Machine Learning Tutorials (Jupyter Notebooks) — English & فارسی

A bilingual, notebook-first machine learning course delivered in Jupyter Notebook (.ipynb) format. The repository is designed as a structured, chapter-by-chapter curriculum for learners who want a practical and theoretical path through classical machine learning, model evaluation, time series, reliability, MLOps, and applied case studies.

Each lesson is provided in:

• English: Tutorials/English/...
• Persian (Farsi / فارسی): Tutorials/Persian/... with Persian notebooks ending in _Fa.ipynb

The English and Persian notebooks cover the same material, with the Persian version being a translation of the English content.

---

Repository layout

Datasets/
  Classification/   # tabular datasets for classification examples
  Regression/       # tabular datasets for regression examples
  Clustering/       # datasets for unsupervised learning and clustering
Tutorials/
  English/
    Chapter1/
      Chapter1_Lesson01.ipynb
      Chapter1_Lesson02.ipynb
      ...
    ...
    Chapter37/
      Chapter37_Lesson01.ipynb
      Chapter37_Lesson02.ipynb
      ...
  Persian/
    Chapter1/
      Chapter1_Lesson01_Fa.ipynb
      Chapter1_Lesson02_Fa.ipynb
      ...
    ...
    Chapter37/
      Chapter37_Lesson01_Fa.ipynb
      Chapter37_Lesson02_Fa.ipynb
      ...
css/
  rtl.css           # RTL styles used by Persian notebooks

• Datasets/ contains CSV files used in examples, exercises, and projects.
• Tutorials/ contains the notebooks organized by Chapter → Lesson.
• css/rtl.css supports right-to-left rendering for Persian notebooks.

---

Course size

This curriculum contains:

• 37 chapters
• 330 lessons
• Introductory, advanced, and comprehensive extension modules
• End-to-end capstone projects for applied classical machine learning

---

What you will learn

This course spans an end-to-end classical machine learning curriculum, progressing from foundations to advanced topics and applied case studies:

• Machine learning foundations, problem formulation, workflow design, and reproducibility
• Data preprocessing, feature engineering, leakage prevention, and pipeline hygiene
• Exploratory data analysis, diagnostics, statistical testing, and data-quality forensics
• Supervised learning: regression, classification, GLMs, regularization, robust methods, and cost-sensitive learning
• Decision trees, tree variants, ensemble learning, boosting, calibration, and imbalanced learning
• SVMs, kernel methods, instance-based learning, and probabilistic models
• Unsupervised learning, clustering, dimensionality reduction, and association rules
• Cross-validation, model evaluation, uncertainty estimation, and decision analysis
• Time series modeling, forecasting, rolling validation, and anomaly detection
• Bayesian networks, graphical models, Gaussian processes, causal inference, and experimentation
• Fairness, privacy, robustness, security, reliability, conformal prediction, and MLOps
• Scalable, online, specialized, and capstone machine learning workflows

---

Full course curriculum

Introductory Course of Machine Learning

Chapter 1: Introduction to Machine Learning

• Lesson 1: What is Machine Learning?
• Lesson 2: Types of Machine Learning (Supervised, Unsupervised, Semi-Supervised, Reinforcement Learning)
• Lesson 3: Applications and Real-World Use Cases
• Lesson 4: ML Workflow (Data, Model, Evaluation, Deployment)
• Lesson 5: History and Evolution of ML
• Lesson 6: Key Mathematical Foundations (Linear Algebra, Probability, Optimization)
• Lesson 7: Common Misconceptions and Challenges in ML
• Lesson 8: ML vs. Statistics vs. Data Mining vs. AI (Boundaries and Overlaps)
• Lesson 9: Problem Formulation (Inputs/Outputs, Objective, Constraints, Costs)
• Lesson 10: Learning Paradigms and Task Taxonomy (Regression, Classification, Ranking, Forecasting)
• Lesson 11: What “Good Performance” Means (Generalization, Robustness, Reliability)
• Lesson 12: Reproducibility Basics (Random Seeds, Determinism, Experiment Tracking Concepts)

