Skip to main content

Table 2 Network structure of the proposed method. The basic block is engaged from ResNet18 [20]

From: Using deep learning to predict the outcome of live birth from more than 10,000 embryo data

Layer Filter Size Output Size
Conv1_x 7 × 7, 64
3 × 3, 64
3 × 3, 128, stride 2
224 × 224
224 × 224
112 × 112
Conv2_x \(\left[\begin{array}{c}3\times 3,128\\ {}3\times 3,128\end{array}\right]\times 3\)
3 × 3, 256, stride 2
112 × 112
56 × 56
Conv3_x \(\left[\begin{array}{c}3\times 3,256\\ {}3\times 3,256\end{array}\right]\times 3\)
3 × 3, 512, stride 2
56 × 56
28 × 28
Conv4_x \(\left[\begin{array}{c}3\times 3,512\\ {}3\times 3,512\end{array}\right]\times 3\)
3 × 3, 1024, stride 2
28 × 28
14 × 14
Conv5_x \(\left[\begin{array}{c}3\times 3,1024\\ {}3\times 3,1024\end{array}\right]\times 3\)
3 × 3, 2048, stride 2
14 × 14
7 × 7
Conv6_x \(\left[\begin{array}{c}3\times 3,2048\\ {}3\times 3,2048\end{array}\right]\times 2\)
3 × 3, 2048
7 × 7
5 × 5
Conv7_x \(\left[\begin{array}{c}3\times 3,2048\\ {}3\times 3,2048\end{array}\right]\times 2\)
3 × 3, 2048
5 × 5
3 × 3
Fc1 Max pool  3 × 3
2048-d fc
1 × 1