What is this?
A free multi-month self-study curriculum that takes a working software engineer from mobile development to a machine learning engineer job, built as a GitHub checklist of articles, books, courses and competitions.
A free multi-month self-study curriculum that takes a working software engineer from mobile development to a machine learning engineer job, built as a GitHub checklist of articles, books, courses and competitions.Is it a course I enroll in, or how is it organized?Not a course — a curated list on GitHub: a table of contents leads from ML overviews and algorithms through beginner and practical books to Kaggle competitions, video series, MOOCs, podcasts and interview questions.
Not a course — a curated list on GitHub: a table of contents leads from ML overviews and algorithms through beginner and practical books to Kaggle competitions, video series, MOOCs, podcasts and interview questions.How do I actually work through it?Tackle the items in order from top to bottom: each day pick one subject, read it cover to cover, take notes, do the exercises and write an implementation in Python or R. Fork the repo and check items off with GitHub task lists in your own branch.
Tackle the items in order from top to bottom: each day pick one subject, read it cover to cover, take notes, do the exercises and write an implementation in Python or R. Fork the repo and check items off with GitHub task lists in your own branch.Do I need strong math or a CS degree before starting?No — it was written by a self-taught mobile developer without a CS degree and deliberately goes hands-on, abstracting most of the math. The author started from only basic university calculus, linear algebra and statistics.
No — it was written by a self-taught mobile developer without a CS degree and deliberately goes hands-on, abstracting most of the math. The author started from only basic university calculus, linear algebra and statistics.How long is this going to take?Years at the author's pace of roughly 4 hours of study per night after a full work day — though if you already know much of the material, it will take a lot less.
Years at the author's pace of roughly 4 hours of study per night after a full work day — though if you already know much of the material, it will take a lot less.Is there anything to do before the daily plan itself?Yes, a short prerequisites section: posts on how data analytics, data mining, data science, ML and big data differ, plus meta-skills like Learning How to Learn, learning on your own and Don't Break The Chain.
Yes, a short prerequisites section: posts on how data analytics, data mining, data science, ML and big data differ, plus meta-skills like Learning How to Learn, learning on your own and Don't Break The Chain.Why a top-down, practice-first plan instead of the usual math-first route?The author kept getting discouraged by books and courses that demand multivariate calculus and inferential statistics up front. His method is practice — learning — practice: start with real projects to get familiar with the methods, then study the underlying theory to guide deeper practice.
The author kept getting discouraged by books and courses that demand multivariate calculus and inferential statistics up front. His method is practice — learning — practice: start with real projects to get familiar with the methods, then study the underlying theory to guide deeper practice.Will I have to pay for any of the resources?Essentially no: even the videos locked behind Coursera or EdX classes are free to access. The catch is that a class may not currently be in session, so you might have to wait a couple of months to enroll.
Essentially no: even the videos locked behind Coursera or EdX classes are free to access. The catch is that a class may not currently be in session, so you might have to wait a couple of months to enroll.Does the plan actually aim at getting hired, not just learning?Yes — it opens with the question of whether you can get an ML job without a CS master's or PhD, and it includes sections of interview questions, stories from self-taught engineers, and Kaggle knowledge competitions for hands-on practice.
Yes — it opens with the question of whether you can get an ML job without a CS master's or PhD, and it includes sections of interview questions, stories from self-taught engineers, and Kaggle knowledge competitions for hands-on practice.Which books does the plan build around?It pairs beginner-friendly reads (Data Smart, Data Science for Business, Predictive Analytics) with practical ones that have companion GitHub repos, like Python Machine Learning, Machine Learning in Action and An Introduction to Statistical Learning.
It pairs beginner-friendly reads (Data Smart, Data Science for Business, Predictive Analytics) with practical ones that have companion GitHub repos, like Python Machine Learning, Machine Learning in Action and An Introduction to Statistical Learning.What are similar projects?Machine-Learning-Tutorials, go, d2l-en, DeepLearning, data-science, game-programmer, professional-programming, 100-Days-Of-ML-Code
Machine-Learning-Tutorials, go, d2l-en, DeepLearning, data-science, game-programmer, professional-programming, 100-Days-Of-ML-CodeWhich topics is it in?kaggle, r, machine-learning-algorithms, roadmap, english, interview-questions, curriculum, career, data-science, book, artificial-intelligence, guide
kaggle, r, machine-learning-algorithms, roadmap, english, interview-questions, curriculum, career, data-science, book, artificial-intelligence, guide