Chantland/Hraf_Multilabel_K-foldsCrossValDemo

Continuation of Multi-Label Text classification model used to decode if passages contain a misfortunate event, a cause for misfortune, and/or an action to mollify or prevent some misfortune. This version implements 5 fold cross validation to improve model performance. We added additional training sets growing the model to 7277 passages train. The F1 micro score for 1820 passages not used for training or validation (the test set) was .851. individual class f1 scores shown below.
EVENT: 0.907
CAUSE: 0.822
ACTION: 0.805

Compare this to the old model using the same test set. F1 micro = .828
EVENT: 0.871
CAUSE: 0.81
ACTION: 0.793

For a quick demo, try typing in a sentence or even a paragraph in the Hosted inference API then pressing "compute"!