This repository documents my journey of turning Stanford CS229 math notes into working machine learning projects using Python (Google Colab).

Instead of stopping at equations, each notebook:

• Starts with the math derivation
• Translates it into step-by-step code
• Builds visualizations to connect theory with practice

The goal: strengthen ML foundations by seeing how math directly powers real models.

1. Derived the Mean Squared Error (MSE) cost function
2. Implemented the gradient descent update rule from scratch
3. Plotted Cost vs Iterations to show convergence
4. Built a simple predictor for housing prices

1. Implemented sigmoid function
2. Derived and coded cross-entropy loss + gradient descent
3. Visualized loss curve and decision boundary
4. Compared with scikit-learn’s LogisticRegression

Each project follows the same structure:

1. Start with Math – review CS229 derivation
2. Code Implementation – translate step by step
3. Visualization – graphs/plots to show error minimization
4. ML Intuition – how this improves predictions

• ✅ Linear Regression
• ✅ Logistic Regression
• Regularization (L1 / L2)
• Softmax Regression
• Neural Networks (Backprop basics)
• Support Vector Machines
• PCA & Dimensionality Reduction

Ibrahim Arshad – exploring ML theory & practice 🚀