Python-first · Theory before code · R edition to follow

Machine Learning A to Z

Every algorithm taught the way a professional needs it: what it is, the intuition, the mathematics, the assumptions it rests on, a worked example by hand, the Python implementation — and how to honestly judge the result.

Start Reading → Browse All 33 Chapters
10
Parts
33
Chapters
29
Algorithms implemented
1,000
Quiz questions

The Course in Ten Parts

The journey runs from foundations, through the three families of learning, to deep learning and a professional capstone. Each part builds on the ones before it.

🧱Part I

Foundations & Data Preprocessing

ML literacy, the statistics primer, and the preprocessing toolbox — missing data, encoding, scaling, splitting.

Chapters 1–3
📈Part II

Regression

Six regressors from simple OLS to random forests — with the full assumption suite and a dedicated evaluation chapter.

Chapters 4–10
🎯Part III

Classification

Seven classifiers from logistic regression to random forests — plus precision, recall, F1, ROC-AUC and the imbalance story.

Chapters 11–18
🫧Part IV

Clustering

K-Means and hierarchical clustering: customer segmentation with the elbow method and silhouette analysis.

Chapters 19–20
🛒Part V

Association Rule Learning

Apriori and Eclat: support, confidence and lift derived numerically from real market-basket data.

Chapters 21–22
🎰Part VI

Reinforcement Learning

The multi-armed bandit problem: exploration vs exploitation, Upper Confidence Bound and Thompson Sampling.

Chapters 23–24
💬Part VII

Natural Language Processing

Text cleaning, bag of words, TF-IDF and sentiment classification on restaurant reviews.

Chapter 25
🧠Part VIII

Deep Learning

Neural networks from the single neuron to backpropagation, then CNNs — convolution, pooling and image classification.

Chapters 26–27
📉Part IX

Dimensionality Reduction

PCA, LDA and Kernel PCA: the eigen-mathematics, how many components to keep, and when each method wins.

Chapters 28–30
🏆Part X

Model Selection & Boosting

k-fold cross-validation, grid search, XGBoost — and a capstone that welds the whole course into one pipeline.

Chapters 31–33

All Chapters

Chapters are published in production order. Each ends with a hands-on project and a 30-question self-test.

