Skip to main content
. 2017 Nov 1;11:534. doi: 10.3389/fnhum.2017.00534

Figure 4.

Figure 4

(A) Above shows the ROC curves for the training set, validation set and the test set as derived from the input data in a randomized fashion according to a 75-15-15 split, for KM-real. We see that the MLP training was successful and good. (B) Shows the architecture of the softmax MLP used for all the MLP models here.