Skip to main content

Table 5 The RMSE of training algorithms for the full-featured dataset

From: Comparing artificial neural network training algorithms to predict length of stay in hospitalized patients with COVID-19

Training algorithm

The number of neurons

5

10

15

20

25

30

LM

2.3338

2.3825

2.4444

2.3863

2.3782

3.0873

BR

1.9291

1.732

1.6963

1.6213

1.9454

1.8312

BFGS Quasi-Newton

2.3948

2.3701

2.3726

2.3932

2.2855

2.2507

RP

2.3513

2.3365

2.4433

2.6511

3.2506

2.8938

SCG

2.4085

2.4049

2.3882

2.3791

2.3944

2.322

CGB

2.3927

2.4045

2.4045

2.4301

2.3861

2.3365

CGF

2.4015

2.3845

2.3936

2.3581

2.3642

2.319

CGP

2.3966

2.3989

2.4514

2.4038

2.3545

2.7662

OSS

2.3706

2.3924

2.4454

2.4101

2.5861

2.8005

GDX

2.405

2.3985

2.8746

2.8096

2.6103

2.9711

GDM

16.0903

23.6745

26.0844

25.2706

32.1784

41.2996

GD

16.896

42.1661

36.6554

30.9327

26.1043

29.3946