The latest addition in my quest to implement common machine learning algorithms in Python is a simple implementation of the k-means unsupervised learning algorithm. It is a straightforward implementation using numpy arrays that we test on some data generated with scikit-learn. This algorithm is a popular and simple way to detect clusters within your data set, but it is not without it’s limitations. I’ll explore that in another post to come!
As usual, you can find the jupyter notebook in my ml-python repo if you’d like to follow along, or rendered in HTML here: