GitHub - eriklindernoren/ML-From-Scratch: Machine Learning From Scratch. Bare bones NumPy implementations of machine learning models and algorithms with a focus on accessibility. Aims to cover everything from linear regression to deep learning.

Abstraction: Bare-bones NumPy educational implementations of ML algorithms

Key points:

• GitHub repo by Erik Lindernoren: transparent NumPy implementations of ML models prioritizing readability over performance
• Supervised learning: Linear/Polynomial/Ridge/Lasso/Bayesian/Logistic Regression, Decision Tree, Random Forest, SVM, KNN, Naive Bayes, Adaboost, XGBoost, MLP, Neuroevolution
• Unsupervised: K-Means, DBSCAN, GAN, Autoencoder, Gaussian Mixture Model, RBM, PCA, FP-Growth, Apriori
• Deep learning module: CNN (ConvNet achieves 98.7% accuracy on digit dataset), GAN (handwritten digits), Deep Q-Network (CartPole), RNN layers
• Neuroevolution example: 100-individual population, 3000 generations, achieves 98.5% train / 96.7% test accuracy on digit dataset

Connections: Github · Machine Learning · Deep Learning · Supervised Learning

Source: https://github.com/eriklindernoren/ML-From-Scratch