Advances, Systems and Applications
Table 5 Hyperparameter settings of TBUFCM model
From: Security issues of news data dissemination in internet environment
Parameter Name | Parameter Meaning | Parameter setting |
|---|---|---|
epoch | Number of iterations | 100 |
learning_rate | Learning rate | 0.01 |
optimization | Optimization algorithm | Adam |
loss_function | Loss function | Cross_entropt_loss |
filter_sizes | Kernel size | [3,4,5] |
linear_layers | Number of fully connected hidden layers | 10 |
GRU_layers | Number of GRU hidden layers | 32 |
hidden_size | Number of GRU hidden layer neurons | 5 |
batch_size | Batch sample size | 32 |
drop_rate | The fraction of lost neurons | 0.5 |