Machine learning is the future and I’m not the first person to tell you that, I know!
But it is true all those ads you come across all the interviews that say that Artificial Intelligence and Machine learning is the future it all is true and it’s high time you get into it.
Do you sometimes wonder how Netflix shows just the right movie for you or how most of the recommended videos are often what you wanted to watch, Aren’t you fascinated by the fact that we soon will be riding in a driverless car? These all are not miracles or sheer coincidences these are made possible using Machine learning. Now hop on, let’s start this guide!
Artificial Intelligence, Machine learning, and Data Science are not the same. So first let’s see the differences in these fields.
Artificial Intelligence VS Machine Learning VS Data Science
Here is the best way to put it :
- Data science produces insights
- Machine learning produces predictions
- Artificial intelligence produces actions
Taking the example of Self-driving Cars:
- Collecting the data, analyzing which one will be more useful, making the data better for our model building, these things come under Data Science.
- From the data we have we get after applying Data Science on it, we will build our model then we will try to maximize its accuracy, this comes under Machine Learning.
- Now comes the last and final step, which is to use this model and apply it in the real world, and this comes under Artificial Intelligence.
If you want to explore these differences to the core, here is a good article for you AI vs ML vs DS.
Why should I learn machine learning?
- Jobs on the rise: With every company wanting to employ AI and regulate their work using computers via Machine learning in their domain, there is a high demand for machine learning engineers in the industry.
- A decent pay: It is one of the most handsomely paid industry with an average salary of a machine learning engineer(fresher) is around ₹745k.
- It helps increase efficiency: Sometimes Machine learning models can make even better decisions than a human being. And so in a lot of fields such as advertisements, market-strategy for malls and stores, online business, etc Machine learning can be of great help.
Where to start?
- Pick a programming language: First things first if you are going to enter any field in the IT industry it’s essential to learn a programming language, for machine learning you can choose between Python, R, Java, or Scala.
The best choice (in my opinion) would be Python as it is the easiest to learn and has a lot of libraries mainly for Machine learning.
If you want to start learning Python, this is all you need: https://blog.usejournal.com/learning-python-the-why-and-where-87e04347c2dc
- Study the required mathematics: Mathematics is one of the most important things for a machine learning career, so be well equipped with knowledge of all the mathematical knowledge required, here are some resources for you if you want to start :
‣ StatQuest with Josh Starmer (Youtube Channel)
‣ Introduction to Statistics on Udacity
‣ Statistics Lectures by Harvard
For a detailed analysis and description of the mathematics, you can visit this medium post >> The Mathematics of Machine Learning
- Start the beginner’s courses: There are a lot of courses for beginners out there once you have a good grasp of the Mathematics and programming language then start these courses.
Here are some of the best courses out there :
- Machine learning by Andrew Ng
- Machine learning with python by IBM
- Machine Learning A-Z™: Hands-On Python & R In Data Science
Some Tips and Tricks :
- Don’t just watch these videos/courses make notes, proper notes so that you can rely on them whenever you need to have a quick look or revision.
- Andrew Ng is the Legend in this field, I will suggest you start with his courses.
- If you’re going with Coursera, the programming exercises will be hard and they’ll get harder each passing week, bear with it, DO NOT GIVE UP, just try your hardest and take help from your friends, look for your problem in the discussion forum.
- There will be times you won’t understand a few things, for those times here’s a quote for you!
- Try to make projects with everything you learn, even the smallest will do, just try to implement them by yourself, and you’ll see the effect of this soon.
- And lastly, the thing I wish I did when I was learning was to get on Kaggle! Don’t worry if you think you’re a rookie, just read a lot of kernels there, see how people are implementing machine learning, and try to start with the Titanic Problem on Kaggle!
Special Resources: Here are some resources that will help you grasp the knowledge in the best way possible -
In the end, I would just say, go for this field if you have an interest and are ready to give all your heart and soul into it, if you’re just trying this field to fit in the crowd and because ‘everyone else is doing it’, you’re wasting your time and energy.
Here’s a quote to conclude this guide :
The best way to learn is to practice and try out things for yourself as much as you can. For you will learn better by falling than by watching others climb.
Thanks for reading! I hope you found this guide insightful.
In case you want to connect with me, follow the links below →