How To Become A Self-Taught Machine Learning Engineer (For Free)

What should you do to become a self-taught Machine Learning Engineer?

Learn Mathematics

Mathematics is at the core of any Machine Learning or Deep Learning. Therefore, you must understand the mathematics behind Machine Learning to develop intelligent systems.

It doesn't mean you need to spend plenty of your time learning complex mathematical concepts. Just some high school mathematics will be enough in most cases.

At the bare minimum, the maths you should learn includes:
    Linear Algebra: It is at the core of regression algorithms.

    Probability and Statistics: Understanding probability theory will help better understand how to                                                         select the appropriate algorithm with precise accuracies, number of                                                          parameters involved in training time, and model complexity.
    Calculus: It plays a vital role in gradient computations and numerical optimizations.

Most of the ML enthusiasts dive into the algorithms directly and tend to ignore the mathematics requirement mentioned above. So, make sure you don't skip this part.

Dive into Machine Learning

As you are self-learning you need to follow some books/courses to grasp ML concepts and find a quick way to implement them.

My top ML books I usually recommend are:

Hands-On Machine Learning with Scikit Learn and Tensorflow

Introduction to Statistical Learning: with Applications in R

Deep Learning by Ian Goodfellow

Some essential topics

These are building block topics that collectively represent the simple value proposition of machine learning: taking data and transforming it into something useful.

Machine Learning projects

Head to Kaggle and begin to tackle a complete ML project that ML engineers like you are solving. Some kaggle projects to get you started are listed below:

1. Supervised Learning

a) Regression Problems

How much did it rain :- https://www.kaggle.com/c/how-much-did-it-rain-ii/overview
Inventory Demand:- https://www.kaggle.com/c/grupo-bimbo-inventory-demand
Property Inspection predictiion:- https://www.kaggle.com/c/liberty-mutual-group-property-inspection-prediction
Restaurant Revenue prediction:- https://www.kaggle.com/c/restaurant-revenue-prediction/data
IMDB Box office Prediction:-https://www.kaggle.com/c/tmdb-box-office-prediction/overview

b) Classification problems

Employee Access challenge :- https://www.kaggle.com/c/amazon-employee-access-challenge/overview

Titanic :- https://www.kaggle.com/c/titanic
San Francisco crime:- https://www.kaggle.com/c/sf-crime
Customer satisfcation:-https://www.kaggle.com/c/santander-customer-satisfaction
Trip type classification:- https://www.kaggle.com/c/walmart-recruiting-trip-type-classification
Categorize cusine:- https://www.kaggle.com/c/whats-cooking

2. Unsupervised Learning

Vehicle Identification:- https://www.kaggle.com/c/st4035-2019-assignment-1

At this point, you already have a good grasp on ML algorithms and practices.

I hope the article has been informative.

In case you need clarifications, please get back to us in the comments section.