# Chapter Focus Status Quiz
Part I — Foundations & Data Preprocessing
01 Machine Learning Foundations ML vs statistics vs AI · taxonomy · workflow · toolchain Published 10 MCQs
02 Statistics Primer for Machine Learning Distributions · correlation · hypothesis testing Published 30 MCQs
03 Data Preprocessing in Python Missing data · encoding · scaling · leakage Published 30 MCQs
Part II — Regression
04 Simple Linear Regression OLS derivation · full assumption suite · R² Published 30 MCQs
05 Multiple Linear Regression Dummy trap · multicollinearity & VIF · backward elimination Published 30 MCQs
06 Polynomial Regression & the Bias-Variance Tradeoff Feature expansion · overfitting made visual Published 30 MCQs
07 Support Vector Regression (SVR) ε-insensitive tube · kernels in regression Published 30 MCQs
08 Decision Tree Regression CART splits · variance reduction · depth as complexity Published 30 MCQs
09 Random Forest Regression Bagging · why averaging trees works Published 30 MCQs
10 Evaluating & Comparing Regression Models MAE/MSE/RMSE · residual audits · model benchmarking Published 30 MCQs
Part III — Classification
11 Classification Foundations & Logistic Regression Sigmoid · odds ratios · assumptions · thresholds Published 30 MCQs
12 K-Nearest Neighbors (K-NN) Distance metrics · choosing k · curse of dimensionality Published 30 MCQs
13 Support Vector Machine (SVM) Maximum margin · support vectors · the C parameter Published 30 MCQs
14 Kernel SVM The kernel trick · RBF & gamma · kernel choice Published 30 MCQs
15 Naive Bayes Bayes' theorem from scratch · conditional independence Published 30 MCQs
16 Decision Tree Classification Entropy · information gain · Gini · pruning Published 30 MCQs
17 Random Forest Classification & Ensemble Learning Bagging vs boosting · feature subsampling Published 30 MCQs
18 Evaluating Classifiers: Beyond Accuracy Precision/recall/F1 · ROC-AUC · imbalance · benchmarking Published 30 MCQs
Part IV — Clustering
19 K-Means Clustering Elbow method · silhouette · customer segmentation Published 30 MCQs
20 Hierarchical Clustering Dendrograms · linkage criteria · vs K-Means Published 30 MCQs
Part V — Association Rule Learning
21 Apriori Support · confidence · lift · market-basket analysis Published 30 MCQs
22 Eclat Frequent itemsets · vertical layout · vs Apriori Published 30 MCQs
Part VI — Reinforcement Learning
23 RL Foundations & the Multi-Armed Bandit (UCB) Exploration vs exploitation · confidence bounds · regret Published 30 MCQs
24 Thompson Sampling Beta distributions · Bayesian exploration Published 30 MCQs
Part VII — Natural Language Processing
25 Text Analytics: Bag of Words and Beyond Cleaning · stemming vs lemmatization · TF-IDF · sentiment Published 30 MCQs
Part VIII — Deep Learning
26 Artificial Neural Networks Neuron model · activations · gradient descent · backprop Published 30 MCQs
27 Convolutional Neural Networks (CNN) Convolution · pooling · augmentation · image classification Published 30 MCQs
Part IX — Dimensionality Reduction
28 Principal Component Analysis (PCA) Eigenvectors · explained variance · component choice Published 30 MCQs
29 Linear Discriminant Analysis (LDA) Fisher's criterion · supervised reduction · vs PCA Published 30 MCQs
30 Kernel PCA Non-linear structure · RBF projection · method choice Published 30 MCQs
Part X — Model Selection & Boosting
31 Model Selection: k-Fold Cross-Validation & Grid Search Fold mechanics · hyperparameters · honest tuning Published 30 MCQs
32 XGBoost & Gradient Boosting Boosting mechanics · regularisation · parameter guide Published 30 MCQs
33 Capstone: End-to-End ML Project Pipelines · model shortlist · business metrics · deployment Published 30 MCQs

Applied Projects — ML in Equity Research

The course, meet the market: full project lessons that weld the chapters into professional equity-research workflows — real market data, finance concepts taught alongside the code, and honest scoreboards.

# Project Focus Status Quiz
MP1 Headline Sentiment & the Next Day's Move (Chapter 25 companion) Bag of words on finance headlines · does sentiment predict returns? Published 30 MCQs
MP2 Will the Loan Default? (Chapter 26 companion) Keras ANN for credit risk · vs logistic + engineered features · profit threshold Published 30 MCQs
P1 Decision Tree: Stock Momentum Technical indicators · interpretable trading rules · honest limits Published 30 MCQs
P2 Random Forest: Earnings & Price Prediction Ensemble advantage · bagging vs one tree · accuracy vs interpretability Planned —
P3 SVR: Stock Price Regression Support vectors on financial series · scaling · kernel choice Planned —

How to Use This Book

The book is designed for self-study: read in order the first time, use as a reference forever after.

1

Read the theory first

Every chapter opens with definition, intuition and mathematics before any code appears — don't skip ahead to the listings.

2

Run the code yourself

Type or paste each listing into your own notebook and execute it block by block. Reading code is not learning code.

3

Do the hands-on project

Attempt it before opening the walkthrough. The struggle is where the learning happens.

4

Pass the quiz

Score 8/10 or better before moving on; retake after re-reading the sections you missed.

📝
Note: Content sections carry stability badges — Core (evergreen theory), Stable (reviewed twice a year) and Volatile (reviewed quarterly) — so you always know how much to trust library-specific details.