Machine Learning Roadmap — Step-by-step Learning Path
Supervised, unsupervised learning and classical ML algorithms
The Machine Learning roadmap
- Python Programming Fundamentals — beginner
- Python Syntax & Data Types — beginner
- Control Flow & Functions — beginner
- Data Structures (Lists, Dicts, Sets, Tuples) — beginner
- Object-Oriented Python — beginner
- File I/O & Working with Modules — beginner
- Virtual Environments & pip — beginner
- Mathematics & Statistics for ML — beginner
- Linear Algebra Essentials — beginner
- Calculus Essentials — beginner
- Probability Fundamentals — beginner
- Descriptive Statistics — beginner
- Probability Distributions — beginner
- Hypothesis Testing & Confidence Intervals — beginner
- Bayesian Thinking Basics — beginner
- Optimization Basics (Gradient Descent Intuition) — beginner
- Python for Data Science — beginner
- NumPy for Numerical Computing — beginner
- Pandas for Data Manipulation — beginner
- Data Visualization with Matplotlib & Seaborn — beginner
- Jupyter Notebooks & Reproducible Workflows — beginner
- Data Analysis & Preprocessing — intermediate
- Exploratory Data Analysis (EDA) — intermediate
- Handling Missing Data — intermediate
- Data Cleaning & Outlier Detection — intermediate
- Feature Scaling & Normalization — intermediate
- Encoding Categorical Variables — intermediate
- Working with Different Data Sources (CSV/SQL/APIs) — intermediate
- Supervised Learning: Regression — intermediate
- Linear Regression — intermediate
- Polynomial Regression — intermediate
- Regularization (Ridge, Lasso, Elastic Net) — intermediate
- Regression Evaluation Metrics (MAE, MSE, RMSE, R²) — intermediate
- Gradient Descent for Regression — intermediate
- Supervised Learning: Classification — intermediate
- Logistic Regression — intermediate
- k-Nearest Neighbors (kNN) — intermediate
- Decision Trees — intermediate
- Naive Bayes — intermediate
- Support Vector Machines (SVM) — intermediate
- Classification Evaluation Metrics — intermediate
- Ensemble Methods — intermediate
- Bagging & Random Forests — intermediate
- Boosting (AdaBoost, Gradient Boosting) — intermediate
- XGBoost, LightGBM & CatBoost — advanced
- Stacking & Blending Models — advanced
- Unsupervised Learning — intermediate
- k-Means Clustering — intermediate
- Hierarchical Clustering — intermediate
- DBSCAN — intermediate
- Dimensionality Reduction with PCA — intermediate
- t-SNE & UMAP for Visualization — advanced
- Anomaly Detection — advanced
- Model Evaluation & Validation — advanced
- Train/Test/Validation Splits — advanced
- Cross-Validation Techniques — advanced
- Bias-Variance Tradeoff — advanced
- Overfitting & Underfitting — advanced
- Hyperparameter Tuning (Grid Search, Random Search) — advanced
- Learning Curves & Diagnostics — advanced
- Feature Engineering & Selection — advanced
- Feature Creation & Domain-Driven Features — advanced
- Feature Selection Techniques — advanced
- Handling Imbalanced Datasets — advanced
- Text Feature Engineering (Bag of Words, TF-IDF) — advanced
- Time-Based Feature Engineering — advanced
- Introduction to Neural Networks — advanced
- Perceptrons & the Building Blocks of Neural Networks — advanced
- Feedforward Neural Networks Basics — advanced
- Activation Functions — advanced
- Backpropagation Intuition — advanced
- When to Use Classical ML vs Neural Networks — advanced
- Model Interpretability & Explainability — advanced
- Feature Importance — advanced
- SHAP Values — advanced
- LIME — advanced
- Partial Dependence Plots — advanced
- Communicating Model Decisions to Stakeholders — advanced
- ML Engineering & MLOps Fundamentals — advanced
- ML Project Structure & Reproducibility — advanced
- Experiment Tracking (MLflow/Weights & Biases) — advanced
- Model Versioning & Registries — advanced
- Data Versioning (DVC) — advanced
- Pipelines with scikit-learn Pipeline/ColumnTransformer — advanced
- Deploying ML Models — advanced
- Serializing Models (Pickle/joblib/ONNX) — advanced
- Serving Models via a REST API (FastAPI/Flask) — advanced
- Containerizing ML Models with Docker — advanced
- Batch vs Real-Time Inference — advanced
- Model Monitoring & Drift Detection — advanced
- A/B Testing for Models — advanced
- Real-World ML Case Studies & Capstone — advanced
- Case Study: Customer Churn Prediction — advanced
- Case Study: Fraud Detection with Imbalanced Data — advanced
- Case Study: Recommendation System Basics — advanced
- Case Study: Time Series Forecasting — advanced
- Capstone: End-to-End ML Project — advanced