Course: Machine Learning for Image Data

What is dimensionality reduction and why is it useful? This video gives a brief overview of what dimensionality reduction is, before mentioning some techniques to perform it, such as PCA …

As we saw in the video, linear regression is the machine learning equivalent of drawing a straight line through your data. In this article we explain the process in more …

In the previous video we saw some of the model evaluation metrics and measures used to evaluate classification models. In this article we will give a quick reminder of those …

Congratulations on completing Week 3 of the course. We’ve covered a lot this week, including an in depth look at some specific techniques which can be broadly categorised as: clustering …

What do we mean by model selection? Once you’ve picked which family of models to use, you may still need to select hyperparameters to define a specific model within that …

This week we looked at several common machine learning methods, and techniques to evaluate our models. In this practical we will build and train an image classifier using the flower …

As we’ve seen, good data visualisation is an important tool in your analysis, and for communicating your results. This article gives a few practical tips for data and model visualisation …

Good visualisation of your data and model results can help both your analysis and communication of your results. This video gives a quick overview of some common techniques used in …

How can we measure model performance? This video looks at some common measures or metrics used in evaluating the performance of machine learning classification models, including: accuracy confusion matrices precision …

All deep learning systems are made up of layers of much simpler ‘building blocks’. In this article we will briefly go through some of the component parts. Perceptron A perceptron …

Linear regression is possibly the most widely known machine learning technique. Most of us are familiar with the idea of fitting a straight line though plotted data. In fact, this …

Decision trees are a commonly used method for classification problems, though they may also be used for regression. They are called trees as they organise sets of smaller decisions in …

As we saw in the previous video, Naive Bayes is a simple supervised learning classification method. In this article, we will review some of the details of the algorithm, before …

What is Naive Bayes classification? Naive Bayes classification is a supervised machine learning algorithm that works on labelled data sets. This video explains how it works in more detail.

Clustering is an unsupervised machine learning technique that divides unlabelled data into distinct clusters occupying similar regions of features space. This video looks at K-means clustering in depth, and introduces …