This repository contains from-scratch implementation of common machine learning algorithms such as Linear Regression, Logistic Regression, K-nearest neighbors, Neural Networks and Backpropagation, etc...
The Matrix class is very incomplete and requires immediate attention; calculating matrix inverse and RREF in Python is tricky but I'm working on it. As of now both gradient descents are functional, logistic regression requires more testing and k-means clustering should work. K-nearest neighbors is 100% correct.
However Python is comparatively slow so some r values combined with certain arrays may take forever to loop (gradient descent) Should have just used Tensorflow....oh well
Note that mimir.py hasn't been packaged as a python module/library properly yet.
Feel free to use and experiment however you like.