day-2.md

Day 2

10. Oct. 9:00 - 17:00

Goals

• Provide an overview of machine learning for classification (and regression)
• Explain the most common concepts and pitfalls
• Give some inspiration and show what can be done with machine learning

Preparation

If possible, have the following python libraries ready:

• numpy
• scikit-learn
• matplotlib

If we don't run out of time, we will also do a quick exercise with pytorch:

conda install pytorch torchvision -c soumith

Program

Tentative program for day 2:

9:00-10:30	Introduction to Machine Learning
10:30-11:00	30 minute break
11:00-12:30	Classical Approaches to Machine Learning
12:30-13:30	Lunch break
13:30-15:00	Machine Learning and Images
15:00-15:30	30 minute break
15:30-17:00	Modern Approaches to Machine Learning

Material

Slides

• Part 1
• Part 2
• Part 3
• Part 4

Thanks to Yuliya Tarabalka for providing some of the material.

Exercises

• PCA
• Classification using k-nearest-neighbors
• Classification using PCA and SVM
• Convnets with PyTorch