Neural network hyperparameters. We specify the types of layers and the hyperparameters used when defining each layer. ‘NN’ indicates the number of neurons and ‘Activ.’ indicates the activation function used.
| Dense Branch A . | Convolutional Branch B . | Interpretation Stage . | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Layer . | Size . | NN . | Activ. . | Layer . | Size . | Filters . | Kernel . | Stride . | Activ. . | Layer . | NN . | Rate . | Activ. . |
| Input | [9,1] | – | – | Input | [11,150,1] | – | – | – | – | Concat. | – | – | – |
| Dense | [64] | ReLU | Conv2D | – | [16] | [3,7] | [1,1] | ReLU | Dense | [64] | – | ReLU | |
| Dense | [32] | ReLU | Conv2D | – | [16] | [3,7] | [1,1] | ReLU | Dense | [32] | – | ReLU | |
| Flatten | – | – | – | Max-Pooling | [1,7] | – | – | – | – | Dropout | – | 0.75 | – |
| Conv2D | – | [16] | [3,19] | [1,1] | ReLU | Dense | [1] | – | Sigmoid | ||||
| Conv2D | – | [16] | [3,19] | [1,1] | ReLU | ||||||||
| Global | – | – | – | – | – | ||||||||
| Avg-Pooling | |||||||||||||
| Dense Branch A . | Convolutional Branch B . | Interpretation Stage . | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Layer . | Size . | NN . | Activ. . | Layer . | Size . | Filters . | Kernel . | Stride . | Activ. . | Layer . | NN . | Rate . | Activ. . |
| Input | [9,1] | – | – | Input | [11,150,1] | – | – | – | – | Concat. | – | – | – |
| Dense | [64] | ReLU | Conv2D | – | [16] | [3,7] | [1,1] | ReLU | Dense | [64] | – | ReLU | |
| Dense | [32] | ReLU | Conv2D | – | [16] | [3,7] | [1,1] | ReLU | Dense | [32] | – | ReLU | |
| Flatten | – | – | – | Max-Pooling | [1,7] | – | – | – | – | Dropout | – | 0.75 | – |
| Conv2D | – | [16] | [3,19] | [1,1] | ReLU | Dense | [1] | – | Sigmoid | ||||
| Conv2D | – | [16] | [3,19] | [1,1] | ReLU | ||||||||
| Global | – | – | – | – | – | ||||||||
| Avg-Pooling | |||||||||||||
Neural network hyperparameters. We specify the types of layers and the hyperparameters used when defining each layer. ‘NN’ indicates the number of neurons and ‘Activ.’ indicates the activation function used.
| Dense Branch A . | Convolutional Branch B . | Interpretation Stage . | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Layer . | Size . | NN . | Activ. . | Layer . | Size . | Filters . | Kernel . | Stride . | Activ. . | Layer . | NN . | Rate . | Activ. . |
| Input | [9,1] | – | – | Input | [11,150,1] | – | – | – | – | Concat. | – | – | – |
| Dense | [64] | ReLU | Conv2D | – | [16] | [3,7] | [1,1] | ReLU | Dense | [64] | – | ReLU | |
| Dense | [32] | ReLU | Conv2D | – | [16] | [3,7] | [1,1] | ReLU | Dense | [32] | – | ReLU | |
| Flatten | – | – | – | Max-Pooling | [1,7] | – | – | – | – | Dropout | – | 0.75 | – |
| Conv2D | – | [16] | [3,19] | [1,1] | ReLU | Dense | [1] | – | Sigmoid | ||||
| Conv2D | – | [16] | [3,19] | [1,1] | ReLU | ||||||||
| Global | – | – | – | – | – | ||||||||
| Avg-Pooling | |||||||||||||
| Dense Branch A . | Convolutional Branch B . | Interpretation Stage . | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Layer . | Size . | NN . | Activ. . | Layer . | Size . | Filters . | Kernel . | Stride . | Activ. . | Layer . | NN . | Rate . | Activ. . |
| Input | [9,1] | – | – | Input | [11,150,1] | – | – | – | – | Concat. | – | – | – |
| Dense | [64] | ReLU | Conv2D | – | [16] | [3,7] | [1,1] | ReLU | Dense | [64] | – | ReLU | |
| Dense | [32] | ReLU | Conv2D | – | [16] | [3,7] | [1,1] | ReLU | Dense | [32] | – | ReLU | |
| Flatten | – | – | – | Max-Pooling | [1,7] | – | – | – | – | Dropout | – | 0.75 | – |
| Conv2D | – | [16] | [3,19] | [1,1] | ReLU | Dense | [1] | – | Sigmoid | ||||
| Conv2D | – | [16] | [3,19] | [1,1] | ReLU | ||||||||
| Global | – | – | – | – | – | ||||||||
| Avg-Pooling | |||||||||||||
This PDF is available to Subscribers Only
View Article Abstract & Purchase OptionsFor full access to this pdf, sign in to an existing account, or purchase an annual subscription.
