I recently finished the free online machine learning course by Stanford University professor Andrew Ng on coursera. It gives a grounding of the core concepts of machine learning such as cost functions and gradient descent and explores how different supervised and unsupervised learning algorithms such as linear regression, neural networks, support vector machines and K-means clustering work.
I was able to follow most of the maths thanks to the level 2 Further Maths course we finished before the lockdown although I did get lost in the optional modules explaining support vector machines.
I found the course extremely interesting and engaging and have learnt a lot. I followed the recommended due dates for the weekly quizzes and programming assignments to finish in 11 weeks.
I had to use Octave for the programming assignments, which is a high-level programming language designed for making numerical computations using matrices and vectors and which is mostly cross-compatible with MATLAB. It was very different to Python and JavaScript but I quite enjoyed using it once I got used to the basics. The main thing that I couldn’t get used to was indexing starting at 1 instead of 0 (the horror!).
I’m looking forward to applying my new knowledge on machine learning problems.
Brychan Thomas' projects 