Skip to main content

4: Classification II & Workshops

Date: 30th of October 2024

💡 Classification helps us to group data into predefined classes. We will recap classification techniques from last week, including Logistic Regression, K-means, and Support Vector Machines (SVM). We will cover a new technique called Decision Trees which uses a tree-like structure to make decisions based on feature attributes. We will introduce 2 practical machine learning challenges:

  1. 🐧 Penguin species classification
  2. 🏠 California housing price prediction

Over the next 3 weeks until 13 Nov, 2024 18:00 we will be running the housing price prediction DOXA challenge

People who manage to get MAE of under 32.6 will be eligible for AI society shirt and a pen (up to 2 or 3 shirts available, so first come first serve).

Check out the last part of our presentation discussing methods to get a better MAE (mean absolute error). We will teach you about ensemble methods including gradient boosting and feature scaling

You can access our demonstration notebook here: 📘 Tutorial 4 Notebook

The folder contains notebooks for Decision Trees as well as the DOXA Challenges.

You can access our slides here: 💻 Tutorial 4 Slides

The recording from this session is available here: 🎤 Tutorial 4 Recording