View profile

You can do Machine Learning too.

You can do Machine Learning too.
By Prashant Brahmbhatt • Issue #1 • View online
Sharing my journey into the realm of Machine Learning, citing all the resources and the approach that I took.

Times have certainly changed since I started in Machine Learning, back in January 2018.
The domain is expanding at a humongous rate surely and everybody is trying to squeeze their way in.
But there is always room for quality people. Many people start but don’t keep up once they get past the beginner phase, you avoid that and you’ll be fine.
“When the going is tough the tough gets going!”
I believe that if a not-so-genius student like me can get a Data Science job, you all can! Absolutely!
Here’s how I went from knowing just the basics of programming to someone working in Data Science
The path that I took wasn’t the most optimal to get a good grip on Machine Learning.
Because when I started, I knew nobody that worked or had knowledge of Data Science which made me try all sorts of different things that were not actually necessary.
I studied C programming as my first language during my freshman year in college. And before the start of my second year, I started learning python just because I knew C is not the way to go. I learned it out of curiosity and I had no idea about Machine Learning at this point.
I did not learn Python by courses initially, but by books.
The first book that I read was this, and for reading about some of the most important topics in detail I checked this book.
The approach I took was just to make the same kind of programs that I had made in C. I just used to replace the syntax with that of python and practiced those. The little experience with a programming language certainly made it easier to learn the second one.
For Machine Learning, the first thing I did was to join Andrew NG’s course which really hit it off for me.
I didn’t have to have any idea about Machine Learning for the course. I completed that in almost a month and it gave me a good intuition of things and the flow of ML. It is super intuitive and I absolutely loved it.
Then I thought of implementing those concepts through python But after a lot of tries, I wasn’t really able to do it.
I was constantly encountering a lot of stuff things in code that I wasn’t really aware of. Even loading simple CSV data seemed like a herculean task because of the unfamiliarity.
I figured that what I was lacking were the Python and ML libraries.
So I started focusing on learning those first and foremost.
The absolute necessary ones were
  • Numpy
  • Pandas
  • Matplotlib
  • Seaborn
  • SciKit Learn
  • Os (an important built-in package).
There were two really good Udemy courses that helped me here.
This one helped me learn the basics of the libraries mentioned above, in manipulating the data as well as visualizations.
This helped me implement the algorithms using SkLearn and I also got my first introduction to Keras for Neural Networks here.
The practice of the taught concepts multiple times was necessary, I did that as much as I could and also read blogs on them
Everything that I did from there on was from Kaggle. It is a platform that’s closest to exposing what real-life problems may look like.
A lot of people solve the same problem differently and you have to read a lot of other people’s code. From there, that’s what kept me growing.
Things I learned from Kaggle:
  • Exploration and visualization of data
  • How to approach a new problem
  • Better code structure for implementing a machine learning solution
You don’t have to be an absolute grandmaster of kaggle but plenty of practice and patience is needed.
Then I just picked up some common projects and implemented them, mostly from kaggle. Like,
  • Titanic Survival
  • Spam Classification
  • Movies Recommendation
  • Boston House Pricing
  • Churn Prediction
You’ll kind of know your way from there, moving to harder problems slowly.
Don’t overthink about maths more than it is necessary. You can always learn the mathematical concepts that you may be missing as you go.
Learning the maths behind will keep things interesting if you won’t enjoy that you will get bored pretty quickly.
And trust me on this, I flunked Maths in my junior college and had to redo the year but still, I kept my chin up.
Finally, it can get quite overwhelming at times, make sure to take one step at a time. Don’t focus on how far ahead other people are focus on how much you’ll grow from here.
And remember you are not too old or too young for this stuff. You can do this too!
All the resources I used don’t have to be the same, see what works for you better. Bend things your way.
Thanks for your time and for bearing with me, I hope it was worth it!
Rate if you liked,
Subscribe if you haven’t,
Share if it seems helpful.
Until next time!
Did you enjoy this issue?
Prashant Brahmbhatt

A weekly newsletter focused on exploring and discussing new concepts each week from the field of machine learning and deep learning.

In order to unsubscribe, click here.
If you were forwarded this newsletter and you like it, you can subscribe here.
Powered by Revue