The average prediction error between the predicted and tested data was around 4% based on the air temperature change across the dry channel.
The network predicts the second virial coefficient with the average prediction error of 1.3%.
These figures highlight that two thirds of the samples are very accurately predicted (with an error of two degrees at most in terms of latitude and eight degrees in term of longitude), but there also exist some isolated samples that are quite inaccurately predicted; these samples increase somewhat disproportionately the average prediction error and its standard deviation.
A comparison between range of predicted values at the design points and the average prediction error shows adequate precision.
Adequate precision compares the range of the predicted values at the design points with the average prediction error.
The results from 121 linear polymers showed that the correlation coefficient (R) between predicted and experimental values was 0.964 and the average prediction error was 0.87%.
PRESS is a signal-to-noise ratio, which compares the range of the predicted values at the design points to the average prediction error.
Adequate precision is a signal to-noise ratio, which compares the range of the predicted values at the design points to the average prediction error.
Comparison of true and predicted values suggested a good agreement with an average prediction error of 13.5 % (Table 3).
After training using back propagation algorithm, this neural network model predicts the concentrations of the output cytokines with an average prediction error of 0.0379 for the five output cytokines concentrations.
The empirical model had an overall average prediction error of 13.7% and the regression between measured and predicted values showed a determination coefficient of 0.77.
