Fig. 5
From: Genome-enabled prediction using probabilistic neural network classifiers
The upper curve is the ROC curve (AUC) with recall vs false positive rate. The lower curve is the precision-recall curve AUCpr with precision vs recall for the a upper 15 % class of grain yield under well-watered conditions (GY-WW) of classifiers MLP (green) and PNN (blue); b upper 30 % class of trait grain yield under well-watered conditions (GY-WW) of MLP (green) and PNN (blue); c middle 40 % class of trait anthesis-silking interval under well-watered conditions (ASI-WW) of MLP (green) and PNN (blue) and d middle 70 % class of trait anthesis-silking interval under well-watered conditions (ASI-WW) of MLP (green) and PNN (blue); e lower 15 % class of trait female flowering under well-watered conditions (FFL-WW) of MLP (green) and PNN (blue); f lower 30 % class of trait female flowering under well-watered conditions (FFL-WW) of MLP (green) and PNN (blue)