Table 5 The ablation study presents the comparison of the coupled model and the baseline model in terms of the same evaluation metrics at the GD and ET hydrological stations in the testing dataset.

From: Leveraging multi-source data and teleconnection indices for enhanced runoff prediction using coupled deep learning models

Evaluation indicators

GD

ET

SVR

SRA-SVR

MLPR

SRA-MLPR

SVR

SRA-SVR

MLPR

SRA-MLPR

NSE

0.6787

0.7383±0.0005

0.7153

0.7679±0.0004

0.6368

0.6869±0.0005

0.6592

0.7146±0.0004

RMSE

636.3489

591.0165±2

620.3271

556.6857±2

798.4798

747.4300±2

779.5903

594.4508±2

MAE

389.8963

356.5494±1.6

381.9365

345.3250±1.5

485.7863

438.0738±1.6

456.8376

407.8214±1.5

MAPE

0.2697

0.2200±0.0002

0.2509

0.2084±0.0001

0.2615

0.2182±0.0002

0.2383

0.2090±0.0001

MSLE

0.0798

0.0676±0.0001

0.0722

0.0624±0.0001

0.0943

0.0723±0.0001

0.0873

0.0652±0.0001

  1. The units of RMSE and MAE in the table are both m3/s. Both SRA-SVR and SRA-MLPR were performed in 100 replicate trials due to the random initial weights of network, thus the metrics were expressed as the mean ± standard deviation.
Back to article page