Table 3.

The AUPRC and AUROC obtained in 25% of the training data set left out for testing

Models	AUPRC	AUROC
RF-HOT	0.802	0.938
RF-TETRA	0.593	0.844
GRU-0	0.752	0.929
GRU-1	0.778	0.934
GRU-2	0.753	0.929
GRU-3	0.728	0.922
GRU-4	0.728	0.923
LSTM-0	0.734	0.923
LSTM-1	0.744	0.927
LSTM-2	0.739	0.923
LSTM-3	0.748	0.924
LSTM-4	0.748	0.928

The data set used has a 1:10 ratio of positive to negative instances. The numbers in bold indicate the models with the highest AUPRC/AUROC per machine learning method