RINA - Démonstrateur d’utilisation de l’intelligence artificielle pour la gestion opérationnelle des risques naturels gravitaires géologiques | FEREC 2020

40 APPENDIX A. RESULTS OF SERIES OF EXPERIMENTS USING DNN ============================================================================ input_6 (InputLayer) [(None, 10)] 0 ____________________________________________________________________________ tf.expand_dims (None, 10, 1) 0 ____________________________________________________________________________ conv1d (Conv1D) (None, 8, 100) 400 ____________________________________________________________________________ batch_normalization (None, 8, 100) 400 ____________________________________________________________________________ dropout_3 (Dropout) (None, 8, 100) 0 ____________________________________________________________________________ flatten (Flatten) (None, 800) 0 ____________________________________________________________________________ main (Dense) (None, 2) 1602 ============================================================================ Total params: 2,402 Trainable params: 2,202 Non-trainable params: 200 ____________________________________________________________________________ RESULTS on VALIDATION SET -> confusion matrix: 0 1 0 805 133 1 80 62 -> precision:[0.91 0.32] -> recall:[0.86 0.44] -> accuracy:0.8 -> balanced_accuracy:0.65 -> fbeta_score:0.37