Table 2 The Impact of LSTM Time Lag Length on Predicting Water Inrush.

From: Predicting mine water inflow volumes using a decomposition-optimization algorithm-machine learning approach

Lag(day)

1

3

5

7

10

15

20

25

30

35

MAE

187.492

186.468

189.754

188.424

193.453

192.962

204.101

201.930

213.504

220.818

MAPE

2.825%

2.811%

2.866%

2.841%

2.924%

2.920%

3.0981%

3.067%

3.258%

3.377%

RMSE

256.459

252.381

255.364

254.289

259.491

259.038

266.827

264.621

276.842

285.704

NSE

0.719

0.724

0.716

0.718

0.705

0.705

0.693

0.700

0.686

0.676

Back to article page