Convolutional Neural Network (CNN) models were created in this project for Optical Proximity Correction (OPC) team to effectively identify hot-spot integrated circuit patterns, so that many months of physical mask respin validation could be saved.
The CNN models were trained on the open ICCAD 2012 Dataset which can be downloaded from: https://drive.google.com/file/d/1In9dr5kaAXbflTDn9uUZR1yDHplWnnU4/view