Machine learning algorithm (or part of them) implemented in python during the module Pattern Recognition, Neural Networks and Deep Learning at King's College, London.
These implementations and were ment to solve toy problems for the module's assignments. The idea was to write the code in an understandable way more than focusing on the optimisation. I used them as a tool to understand how each algorithm worked.
For the same reason, I choose to use jupyter notebooks instead of plain python, in order to comment each step and to have major control on my experiments.
Some notebooks only contain useful functions like the check for linear separability, or how tos, like PCA and LDA.