"Machine Learning is one of the grossing field, where computer learns to perform a specific task without intervention of humans."
Generally learning machine learning is not very difficult if you know the pre-requisites. Many people reached out to me about how to learn machine learning quickly over Linkedin. I thought to create an article for that, So let's start with some prerequisites required for learning Machine learning.
How to learn Machine Learning - Start with Prerequisites:
1. Motivation :) :
The first and foremost prerequisite is "Motivation". If you have enough motivation within you, that you can learn Machine learning Congrats! you are fulfilling the first prerequisite criterion.
Note: Most websites directly start with other prerequisites without knowing that this is the most essential criterion.
Mathematics is the backbone of machine learning. You need to be comfortable with the numbers, to excel machine learning. If you are a computer programmer or a student, you need to brush up your mathematics.
Topics of Mathematics:
- Linear Algebra
- Probability & Statistics.
Having a question? Where to learn Mathematics?
There are different types of learners, some prefer to learn from books, some from MOOCs, etc. So, I am specifying some of the Books & MOOCs for mathematics.
Best Moocs for Mathematics:
Linear Algebra - Khan Academy (Course)
3Blue1Brown - Essence of Calculus (Course).
Statistics and Probability - Khan Academy (Course)
Best Books for Mathematics:
Introduction to Linear Algebra by Gilbert Strang (Book)
Naked Statistics by Charles Wheelan (Book)
Calculus by Michael Spivak (Book)
3. Choose Programming Language:
Now, Pick up a programming language of your choice. If you are a complete novice then choose Python, it's really easy and powerful. If you already know the programming language then you are good with this step.
After getting some grip on a programming language, start with the Data Science with Python - Numpy, Seaborn, Matplotlib, Pandas, sklearn, etc.
Data Science With Python - University of Michigan (Coursera)
Introduction to Data Analysis (Udacity)
Python Data Science Handbook: Essential Tools for Working with Data
Now you are done with all the prerequisites. Let's start with Machine Learning.
The two aspects of Machine Learning - One is theory and one is practical. If you know the intuition or theory behind the algorithm, the practical implementation is very simple in python (Scikit learn).
The best course on the Internet for developing the intuition behind ML is Andrew NG's Stanford youtube playlist, not the Coursera one. Don't worry too much about mathematics, just try to learn the intuition behind the algorithm.
For practicals, Machine Learning A-Z is best
Best MOOC for Theory: Andrew NG Stanford Lectures [Link]
Best MOOC for Practical: Machine Learning A-Z [Link]
- Don't focus too much on the syntax of the code. Just start with the flow. With the passage of time, you will be an expert.
- Start contributing to Kaggle, as soon as you start the Data Science Basics.
- Don't lose motivation, you can't become a master of Machine learning overnight :). No course will help you, which states learn machine learning in 24 hours.
- Talk to the Data science people make your connections.
I have connected with many data scientists, with the majority of voting - created a curriculum for the machine learning journey. Please follow this post - " Best Machine Learning Curriculum for Beginners � Level 101 "
If you still have queries, please feel free to drop a comment below.