Tutorial Challenge: Decision Trees and Forests

We challenge you to do your first Machine Learning coding

Hello fellow KTHAISers!

We believe that many of you are keen to get hands-on experience working with machine learning methods. That is why we want to challenge you who are ML beginners to get your hands dirty with this kaggle tutorial, where you'll be learning some practical skills about machine learning models for predicting house prices.

We encourage you to take on this challenge and share your results by submitting your notebook or a github-repo link to submissions@kthais.com. If you get stuck or need help write in the slack!

Category: Beginner

Type: Skill-building / Tutorial

Learning Goals

  • Understand what an ML model is and why it works (Regression task)
    • Decision Trees
    • Random Forests
  • Work with it practically in a Python Notebook environment
  • Importing and exploring a dataset
  • Learn about model validation and over/underfitting


  • Basic maths and a little python programming

Time investment

You should be able to finish this challenge within 2 hours (or less if you’re quick!).

The challenge

LINK: Kaggle Getting Started: Your First Machine Learning Model

This is a basic tutorial in 7 small parts - here’s the outline:

  1. How Models Work, Decision Trees
  2. Basic Data Exploration
  3. Your First Machine Learning Model
  4. Model Validation
  5. Underfitting and Overfitting
  6. Random Forests
  7. Kaggle Competitions submission

Each part has an interactive exercise where you can work in a Notebook environment in your browser - no need need to install anything! You can also keep track of your progress.

This is the first KTHAIS challenge, good for beginners who are just starting out learning practically about how to build and train models. Stay tuned as more cases including other machine learning methods are being published!