Medical Specialty Classification Based on Semiadversarial Data Augmentation

Figure 9

Visualization comparison for the output of the softmax layer. The samples used are exactly same as in Figure 8. (a1–a3) The visualization results under SRA and (b1–b3) the visualization results under SemiADA. It is easy to observe that the distribution of SemiADA is significantly wider (area covered by the same color).