Short-Term Passenger Flow Prediction of Urban Rail Transit Based on SDS-SSA-LSTM

Table 1

SDS-SSA-LSTM model parameters.

Model parameters	SDS	SSA	LSTM
Number of neurons	—	—	32
Learning rate	—	—	0.001
Look back	—	—	3
Epochs	—	—	50
Window length	—	20	—
Contribution threshold	—	80%	—
Length of decomposition	50	—	—

Note: “—” indicates that the parameter in this row does not exist in this column model.