Table 2 Hyperparameters of all components in each convolutional layer
From: iProL: identifying DNA promoters from sequence information based on Longformer pre-trained model
CNN layer | Conv1d | BatchNorm1d | MaxPool1d | Dropout |
|---|---|---|---|---|
Layer-1 | Filter = 128 Kernel size = 8 Stride = 2 | Num Features = 128 | Kernel size = 3 Stride = 1 | p = 0.8 |
Layer-2 | Filter = 64 Kernel size = 8 Stride = 2 | Num Features = 64 | Kernel size = 3 Stride = 1 | p = 0.8 |
Layer-3 | Filter = 32 Kernel size = 3 Stride = 2 | Num Features = 32 | Kernel size = 3 Stride = 1 | Not used |