Table 11 Training and test statistics of the models for EPAN prediction—Station 2 (Model 4) LSTM-DBO.

From: Metaheuristic-enhanced deep learning for monthly pan evaporation prediction under limited climatic data

Model inputs

Training period

Test period

 

RMSE

MAE

R2

NSE

RMSE

MAE

R2

NSE

M1 (70%−30%)

Tmin, Tmax

0.7476

0.5542

0.8852

0.8841

0.9281

0.7063

0.8482

0.8436

Tmin, Ra

0.7514

0.5861

0.8836

0.8831

0.8703

0.6692

0.8653

0.8614

Tmax, Ra

0.7973

0.6004

0.8728

0.8692

0.8326

0.6154

0.8685

0.8647

Tmin, Tmax, Ra

0.7708

0.5605

0.8784

0.8775

0.8242

0.6038

0.8738

0.8706

Tmin, Tmax, Ra, α

0.6972

0.5482

0.9026

0.8994

0.8058

0.5985

0.8804

0.8728

M2 (75%−25%)

Tmin, Tmax

0.4863*

0.3808

0.9492

0.9481

0.5293

0.4406

0.9382

0.9346

Tmin, Ra

0.7828

0.5806

0.8674

0.8656

0.8106

0.6258

0.8549

0.8535

Tmax, Ra

0.6553

0.5046

0.9076

0.9058

0.6584

0.5137

0.9038

0.9026

Tmin, Tmax, Ra

0.5167

0.4608

0.9385

0.9346

0.6708

0.4934

0.9156

0.9128

Tmin, Tmax, Ra, α

0.6309

0.5237

0.9003

0.8982

0.6839

0.5395

0.9064

0.9053

M3 (80%−20%)

Tmin, Tmax

0.6708

0.5356

0.9082

0.9061

0.7682

0.6144

0.8904

0.8868

Tmin, Ra

0.7136

0.5659

0.8934

0.8928

0.9236

0.7303

0.8328

0.8304

Tmax, Ra

0.8594

0.6508

0.8456

0.8449

0.9674

0.7738

0.8126

0.8089

Tmin, Tmax, Ra

0.7056

0.5337

0.8967

0.8953

0.8438

0.6356

0.8703

0.8664

Tmin, Tmax, Ra, α

0.6608

0.5205

0.9092

0.9081

0.7596

0.6019

0.8924

0.8896

  1. *Bold numbers indicates the best value (minimum RMSE or MAE and maximum R2 or NSE).
Back to article page