Yet Another Machine Learning From Scratch (Python)
During my studies, I encounter various algorithms, and it helps me to implement them from scratch to better understand them. These libraries are meant as a learning tool, so code is written verbosely and is largely unoptimized.
In Python, scikit-learn is pretty much the de facto library for machine learning. In an effort to provide a good comparison, I tried to stick to the scikit way of doing things like...
model = MyClassifier(params) model.fit(X_train, y_train) predictions = model.predict(X_test)
How To Run
sudo pip install -r requirements.txt to grab libraries.
examples directory to see algorithms and comparisons.