MLP: Machine Learning Practical

Machine Learning Practical — 2026/27

This page is an asset repository for people not registered for Machine Learning Practical who wish to audit the course.

Materials are added periodically. This repository is not maintained in step with course activities and should not be relied upon for current teaching schedules, coursework requirements or deadlines.

Registered students should use the current year’s Learn course for course materials and information. They should not use this repository as part of taking the course.

Course resources

• Course GitHub repository — code and practical notebooks
• Full course information — DRPS 2026/27

About the course

Machine Learning Practical (MLP) focuses on the design, implementation, training and evaluation of machine learning systems, with particular emphasis on practical and experimental aspects of deep learning.

Semester 1 combines lectures, labs and individual coursework. Students develop a deep learning framework using Python and Jupyter notebooks, experimenting with handwritten-digit classification using MNIST.

Semester 2 centres on small group projects using established frameworks such as PyTorch or TensorFlow, supported by tutorials and guest lectures.

MLP is not a standalone introduction to machine learning. Please consult the DRPS entry for entry requirements, learning outcomes and full course information.

Topics

• Feed-forward neural networks
• Optimisation and learning rules
• Regularisation and normalisation
• Neural networks for classification
• Autoencoders
• Convolutional neural networks
• Recurrent neural networks
• Transformers

Course details

Weekly reading guides

The weak numbers organise the materials by their place in the course. They do not indicate when materials will be published in this repository. Click on the links below for the reading guide and slides for that week's topic.