Improved Classification of White Blood Cells with the Generative Adversarial Network and Deep Convolutional Neural Network

Table 5

10-fold cross-validation accuracy of the DNN models initialized using random weights.

Model	Trans_aug1 (100 inst./class) (%)	Trans_aug2 (150 inst./class) (%)	Trans_aug3 (200 inst./class) (%)
VGG-16	91.5	92.1	92.9
VGG-19	92.3	92.8	93.4
ResNet-18	91.4	92.6	93.2
ResNet-50	93.5	94.0	94.7
DenseNet-121	94.4	94.8	95.4
DenseNet-169	94.9	95.4	95.8