/ml-from-scratch

Implementations of basic machine learning algorithms using only numpy, pandas, and matplotlib.

Primary LanguageJupyter Notebook

Just some playing around with the most bare-bones implementations of core ML algorithms I can manage. TODO: remove numpy in favor of util fncs, only use pandas to load, do evaluation separation on all, cross-validation fnc, SVM, Random Forests, n-layer MLP, CNN, RNN.