How to get started in machine learning? Python is here because if you are new to machine learning and new to programming then Python would be a really good choice. But really machine learning is all about math so if you already know another language does not worry about learning Python do it in the language that you already know. It's a very versatile language you can do a lot of things with it, but the choice of language is up to you.
If you're going down the Python route and you're going to need to learn it well enough to be able to implement Numpy, pandas, Matplotlib and Scikit-learn, Numb, Polyfill, Linear Algebra, Pandas for data wrangling and data organization. Matplotlib for data visualization and Scikit-learn has all the sort of meaty machine learning functions in it. And its library it's a great resource for machine learning and you're going to need to practice to get your programming up to scratch.
In order to be able to do it well enough to do the machine and you want to do but really as I said machine learning is all about math and you're going to have to learn in linear algebra probability and statistics and calculus. And once you can do all of that then you will be able to have a very good grasp of machine learning and you'll be able to choose which sort of machine learning roots you want to go down. Because it's a vast field and you're never going to be able to do all of it all at once. This will help you to decide where exactly you want to go.
How to learn Python?
There are books that you can use, there are online courses it's so automated the boring stuff with Python is probably the best Python book I have seen it's a great way to learn Python. There's a review of the book on my channel so you can view that here on this youtube channel also the book is available for free online. The contents of the book are available on a website and if you google to automate the boring stuff with Python online version you will find that.
Python crash course is another excellent book I would definitely recommend that effective computation in physics is aimed at physicists that don't know how to code and so it teaches them all the things that physicists would need to know. And that includes most of the stuff that you'd need to be able to do in Python for machine learning. So I definitely recommend that book and then there's learning Python the hard way.
Online courses there's the Udemy automate the boring stuff with Python course which is the companion video course to the book, that is an excellent course as are you to me complete Python Bootcamp that covers absolutely everything. With these Udemy courses every now and then they have special offers on where you can buy them for about $10. So you know that they don't have to be very expensive. The learn Python Website is very good and that is you can see here that covers the Python basics and then goes on to data science tutorials and advanced tutorials. I would certainly recommend that you know for getting started learning Python for applications in machine learning.
There's Google's Python class which is not bad it wouldn't be my first choice but it's definitely worth having a looking at. Then there's my own course which I will let you be the judge of hacker rank is very good for testing your skill with Python coding or with any type of coding. effective computation in physics again for all the reasons. I said before online Udacity has a course intro into data analysis worth taking a look at and it's free and then these two Udemy courses are paid for. There's Python for data science and machine learning that the Bootcamp which is by the same chap that did the complete Python Bootcamp again a very good course and then there's this Udemy machine learning dessert hands.
On Python in data science which is a little dry but covers everything and explains it well linear algebra. If you're looking for a free linear algebra course then this one from University College London is excellent and I definitely recommend it and then, of course, there are all the MIT courses and they have them in just about everything. You're ever going to need they have a linear algebra one too and you can access that here for calculus. If you don't already know calculus MIT has several courses on calculus there's the single variable calculus cause and the multivariable calculus course.
There is a link to the single variable calculus course here Udemy also has some calculus courses which are good but you have to pay for them and then if a probability and statistics. This is an excellent resource information theory inference and learning algorithms by David McKay. This is a free pdf to that book it is quite mathematical so maybe brush up on your math first otherwise you might be put off by just the amount of algebra and equations that are there are in this book. But it is excellent and it has different learning paths for different types of machine learning that you might want to do and I heartily recommend it as one of the best end-to-end foundation courses that there is on machine learning. It's the Andrew neg machine learning course offered by offered through Coursera and created by Stanford University and is one of the most highly regarded introductions to machine learning that exists online.