Skip to main content
. 2024 Mar 11;24(6):1792. doi: 10.3390/s24061792

Table 6.

ResNet model parameters.

Layer Name Output Size 18-Layer 34-Layer 50-Layer 101-Layer 152-Layer
Input layer -- (200,200,3)
Conv1 112 × 112 7 × 7, 64, stride 2
56 × 56 bn, 3×3 max pool, stride 2
Conv2_x 56 × 56 3×3,643×3,64×2 3×3,643×3,64×2 1×1,643×3,641×1,256×3 1×1,643×3,641×1,256×3 1×1,643×3,641×1,256×3
Conv3_x 28 ×28 3×3,1283×3,128×2 3×3,1283×3,128×4 1×1,643×3,641×1,256×4 1×1,643×3,641×1,256×3 1×1,643×3,641×1,256×8
Conv4_x 14×14 3×3,2563×3,256×2 3×3,2563×3,256×6 1×1,643×3,641×1,256×6 1×1,643×3,641×1,256×23 1×1,643×3,641×1,256×36
Conv5_x 7 × 7 3×3,5123×3,512×2 3×3,5123×3,512×3 1×1,643×3,641×1,256×3 1×1,643×3,641×1,256×3 1×1,643×3,641×1,256×3
Output layer 1 × 1 Average pool, 1000-d fc, softmax