This is an implementation of a Support Vector Machine. It was part of a school assignment in my machine learning course. The algorithm is built from scratch, using Python packages NumPy and SciPy, and following assignment instructions in “instructions.pdf”.

It can be used to classify data points with linear and non-linear functions, as seen in Figure 1.

Test the Support Vector Machine by navigating to the repository in the terminal and type:

python experiments.py

The experiments run with Linear Kernel as default. Change the kernel function by following instructions in "experiments.py".