Fig. 1From: Interpreting patient-Specific risk prediction using contextual decomposition of BiLSTMs: application to children with asthmaValidation of contextual decomposition for LSTM and BiLSTM for the class c=1. The attribution is correct if the highest contribution among all visits is assigned to the artificial visit. The prediction curves indicate the prediction accuracy for class c=1, which also represents the upper bound for the attribution accuracyBack to article page