Table 2 The statistical parameters of developed GPRs.

From: An insight into the estimation of frost thermal conductivity on parallel surface channels using kernel based GPR strategy

 

R2

MRE (%)

MSE

RMSE

STD

GPR (rational quadratic)

Train

0.987

4.08134

0.00005

0.00707

0.00471

Test

0.983

4.67914

0.00004

0.00636

0.00439

Total

0.986

4.22817

0.00005

0.00636

0.00460

GPR (exponential)

Train

0.995

2.82505

0.00002

0.00444

0.00334

Test

0.990

3.82609

0.00003

0.00554

0.00436

Total

0.993

3.07092

0.00002

0.00554

0.00359

GPR (squared exponential)

Train

0.990

3.71123

0.00004

0.00618

0.00475

Test

0.985

3.95804

0.00004

0.00603

0.00419

Total

0.989

3.77185

0.00004

0.00603

0.00458

GPR (matern)

Train

0.997

1.83752

0.00001

0.00335

0.00257

Test

0.994

2.38225

0.00001

0.00370

0.00282

Total

0.996

1.97131

0.00001

0.00370

0.00261

Back to article page