Figure 6
From: DNdisorder: predicting protein disorder using boosting and deep networks
Performance of a boosted ensemble. To determine the effectiveness of boosting we evaluated the performance of an ensemble as a function of the number of rounds of boosting. The figure displays the balanced accuracy and area under the ROC curve for the DISORDER723 dataset. The predictions were made using a 10 fold cross validation procedure with input windows 30 residues long and a target window of 3 residues.