Scikit-learn: Practical Machine Learning

Learn how to do machine learning with real-world data


  • Ionescu Vlad


Number of students:

40 / lecture group, 20 / lab group

Duration and planning:

6 weeks, 24 hours:

  • 2 hours lecture / week  
  • 2 hours lab / week


Necessary extra effort

4 hours / week

Pricing and payment

The price shown on our learning platform, pay by card or PayPal.

About the tutor

I started learning to code over 20 years ago when I was around 11, to make changes to various applications and games. I know how frustrating it can be to not find answers to your questions anywhere and to not be able to tell good resources from bad ones. This is why I decided to start Evobyte: to deliver quality courses that truly help people.

I have a PhD in Computer Science, two courses published on Udemy (here and here) and many well-received answers on StackOverflow. Most importantly however, I have a passion for programming and teaching that I hope I can also instill in my students.

Course description

This course focuses on quickly getting you able to apply machine learning on real-world data sets using the Scikit-learn Python machine learning library. We will assume that you have at least basic programming skills. You should be able to follow along nicely if you have used any programming language before, even if it wasn’t Python.

The main course objective is for you to gain an intuitive understanding of the most common machine learning models that exist and to be able to properly apply them on real-world data and in real scenarios.

We will use industry-standard data sets taken from places such as Kaggle and by the end of the course you will be able to implement a professional machine learning pipeline that follows all the best practices in the field and performs well.

We do not focus on mathematical details in this course.

Skills gained after this course

Participants who will actively take part in our activities and make a concerted effort to learn will:

  1. Understand the basic concepts of machine learning;
  2. Understand how and when to apply supervised and unsupervised learning;
  3. Know the best practices to follow when developing a production-ready machine learning system;
  4. Know what to study next to expand their knowledge in the areas they are interested in;
  5. Be able to avoid common mistakes when doing machine learning development;
  6. Be able to use real-world data sets for the experiments;
  7. Be able to start applying machine learning for their own needs and on their own data;

Online activities

All activities will take place online, mainly on Google Meet and our Moodle platform, after a timetable agreed upon with all the participants. All video activities (lectures and labs) will be recorded and you can rewatch them at any time.

Other than the above, we will also be using the following:

We recommend that you have a functional microphone and webcam to easily communicate during activities, but it is not mandatory.

Course structure

The course will run over 6 weeks with the following main activities each week:

Outside of these activities you can always contact us on the discussion forums and on Discord.

Necessary tools

All you need beforehand is a PC or a Laptop with a web browser (such as Google Chrome).

We will tell you what else to install during the activities. We will only use free tools.

How to start

Register on our online learning platform and sign up for this course: 

What we expect from you

We expect you to actively participate in the main activities (lecture and lab) and ideally also in the secondary ones, such as quizzes and exercises. It is also important that you take into account any feedback that you receive.

Individual study is also very important, which is why we recommend that you invest at least 4 hours a week for further reading (we will give you curated lists of tutorials and videos for this), doing exercises and quizzes, rewatching our recordings and so on.

There are no completion conditions for this course. Everyone will receive a completion certificate at the end of the 4 weeks.

What you will learn - course contents

Here is a summary of what we will be studying each week.


Summary of activities


Introduction to Scikit-learn: installation, basics, working with data; supervised learning: linear regression.


Logistic regression, real-world data sets, visualizations.


Model evaluation and tuning: train-test splits, cross-validation, learning curves, grid search and randomized search.


Non-linear models: SVMs, MLPs, Decision Trees, Random Forests. Different metrics for different tasks.


Unsupervised learning, clustering, K-Means, other clustering algorithms; comparison with supervised learning.


Working with text data: preprocessing and suitable algorithms.