. 2020 Jan 28;10:1631. doi: 10.3389/fphar.2019.01631

Table 2.

Algorithm and training procedure of convolution-capsule networks (Conv-CapsNet).

Algorithm: Conv-CapsNet training algorithm, using a mini-batch stochastic gradient descent (SGD) for simplicity.

Input: mini batch feature vector (x);
Number of Conv-CapsNet training epoch (S);
Number of dynamic routing iterations (iter).
Output: Length of each capsules (Len).

1: For n=1 to S do
2: conv_layer ← conv(x, CW)
3: hf_layer ← fc1(conv_layer, W1)
4: pc_layer ← fc2(hf_layer, W2)
5: u ← Encapule(pc_layer)
6: For all capsule i in PrimaryCaps layer:

{\hat{u}}_{j | i} \leftarrow W_{i j} u_{i}

………… {contribution computes Eq. 1}
7: For all capsule i in PrimaryCaps layer and capsule j in DigitCaps layer:b_ij ← 0
8: For m=1 to iter do
9: For all capsule i in PrimaryCaps layer: c_i ← softmax(b_i) ……{softmax computes Eq. 2-1}
10: For all capsule j in DigitCaps layer:

s_{j} \leftarrow \sum_{i} c_{i j} {\hat{u}}_{j | i}

…{dynamic computes Eq. 2-2}
11: For all capsule j in DigitCaps layer: v_j ← squash (s_j) ………{squash computes Eq. 2-3}
12: For all capsule i in PrimaryCaps layer and capsule j in DigitCaps layer:

b_{i j} \leftarrow b_{i j} + {\hat{u}}_{j | i} \cdot v_{j}

13: End for
14:

L e n \leftarrow L e n g t h o f v

15:

L \leftarrow l o s s o f v

………………………………………{loss computes Eq. 3}
16:

W \leftarrow W - \partial L / \partial W

17:

C W \leftarrow C W - \partial L / \partial C W

18:

W 1 \leftarrow W 1 - \partial L / \partial W 1

19:

W 2 \leftarrow W 2 - \partial L / \partial W 2

20: End for