Skip to main content
. 2024 Feb 15;10(4):e26182. doi: 10.1016/j.heliyon.2024.e26182

Table 5.

Comparison between our CNN and other states of the art CNN classifier on GTSRB.

Model param(M) FLOPs(G) Top-1 Acc. (%)
VGGNet [50] 138.36 15.47 91.74%
ResNet [50] 44.55 7.87 95.57%
ENet [21] 0.9 0.21 98.60%
MCDCNN [51] 38.5 7.14 99.50%
PFANet [30] 27.7 4.99 97.21%
ConvNeSe(ours) 26.87 4.46 99.85%