently on this path 😃
Please feel free to make any contributions to make this better.