MLSS 2015 NICTA Labs

A collection of labs for the 2015 Machine Learning Summer School from NICTA authored by:

• Finn Lattimore
• Lachlan McCalman
• Simon O'Callaghan
• Alistair Reid
• Daniel Steinberg
• Brian Thorne
• John Vial

The labs are all self contained in ipython notebooks which are python environments that are locally hosted within a browser. Both the instructions and the code input cells are in the notebook, and so all you need to do to complete a lab is to open the corresponding notebook in the directory where you have downloaded the tutorials, and then work through the exercises in your browser.

This repository contains the first four labs for MLSS 2015 Syndey:

1. Introduction to Python and PCA
2. Linear Regression
3. Classification
4. Clustering and Latent Variable Models

You can find the general lab instructions here, as well as how to set up your python environment.

Preview and Solutions

You can preview all of the lab solutions here:

1. Introduction to Python and PCA
2. Linear Regression
3. Classification
4. Clustering and Latent Variable Models

Dependencies

General:

• scikit-learn
• numpy
• matplotlib
• scipy
• ipython[all]

The above can all be checked using the Dependency checker.ipynb script in this repo. Optionally for the clustering lab:

• lda (may require you to also pull down pbr)
• Pillow