My Machine Learning journey (so far)

Posted: February 19, 2022

In this post, I will share with you the 2 years training plan I took to start in Machine Learning. The plan itself evolved as I went through my journey.

TL;DR

In a chronological order:

  • Machine Learning by Andrew Ng (Coursera)
  • Applied Data Science with Python - Specialization (Coursera) :
    1. Introduction to Data Science in Python
    2. Applied Plotting, Charting & Data Representation in Python
    3. Applied Machine Learning in Python
    4. Applied Text Mining in Python
    5. Applied Social Network Analysis in Python
  • AWS Certified Developer - Associate (ACG)
  • Machine Learning by Andrew Ng (Coursera) I took notes this time 😅
  • AWS Certified DevOps Engineer - Professional (ACG)
  • Deep Learning Specialization (Coursera or DeepLearning.AI)
    1. Neural Networks and Deep Learning
    2. Improving Deep Neural Networks: Hyperparameter tuning, Regularization, and Optimization
    3. Structuring Machine Learning Projects
    4. Convolutional Neural Networks
    5. Sequence Models
  • AWS Certified Machine Learning - Specialty (ACG)

Why I chose these courses ?

I started with the Machine Learning course by Andrew Ng on Coursera. This was a recommendation by a friend who at the time was already ankle-deep in Machine Learning. This course is more than just an introduction, it’s a good starting point to build a solid foundation, it covers among other things: types of learning, most common learning algorithms, model evaluation, etc. I later completed this course a second time.

The second course I took was the Applied Data Science with Python Specialization also on Coursera. This series of courses starts by introducing Data Science as a whole, then diving into Machine Learning, this course also covers plotting, charting and data representation as well as text mining. Contrary to the first course, which was more on the theoretical side, this one is very hands-on.

Coming from a DevOps background I know how important automation and tooling are, nowadays developing a piece of software is not enough, building a Continuous Integration/Continuous Delivery pipeline is also required. To solve this issue I have decided to take advantage of AWS.

Using my free tier account and the sandboxes provided by ACloudGuru, I was able to understand the services provided by AWS well enough to take and pass the certification exams.

The Deep Learning Specialization on Coursera was by far the most complete series of courses out there, it dives into CNNs, RNNs, optimization, etc. You must check it out.

My advice to you 📓

If you are planning on starting your journey in Machine Learning, my advice to you is to revise the basics and don’t be afraid of the math, because the algorithms will only get more complex, if you get familiarized with the math from the start, you will have a better chance to grasp the concepts later.

Also, if you're planning to land a job as a Machine Learning Engineer (or any other tech job actually) start building a portfolio as soon as possible, most recruiters tend to regard the previous experiences as irrelevant to the new role you’re seeking.

My last piece of advice is to always take notes, with time, some details can fade from our memory, so it's always nice to write down the important stuff.