Machine Learning — Start with Baby Yoda Steps
One simple way to get started with Machine Learning.
Welcome to the world of Machine Learning! If you are new to this field, you might feel overwhelmed by the vast amount of information available, all the fields and subfields, bunch of terms and notations (ML, AI, CV, DL, MLOps, RL, OMG). It is indeed a hot area of research and innovation lately, many of the ML algorithms being able to solve real world problems. Ranging from identifying credit-card fraud, recommending movies you can watch on streaming services, monitoring assembly lines for faulty products, autonomous cars, healthcare services, seeing more of the content you’ve liked on TikTok and many more.
With the latest advancements in todays’ technology there’s a plethora of ML powered tools and solutions that can save you time, make you more productive and increase the quality of your work. Software development, marketing analysis, writing, studying or pretty much any other activity can harness the power of AI.
With this article, I’m aiming to outline some resources, tips and courses that helped me getting started. Although I cover the basics, I’m aware that for some readers who’ve worked with ML a bit, these things will seem trivial and in this case — I’d recommend you to read only the “First and foremost” paragraph, maybe you can find some value there 😉.
First and foremost
Note: Let’s go through some of the core social principles that, I think, are good to keep in mind.
Motivation — Sometimes, It’s ok to feel overwhelmed.
Each journey has it’s ups and downs. ML it’s a complex field, programming, statistics, math, theory etc. When learning something new, and feel like dropping out, always try to compare today version of yourself to yesterday’s version. Even if you did 0.001% progress today, you’re on the right track.
Perspicacity — Keep the end goal in mind.
This is a habit I’ve learned from the book “7 Habits of Highly Effective People — Stephen R. Covey” that states to “Begin With the End in Mind”. Often we get caught-up in minor details and lose our focus, so whenever you feel like you’re not progressing, take a break, relax and look at the bigger picture and what you’re trying to achieve.
Curiosity — Look to grasp what’s under the hood.
Taking everything as it is and not looking into understanding it, is similar to fixing something with duct tape. Always try to ask yourself “why is that?” when learning new concepts, it will serve you well in the long run.
Technical aspects
Programming — Learn Basics of Python 🐍
Python is the easiest multi-purpose programming language you could learn. Simple syntax, easy to understand and highly versatile, it can be used for both small and complex tasks, and is widely popular within data science and machine learning fields. Hence, learning python is a must.
- 📚Learn — Python in 1 Hour (ProgrammingWithMosh)
- Tutorial — Interactive Python Basics
Machine Learning — Try Linear Regression 📈
Linear Regression is the “Hello World” of Machine Learning. It’s a statistical approach that provides a relationship between an independent variable and a dependent variable in order to predict outcomes of future events. In simpler words, it predicts a continuous variable (e.g price, height, sales, marks etc).
- Tutorial — Predicting marks based on hours of study
- Tutorial — Predicting sales based on money spend on marketing platforms
- 📚Learn — Linear Regression in Python (codebasics)
Machine Learning — Try Logistic Regression 🧮
Logistic Regression, compared to Linear Regression helps predicting a discrete output given an input variable, like a classification decision. To put it simply, it returns Yes/No, True/False, 1/0 types of outputs.
- 📚Learn — Simple Classification with Logistic Regression (codebasics)
- Comparison — Logistic vs Linear Regression
Data Visualization — Get familiar with Seaborn 📊
Seaborn is data visualization library based on matplotlib (python). It provides a high-level interface for drawing attractive and informative statistical graphics.
- 📚Learn — Charts and Visualization with Seaborn
- 📚Learn — Seaborn in 1–2 hours
Data Processing — Get familiar with Pandas 🐼
Pandas is the most famous Python library for working with structured data, and machine learning is about data, thus being able to represent, filter, delete, modify data is key.
- 📚Learn — Pandas in 1–2 hours
Reproducibility — Learn how to use Jupyter Notebooks 📙
As a Machine Learning Engineer, most of your work is about experimenting and testing different approaches. Running an entire experiment in a monolithic way often results in hours of debugging and no progress. With Jupyter Notebooks you’ll be able to run code in cells which is a very effective way to experiment and develop in the preliminary stages of a Machine Learning Project ( sounds like a teleshopping ad 😃).
- 📚Learn — Jupyter Notebooks in 15min
Experiment — Apply ML on your own ideas/projects 🚀
Once you’ve got the basics on all the points elaborated above, I think you’re in a good spot to try and apply the algorithms you’ve learned, on an idea of your own. You could limit the scope to the Linear/Logistic regression approaches and do the followings:
- Pick a topic, think of a really basic problem
- Figure out what you’re trying to predict with ML
- Gather some data that describes the problem
- Use Pandas to load, process transform and extract data
- Use Seaborn for Visualization
- Follow the Logistic/Linear Regression tutorials
- Spin-up a new Jupyter Notebook and start experimenting
In the end, I was looking to cover the introductory steps on how to get started with the basics of ML, and how to grasp an idea of what it does. It is worth to keep in mind that this article is aiming to offer a short overview, just enough to get you started with key tools (Pandas, Jupyter, Seaborn) used widely in Machine Learning, so if you didn’t heard of ML before — then it’s a good fit for you.
Of course, you’ll find lots of roadmaps on how to get started, various approaches, more theoretical or more practical, and I encourage you to follow those ones that tailor best to your style.
Thanks for reading!
Feel free to leave your thoughts on this article in the comment section, regardless of the feedback, I’ll appreciate it ;)
If you enjoy technology and also giggle at memes from time to time, stay tuned for more articles. I’m planning to write and explain about AI/ML using games/movies references, memes, jokes and more.
I think that learning and fun should come hand in hand.
Let’s get in touch, you can find me on:
- 💻 GitHub
- 🏄🏻♂️ NeuraLeaps
- 🎁 Buy me a coffee ( comes with a surprise from my part ;) )
