Project brief — image or text classification — Deep Learning & Neural Networks

The M6 Capstone is the culminating project of Course 5 — Deep Learning and Neural Networks. Over five lessons you will design, build, train, optimise, and deploy a complete deep learning system of your own choosing, applying every major technique from Modules 1 through 5 in a single cohesive project. The Capstone is deliberately open-ended: you may tackle either an image classification task (applying CNN, transfer learning, and data augmentation from M3) or a text classification task (applying transformers, BERT fine-tuning, and tokenisation from M5). Both paths share the same evaluation criteria — dataset design, model architecture justification, training discipline, evaluation rigour, and deployment quality. The project brief defines two concrete Capstone scenarios purpose-built for the cricket domain: (1) IPL Shot Type Classifier — a computer vision model identifying batting shot types from match frame images, and (2) IPL Commentary Sentiment and Event Classifier — an NLP model classifying live commentary into fine-grained event categories. Either project, when completed to the rubric standards, demonstrates production-ready deep learning engineering skill that directly maps to real IPL franchise analytics roles.

Analogy🏏Cricket

🏏 Think of it like cricket: The Duckworth-Lewis-Stern (DLS) method, the ICC bowling economy metric, and the net run rate each measure team performance but optimise for completely different things — DLS cares about wickets-in-hand, economy cares about runs-per-over, NRR cares about tournament-wide margins. Using the wrong metric to evaluate a bowler is like using MSE for classification: technically computable but optimising the wrong thing entirely. Just as you would never rank a spinner by batting average, you should never use MSE when your model predicts probabilities. Just as focal DLS adjustments down-weight easy chases and amplify close finishes, focal loss down-weights easy examples and amplifies hard minority-class examples. The choice of scoring metric defines what excellence means — and so does the choice of loss function.