Table 3 Comparison of prediction accuracy for each model.
From: Regional subsidence monitoring and prediction along high-speed railways based on PS-InSAR and LSTM
Point | Model | RMSE/mm | MAE/mm | MAPE/% | R2 |
|---|---|---|---|---|---|
3703 | BP | 5.67 | 0.55 | 10.91 | 0.82 |
LSTM | 4.46 | 0.45 | 7.72 | 0.86 | |
BP-Adaboost | 4.25 | 0.44 | 7.43 | 0.86 | |
LSTM-Adaboost | 3.47 | 0.37 | 6.91 | 0.89 | |
VMD-BP-LSTM | 2.21 | 0.33 | 7.45 | 0.92 | |
VMD-BPAda-LSTMAda | 1.05 | 0.29 | 6.87 | 0.93 | |
VMD-LSTM-LSTM | 1.15 | 0.25 | 6.36 | 0.94 | |
VMD-LSTMAda-LSTMAda | 0.82 | 0.25 | 6.31 | 0.94 | |
522 | BP | 10.34 | 1.10 | 9.85 | 0.87 |
LSTM | 9.95 | 0.94 | 8.13 | 0.88 | |
BP-Adaboost | 8.81 | 0.87 | 7.40 | 0.88 | |
LSTM-Adaboost | 8.04 | 0.79 | 6.94 | 0.90 | |
VMD-BP-LSTM | 7.43 | 0.73 | 6.82 | 0.91 | |
VMD-BPAda-LSTMAda | 4.58 | 0.57 | 5.57 | 0.93 | |
VMD-LSTM-LSTM | 4.46 | 0.56 | 5.40 | 0.94 | |
VMD-LSTMAda-LSTMAda | 1.32 | 0.46 | 4.92 | 0.95 |
