Wafer map defect pattern classifier

The brief

Defect patterns on a wafer (rings, scratches, edge failures) point to specific process problems. Train an image model that recognises these patterns from wafer maps, handles the heavy class imbalance, and shows where it gets confused. Good looks like: accuracy per pattern, imbalance handled, and a confusion matrix engineers can act on.

What you will build

1. Load and clean the wafer maps
2. Handle the imbalance with augmentation or weighting
3. Train a CNN and compare with a simpler model
4. Evaluate per defect pattern
5. Study confusions between similar patterns

Free, open data to build it with

• WM-811K wafer map dataset ↗ Kaggle

Check each dataset's licence before you publish your work. Your mentor gives your team its own version of the brief (data, scope or region), so every team's project is its own.