Text classification model used to decode passages that contain misfortunate events. Current F1 score of 140 passages not used for training is .94. <br><br><br> For a quick demo, try typing in a sentence or even a paragraph in the <b>Hosted inference API</b> then pressing "compute"!