Curriculum: Beginner to Professional

The course is organised into four progressive levels. Each one develops practical skills through guided exercises, assignments, and real projects.

Beginner Level – Python and Data Preparation

Start with Python, the language most widely used in machine learning. Learn variables, data types, control flow, functions, collections, and file handling. Move into the core data libraries used in ML, covering arrays, dataframes, and numerical operations. Study data cleaning, handling missing values, outlier detection, and data type conversion. Learn exploratory data analysis, including summary statistics and visualisation, so you understand the shape and quality of your data before modelling it. Get comfortable with notebooks, version control, and reproducible workflows.

Intermediate Level – Core Supervised and Unsupervised Learning

Learn the fundamental ideas behind how models learn from data, what overfitting and underfitting mean, and why validation matters. Study supervised learning algorithms for regression, including linear models and regularised variants, and for classification, including logistic regression and decision trees. Explore unsupervised learning techniques such as k-means clustering, hierarchical clustering, and dimensionality reduction with principal component analysis. Learn the appropriate evaluation metrics for each problem type, including accuracy, precision, recall, F1 score, mean squared error, and R-squared. Study train-test splits, cross-validation, and the bias-variance trade-off. Build complete pipelines from raw data to evaluated model.

Advanced Level – Feature Engineering, Ensembles, and Tuning

Move into the techniques that separate competent models from strong ones. Learn feature engineering, including encoding categorical variables, scaling, binning, interaction features, and date handling. Study feature selection methods to reduce noise and improve interpretability. Explore ensemble methods, including random forests, gradient boosting, and stacking, and understand why combining models often outperforms single ones. Learn hyperparameter tuning through grid search, random search, and Bayesian optimisation. Study techniques for handling imbalanced data, including resampling strategies and class weighting. Learn to interpret models using feature importance and explanation techniques.

Professional Level – Deployment and Production Workflows

Bring everything together by turning models into working systems. Learn to save and load trained models, package them into reusable components, and expose predictions through APIs. Study deployment options across cloud and on-premise environments. Learn monitoring practices for detecting data drift and performance degradation after launch. Explore retraining strategies and versioning practices that keep models reliable over time. Study documentation and governance so that your models are understandable and auditable. Complete a capstone project that demonstrates your full skill set, from data preparation and feature engineering through modelling and evaluation to deployment.