Fig. 4
From: Using autoencoders as a weight initialization method on deep neural networks for disease detection
ROC-AUC and Precision-Recall Curves for the RNA-Seq data, with Approach B. Classification Model ROC-AUC — a to f — and b Precision-Recall — g to l — curves when detecting breast cancer detection in the RNA-Seq input data, with Approach B (allowing fine tune on the weights imported from the AE) combined with both AE importing strategies. Each line represent the different 10 folds for the cross validation