This was a kaggle competition in which I had trained an object detection model on chest X-rays for detecting several medical conditions. There were 14 types of abnormalities found in the chest Xray. The task was to localize and classify these anomalies.
For more information please visit https://www.kaggle.com/c/vinbigdata-chest-xray-abnormalities-detection/overview
The 'retinanet' directory inside 'codes' directory was borrowed directly from https://github.com/yhenon/pytorch-retinanet/tree/master/retinanet