AI & ML · Intermediate
Applied Machine Learning for Scientists & Engineers
Build, validate and interpret ML models on environmental, agricultural and engineering data with Python and scikit-learn.
Course modules
- 01Problem framing, data audit and feature engineering
- 02Regression, classification and ensemble methods (RF, XGBoost)
- 03Cross-validation, spatial & temporal leakage, uncertainty
- 04Explainability with SHAP and partial dependence
- 05Reproducible pipelines and model reporting
- 06Capstone: an end-to-end model on your own data
Learning outcomes
- Choose and justify the right model for a scientific question
- Avoid the validation mistakes reviewers reject
- Explain model behaviour to non-technical stakeholders
- Leave with a portfolio-ready, documented project
Project-based training
Every participant completes a capstone: Capstone: an end-to-end model on your own data. Labs use real environmental, agricultural or engineering datasets.
Tools
- Python
- pandas
- scikit-learn
- XGBoost
- SHAP
Who it’s for
PG students, researchers, engineers and analysts
Prerequisites
Basic Python and statistics
