Unit 15 of 15
Integrated Projects
- 3 lessons
- Videos 89 to 91
The course ends with complete projects. Two of them follow the shared workflow for regression and for rare-event classification, and a capstone designs, audits and defends a full study.
The regression project audits the data, protects a final test set, compares a baseline, a regularized linear model and a tree ensemble, and reports the residuals and the limits of the result. The classification project adds calibration on a separate sample, a threshold chosen from review costs and capacity, and a subgroup audit. The capstone starts from a flawed study with a perfect score, writes a corrected protocol before the final comparison, resamples the independent unit to measure uncertainty and reports a failed alternative and an unresolved limitation.
Every lesson will be one video. The videos are coming soon, and each lesson page already has the definitions and an example for every key term.
Regression and Classification Projects
Regression Project from Data to Evaluation
How a regression project runs from a data audit and a protected test set through a baseline, a regularized linear model and a tree ensemble to residual analysis, signed mean error and out-of-range inputs.
Video 89
Rare-Event Classification Project
How a rare-event classification project links an event definition, a calibration sample, a cost-based threshold under review capacity, a frozen procedure and a subgroup audit to a final report with honest limits.
Video 90
Capstone Study
Design and Defend a Machine Learning Study
How to design, audit and report a machine learning study, with a study protocol, no post-outcome information, the independent unit of analysis, a failed alternative, reproducibility versus validity and a stated limitation.
Video 91