🇬🇧 If you want to learn Machine Learning, take this Scikit-learn course!


Teaching Machine Learning

Teaching AI and Machine Learning is far from easy. You can find so many resources on this subject: youtube videos, blog posts, MOOCs… As you can imagine, the quality is very heterogeneous.

I personally know this as I give AI courses for different companies. According to the background of your students, you need to adapt the way your approach such abstract concept as overfitting or cost function.

A few months ago, I saw that the team behind the amazing machine learning library Scikit-Learn released a full course about Machine Learning. I personally teach (and use) scikit-learn. This is the de-facto library to start with Machine Learning.

I bookmarked the link for later. And “later” was this week as I cleared my agenda for July.

Well, I sincerely invite you to take a look at this course if you want to dive in this subject. It is very well done. The approach is very educational: clear explanations, a tons of examples, not too many ideas at the same time, a lot of quizzes… It is really great ! In fact, I’m going to take some inspiration from it for my next courses.

Scikit-learn course

Alright, I’ll describe you some of the key elements of this course.

This course is divided into 10 modules:

  • Machine Learning Concepts
  • The predictive modeling pipeline
  • Selecting the best model
  • Hyperparameter tuning
  • Linear models
  • Decision tree models
  • Ensemble of models
  • Evaluating model performance
  • Feature selection
  • Interpretation

In every module, first you have a clear description about the key concepts. Then, you are guided through some examples, using scikit-learn. Then, you have an exercise you need to do. The solution is available (of course). And several quizzes punctuate the course.

If you need a complete introduction in Machine Learning you’ll learn everything you need thanks to this course :

  • data exploration
  • handle numerical and categorical data
  • overfitting and underfitting
  • hyperparameter tuning
  • linear and tree-based models
  • regularization
  • ensemble methods
  • cross-validation
  • feature selection …

The only prerequisite is to know a bit about python. Then you’ll discover numpy, pandas and scikit-learn.


This Scikit-learn course is a must-do if you want to discover Machine Learning.

If you should take one course online, take this one! You won’t be disappointed.

This is very well written, crystal clear explanations, real examples, a lot of graphs.

If you have others resources to share, please, feel free to send them to me. I’ll be glad to hear about them.

Have a nice day

Maxime 🙃

#dataScientist #techplorator #prototypeur #entrepreneur