Chapter 2: Basics of Data and Preprocessing

• Lesson 1: Understanding Data Types and Structures
• Lesson 2: Data Cleaning and Missing Values
• Lesson 3: Data Transformation and Encoding
• Lesson 4: Feature Scaling (Normalization & Standardization)
• Lesson 5: Handling Outliers and Imbalanced Data
• Lesson 6: Feature Engineering Fundamentals
• Lesson 7: Data Leakage and Prevention Techniques
• Lesson 8: Data Collection and Label Quality (Noise, Ambiguity, Measurement Error)
• Lesson 9: Missingness Mechanisms (MCAR, MAR, MNAR) and Practical Implications
• Lesson 10: Imputation Methods (Simple, kNN, Iterative/Multiple Imputation Concepts)
• Lesson 11: Encoding Categorical Features (One-Hot, Ordinal, Hashing, Target Encoding)
• Lesson 12: Train/Validation/Test Hygiene (Temporal Splits, Group Splits, Entity Leakage)
• Lesson 13: Building Preprocessing Pipelines (Fit/Transform Discipline, Column-Wise Pipelines)

Chapter 3: Exploratory Data Analysis (EDA)

• Lesson 1: Visualizing Data Distributions
• Lesson 2: Pairwise Relationships (Correlation, Scatterplots)
• Lesson 3: Detecting Patterns in Data
• Lesson 4: Dimensionality Reduction (Intro to PCA)
• Lesson 5: Using Tools like Pandas, Matplotlib, and Seaborn
• Lesson 6: Statistical Hypothesis Testing in ML Context
• Lesson 7: EDA for Data Quality (Duplicates, Inconsistencies, Drift, Label Issues)
• Lesson 8: Multicollinearity and Confounding Signals (Detection and Mitigation)
• Lesson 9: Leakage Forensics in EDA (Suspicious Features, Post-Outcome Variables)
• Lesson 10: EDA Reporting (Narratives, Assumptions, and Actionable Insights)

Chapter 4: Supervised Learning Basics

• Lesson 1: Introduction to Regression and Classification
• Lesson 2: Linear Regression: Concept and Applications
• Lesson 3: Logistic Regression: Binary Classification
• Lesson 4: Overfitting and Regularization (Ridge, LASSO, Elastic Net)
• Lesson 5: All Model Evaluation Metrics (MAE, MSE, RMSE, Accuracy, etc.)
• Lesson 6: Bias-Variance Tradeoff and Model Complexity
• Lesson 7: Polynomial and Interaction Terms in Regression Models
• Lesson 8: Generalized Linear Models (GLMs) Overview (Link Functions, Likelihood)
• Lesson 9: Multiclass Logistic Regression (Softmax) and Evaluation
• Lesson 10: Ordinal Regression (When Class Order Matters)
• Lesson 11: Robust Regression (Huber, RANSAC Concepts and Use Cases)
• Lesson 12: Quantile Regression and Prediction Intervals (Intro)
• Lesson 13: Cost-Sensitive Learning Basics (Thresholding, Costs, Utility)

Chapter 5: Decision Trees and Variants

• Lesson 1: Concept of Decision Trees
• Lesson 2: CART (Classification and Regression Trees)
• Lesson 3: Pruning and Overfitting in Trees
• Lesson 4: CHAID and M5 Model Trees
• Lesson 5: C4.5 and C5.0 Decision Trees
• Lesson 6: Interpretability and Feature Importance in Trees
• Lesson 7: Split Criteria (Gini, Entropy, Gain Ratio, Variance Reduction)
• Lesson 8: Handling Missing Values and Categorical Variables in Trees
• Lesson 9: Tree Stability, Variance, and Sensitivity Analysis
• Lesson 10: Monotonic Constraints and Business Rules in Tree Models (Concepts)

Chapter 6: Ensemble Learning Techniques

• Lesson 1: What is Ensemble Learning?
• Lesson 2: Bagging Algorithms (Random Forest, Bootstrap Aggregation)
• Lesson 3: Boosting Algorithms (AdaBoost, Gradient Boosting)
• Lesson 4: Stacking and Blending Techniques
• Lesson 5: Comparison of Ensemble Methods
• Lesson 6: Voting Classifiers and Averaging Methods
• Lesson 7: Bias-Variance Reduction via Ensembles
• Lesson 8: Out-of-Bag (OOB) Estimation and When It Works
• Lesson 9: Extremely Randomized Trees (ExtraTrees) and Diversity
• Lesson 10: Calibration with Ensembles (Platt Scaling, Isotonic Regression Overview)
• Lesson 11: Imbalanced Data with Ensembles (Class Weights, Balanced RF, Thresholding)

Chapter 7: Support Vector Machines (SVM)

• Lesson 1: Concept of SVM for Classification
• Lesson 2: Kernel Functions in SVM
• Lesson 3: Soft Margin and Hyperparameters
• Lesson 4: Support Vector Regression (SVR)
• Lesson 5: Applications of SVM in Real-World Problems
• Lesson 6: Mathematical Formulation of the SVM Optimization Problem
• Lesson 7: Kernel Trick Intuition and Feature Spaces
• Lesson 8: Multi-Class SVM Strategies (OvR, OvO) and Practical Tradeoffs
• Lesson 9: SVM Probability Estimates and Calibration Considerations
• Lesson 10: Scaling SVMs (Approximate Kernels, Linear SVMs, Complexity)

