BERT-BiGRU Intelligent Classification of Metro On-Board Equipment Faults Based on Key Layer Fusion

Table 2

Parameter setting.

Parameter	Value
Batch_size	128
Learning rate	8-4
Optimizer	Adam
Hidden layer dimension	768
Epoch	30
LSTM_dim	256
GRU_dim	256
Activation function	Relu
Max_seq_len	128
Dropout	0.5