Machine Learning

What you will learn

• Prepare features and split data honestly
• Train regression and classification models with scikit-learn
• Evaluate, tune and compare models
• Apply the same workflow to text

Syllabus

89 lessons in 12 sections. Take them in order, or open any lesson directly.

1. Machine Learning5 min, 3 exercises
2. Phase 1: The ML Foundation8 lessons

   1. Phase 1 - The ML Foundation6 min
   2. What is Machine Learning?9 min, 6 exercises
   3. ML vs Traditional Programming9 min, 6 exercises
   4. The Machine Learning Roadmap8 min, 6 exercises
   5. Artificial Intelligence vs Machine Learning vs Deep Learning8 min, 6 exercises
   6. Types of Machine Learning (Supervised, Unsupervised, Reinforcement)8 min, 6 exercises
   7. The ML Lifecycle - From Data to Deployment9 min, 6 exercises
   8. Setting up the ML Environment (Scikit-Learn, TensorFlow, PyTorch)7 min, 6 exercises
3. Phase 2: Data Preprocessing & Feature Engineering9 lessons

   1. Phase 2 - Data Preprocessing & Feature Engineering4 min
   2. Framing an ML Problem & Getting the Data7 min, 6 exercises
   3. Creating a Test Set (Avoiding Data Snooping)6 min, 6 exercises
   4. Exploratory Data Analysis & Correlations7 min, 6 exercises
   5. Data Cleaning & Handling Missing Values8 min, 6 exercises
   6. Handling Text & Categorical Attributes7 min, 6 exercises
   7. Feature Scaling (Normalization & Standardization)7 min, 6 exercises
   8. Transformation Pipelines & Custom Transformers8 min, 6 exercises
   9. End-to-End Machine Learning Project (California Housing)8 min, 6 exercises
4. Phase 3: Supervised Learning Regression9 lessons

   1. Phase 3 - Supervised Learning - Regression4 min
   2. Introduction to Regression Analysis7 min, 6 exercises
   3. Simple Linear Regression11 min, 6 exercises
   4. Multiple Linear Regression9 min, 6 exercises
   5. Polynomial Regression9 min, 6 exercises
   6. Cost Functions - Mean Squared Error (MSE)9 min, 6 exercises
   7. Gradient Descent Explained9 min, 6 exercises
   8. Regularization - Ridge and Lasso Regression9 min, 6 exercises
   9. Metrics - R-Squared and Adjusted R-Squared8 min, 6 exercises
5. Phase 4: Supervised Learning Classification10 lessons

   1. Phase 4 - Supervised Learning - Classification5 min
   2. Introduction to Classification7 min, 6 exercises
   3. Logistic Regression (Binary vs Multiclass)8 min, 6 exercises
   4. K-Nearest Neighbors (KNN)9 min, 6 exercises
   5. Support Vector Machines (SVM)10 min, 6 exercises
   6. Decision Trees - Entropy and Gini Impurity8 min, 6 exercises
   7. Naïve Bayes Classifier11 min, 6 exercises
   8. Evaluation Metrics - Confusion Matrix10 min, 6 exercises
   9. Precision, Recall, and F1-Score9 min, 6 exercises
   10. The ROC Curve and AUC11 min, 6 exercises
6. Phase 5: Ensemble Learning6 lessons

   1. Phase 5 - Ensemble Learning5 min
   2. The Power of Ensembles - Why Combine Models?8 min, 6 exercises
   3. Bagging - Random Forest Regressor/Classifier8 min, 6 exercises
   4. Boosting - Introduction to AdaBoost8 min, 6 exercises
   5. Gradient Boosting (XGBoost, LightGBM, CatBoost)9 min, 6 exercises
   6. Stacking and Voting Classifiers8 min, 6 exercises
7. Phase 6: Unsupervised Learning & Dimensionality Reduction9 lessons

   1. Phase 6 - Unsupervised Learning & Dimensionality Reduction6 min
   2. Introduction to Clustering13 min, 6 exercises
   3. K-Means Clustering Algorithm11 min, 6 exercises
   4. Hierarchical Clustering (Dendrograms)13 min, 6 exercises
   5. DBSCAN - Density-Based Clustering13 min, 6 exercises
   6. Anomaly Detection with Isolation Forests12 min, 6 exercises
   7. Association Rule Learning (Apriori Algorithm)15 min, 6 exercises
   8. Principal Component Analysis (PCA)12 min, 6 exercises
   9. t-SNE and Manifold Learning14 min, 6 exercises
8. Phase 7: Model Optimization & Tuning7 lessons

   1. Phase 7 - Model Optimization & Tuning4 min
   2. Underfitting vs Overfitting8 min, 6 exercises
   3. Bias vs Variance Tradeoff10 min, 6 exercises
   4. K-Fold Cross-Validation8 min, 6 exercises
   5. Hyperparameter Tuning with GridSearchCV7 min, 6 exercises
   6. RandomizedSearchCV for Large Parameter Spaces8 min, 6 exercises
   7. The ML Pipeline - Automating the Workflow7 min, 6 exercises
9. Phase 8: Model Deployment (MLOps)6 lessons

   1. Phase 8 - Model Deployment (MLOps)5 min
   2. Saving and Loading Models (Pickle, Joblib)7 min, 6 exercises
   3. Building an ML API with Flask/FastAPI6 min, 6 exercises
   4. Deploying ML Models to Streamlit7 min, 6 exercises
   5. Dockerizing an ML Application8 min, 6 exercises
   6. Monitoring Model Drift10 min, 6 exercises
10. Phase 9: Interpretability & Responsible ML6 lessons

    1. Phase 9 - Interpretability & Responsible ML7 min
    2. Why Interpretability Matters10 min, 6 exercises
    3. Feature Importance and Its Traps9 min, 5 exercises
    4. Partial Dependence and ICE Plots10 min, 5 exercises
    5. SHAP Values from Scratch10 min, 5 exercises
    6. Fairness Metrics and Bias Auditing12 min, 5 exercises
11. Phase 10: Applied ML Problems7 lessons

    1. Phase 10 - Applied ML Problems7 min
    2. Imbalanced Classification and Fraud Detection11 min, 5 exercises
    3. Cost-Sensitive Learning and Decision Thresholds10 min, 5 exercises
    4. Time Series Forecasting Fundamentals10 min, 5 exercises
    5. Text Classification with TF-IDF10 min, 5 exercises
    6. Recommender Systems from Scratch9 min, 5 exercises
    7. Anomaly and Outlier Detection10 min, 5 exercises
12. Phase 11: ML Engineering5 lessons

    1. Phase 11 - ML Engineering6 min
    2. Experiment Tracking and Reproducibility10 min, 5 exercises
    3. Data Validation and Schema Contracts8 min, 5 exercises
    4. Feature Stores and Training-Serving Skew10 min, 5 exercises
    5. Testing ML Code9 min, 5 exercises
13. Phase 12: Capstone Projects6 lessons

    1. Phase 12 - Capstone Projects5 min
    2. Capstone 1 - Fraud Screening End to End8 min, 5 exercises
    3. Capstone 2 - Churn with Point-in-Time Features8 min, 5 exercises
    4. Capstone 3 - Demand Forecasting for Ordering8 min, 5 exercises
    5. Capstone 4 - Ticket Triage with Text8 min, 5 exercises
    6. Capstone 5 - A Recommender with an Honest Evaluation7 min, 5 exercises