The plan for this project is to make a tool that can load weights, train, a model, and evaluate a test dataset to produce an output file, for a fully connected n-layer neural network. This proccess is horribly inefficient, because the proffesor required us to avoid all matrix algebra so "we understand better."
I then plan to add an exporting feature to json, which can be used with a Python Script will generate a graphic which shows how the weights train overtime.