Project10X: ML Models

Project10X-ML-Models series: a structured, model-centric ML mastery journey based on Chapter 4 of Hands-On Machine Learning by Aurélien Géron.

Each model is explored through a 10-project progression, starting from intuition and basic math, building up to experimentation, visualization, real-world application, and best practices like regularization and early stopping.

---

Structure

This repository is divided into 6 major model categories, each with its own dedicated subfolder and 10 progressive projects.

Each category teaches to not just train models — but deeply understand how and why they work.

---

Core Tracks

Track #	Model Family	Folder	Key Topics
1	Linear Regression	linear_regression/	Normal Equation, Predictions, Fitting
2	Gradient Descent	gradient_descent/	Batch, SGD, Mini-Batch GD
3	Polynomial Regression	polynomial_regression/	Feature Expansion, Nonlinear Fit
4	Learning Curves	learning_curves/	Bias-Variance, Sample Size Effects
5	Regularized Linear Models	regularized_models/	Ridge, Lasso, ElasticNet, Early Stopping
6	Logistic Regression	logistic_regression/	Classification, Decision Boundaries, Softmax

---

Format

Each subfolder (e.g. linear_regression/) includes:

• projectXX_name/ folders with:
  • README.md for project overview and learning goals
  • notebook.ipynb for implementation
  • Visualizations and experiments
  • datasets/ folder if datasets are included or generated

Each 10x project builds on the last, increasing in complexity and insight.

---

Philosophy

Don’t just train the model. Master the model.

This repo is designed to help me: - Build practical intuition through implementation and visualization - Understand where models break down (not just how to use them) - Run real diagnostics and interpret results meaningfully - Develop ML maturity from experimentation to deployment-readiness

---

Source Inspiration

• Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow – Chapter 4
• Official scikit-learn docs
• Real ML experiments and diagnostics

---

Prerequisites

scikit-learn
matplotlib
numpy
pandas
jupyternotebook

---

License

This repository is open source under the MIT License.

---

Created and maintained by RM Villa.