Skip to main content
Fig. 5 | BMC Medical Informatics and Decision Making

Fig. 5

From: A novel generative adversarial networks modelling for the class imbalance problem in high dimensional omics data

Fig. 5

Classic WGAN-WP Comparison Results. Validation scores are summarised across experiments using the microarray data for GAN, ExpandGAN, ClassicGAN and ClassicExpand methods. Boxes represent 1 standard deviation from the mean, and the horizontal line, of the box, the mean. The ‘AUC Diff’ defines the score difference between the 0.5 baseline (red, separated lines) and the validation score. GAN and ExpandGAN results shown are those using the alpha hyperparameter combination resulting in the greatest mean validation score. Area under the receiver operating characteristic curve (AUC); Wasserstein generative adversarial network with weight penalty (WGAN-WP)

Back to article page

BMC Medical Informatics and Decision Making

ISSN: 1472-6947

Contact us