Chapter 8: Instance-Based Learning

• Lesson 1: K-Nearest Neighbors (KNN) Algorithm
• Lesson 2: Choosing the Right K
• Lesson 3: Distance Metrics and Weighting
• Lesson 4: Locally Weighted Learning (LWL)
• Lesson 5: Applications and Challenges
• Lesson 6: Curse of Dimensionality and Its Impact on KNN
• Lesson 7: kNN Regression and Local Smoothing Bias/Variance
• Lesson 8: Approximate Nearest Neighbors (KD-Trees, Ball Trees, ANN Concepts)
• Lesson 9: Metric Learning Overview (When Distances Should Be Learned)

Chapter 9: Probabilistic Models

• Lesson 1: Naïve Bayes Classifier
• Lesson 2: Gaussian Naïve Bayes
• Lesson 3: Bayesian Linear Regression
• Lesson 4: Assumptions and Limitations
• Lesson 5: Case Studies with Probabilistic Models
• Lesson 6: Maximum Likelihood vs. Bayesian Estimation
• Lesson 7: Bayesian Decision Theory (Risk, Loss, Bayes Optimal Classifier)
• Lesson 8: Priors and Conjugacy (Conceptual Toolkit for Fast Bayesian Updates)
• Lesson 9: MAP Estimation, Regularization Connections, and Interpretations
• Lesson 10: Expectation-Maximization (EM) Intuition (Preview for Later Chapters)

Chapter 10: Unsupervised Learning Basics

• Lesson 1: Introduction to Clustering
• Lesson 2: K-Means Clustering
• Lesson 3: Hierarchical Clustering
• Lesson 4: Gaussian Mixture Models (GMM)
• Lesson 5: Applications of Clustering Techniques
• Lesson 6: Evaluation Metrics for Clustering (Silhouette, Davies-Bouldin)
• Lesson 7: K-Means Variants (k-medoids, k-means++ initialization)
• Lesson 8: Model-Based Clustering and Selecting Number of Clusters (AIC/BIC concepts)
• Lesson 9: Constraints and Practicalities (Must-Link/Cannot-Link, Business Constraints)
• Lesson 10: Clustering at Scale (Mini-Batch K-Means and Sampling Strategies)

Chapter 11: Dimensionality Reduction

• Lesson 1: Principal Component Analysis (PCA)
• Lesson 2: t-SNE for Visualization
• Lesson 3: Linear Discriminant Analysis (LDA)
• Lesson 4: Feature Selection vs Feature Extraction
• Lesson 5: Applications of Dimensionality Reduction
• Lesson 6: Independent Component Analysis (ICA)
• Lesson 7: Random Projections and Johnson–Lindenstrauss Intuition
• Lesson 8: Non-negative Matrix Factorization (NMF) for Parts-Based Representations
• Lesson 9: Sparse PCA and Interpretability Tradeoffs
• Lesson 10: Embedded Feature Selection (L1, Tree-Based, Permutation Importance Overview)

Chapter 12: Association Rule Learning

• Lesson 1: Concept of Association Rules
• Lesson 2: Apriori Algorithm
• Lesson 3: Eclat Algorithm
• Lesson 4: Market Basket Analysis
• Lesson 5: Challenges and Limitations
• Lesson 6: Evaluation Metrics (Support, Confidence, Lift)
• Lesson 7: FP-Growth (Frequent Pattern Growth) and When It Outperforms Apriori
• Lesson 8: Rule Pruning and Redundancy Control
• Lesson 9: Sequential Pattern Mining (Concepts and Use Cases)

Chapter 13: Cross-Validation and Model Evaluation

• Lesson 1: Train-Test Split and Validation
• Lesson 2: K-Fold Cross-Validation
• Lesson 3: Stratified Sampling in Cross-Validation
• Lesson 4: Performance Metrics for Classification
• Lesson 5: Performance Metrics for Regression
• Lesson 6: Model Calibration and ROC Curves
• Lesson 7: Learning Curves and Validation Curves
• Lesson 8: Nested Cross-Validation for Model Selection (Leakage Avoidance)
• Lesson 9: Bootstrap Methods for Performance Estimation and Uncertainty
• Lesson 10: Statistical Tests for Comparing Models (Paired Tests, Practi