Table 2 Computational cost of five methods on PEMS04 dataset.
From: Multi-scale spatio-temporal graph neural network for urban traffic flow prediction
Model | Training Time (s/epoch) | inference time on testset(s) |
|---|---|---|
DSTAGNN | 1380.68 | 663.41 |
ISTNet | 548.52 | 79.06 |
PDFormer | 557.39 | 71.16 |
DEC-Former | 543.03 | 70.31 |
STGMS | 538.59 | 66.79 |
