Machine Learning Models from Scratch

This ongoing repository is my implementation of machine learning models covered in Stanford's CS229 with Python and NumPy.

About

In this repo, I have included most of the models covered in CS229, all entirely built from scratch. Each model came with detailed explanations and was tested with a classic dataset from Kaggle.

Those 8 models include (sorted in category):

• Supervised Learning:

  • Discriminative Algorithms:
    • Linear Regression - House Price Prediction
    • Logistic Regression - Brain Tumor Classification
    • Decision Trees - Titanic Survival Classification
    • Supported Vector Machines (SVM) - Iris Flower Classification (still in progress)
  • Generative Algorithms:
    • Naive Bayes - Email Spam Classification
  • Deep Learning:
    • Neural Network - Handwritten Digit Recognition (MNIST) (still in progress)

• Unsupervised Learning:

  • Clustering
    • K-means- Customer Segmentation (still in progress)
  • Dimensionality Reduction
    • Principal Component Analysis (PCA) - Face Recognition (still in progress)

You could also find my handwritten notes here. Thank you!