Skip to main content

Table 4 Parameters of the ANN model

From: A quantitative structure- property relationship of gas chromatographic/mass spectrometric retention data of 85 volatile organic compounds as air pollutant materials by multivariate methods

optimized parameters of the ANN

Transfer function

 

Logsig

Training function

 

Traingdx

Number of input neurons

 

6

Number of hidden neurons

 

6

Number of output neurons

 

1

Learning rate

 

0.5

Momentum

 

0.5

R 2

R 2 cal

R 2 pre

RMSEC a

RMSEP b

0.977

0.983

0.967

0.173

0.315

  1. a Root mean squared error of calibration
  2. b Root mean squared error of prediction