Developed Artificial Neural Networks from first principles

• Learned and implemented the theory of Artificial Neural Networks using Python with NumPy, without the use of any 3rd party machine learning libraries.
• I also allowed for training and saving of multiple models with custom hyper-parameters via a GUI, as well as experimenting via Jupyter notebooks, and achieved a prediction accuracy of 96% on the MNIST dataset.
• Provided a Docker file and set up a GitHub workflow to run unit tests as part of the CI/CD.
• GitHub Repository: github.com/max-cotton/school-project