STGNN-FAM: A Traffic Flow Prediction Model for Spatiotemporal Graph Networks Based on Fusion of Attention Mechanisms

Table 2

Experimental parameter configuration table.

Parameter	Meaning	Value
Epoch	Iteration round	300
BatchSize	Number of batches	16
Initial_LR	Initial learning rate	5e-4
Heads	Number of GAT heads	8
GRU_HiddenSize	GRU hidden layers	128