Pearson's correlation between the predicted values (mean of the predictive distribution) and the observed PTA, Cor y ^ test, y PTA ), and predictive mean-squared error (MSE) defined as ∑ i = 1 2499 (ŷ i test − y i PTA ) 2 / n were computed to evaluate the predictive ability of the two kernels.
The temperature evolution due to localized microplasticity was considered as the fatigue damage indicator, and good predictions were confirmed between the predicted values and the traditional values.
The Adequate Precision (AP) representing the error between the predicted values at the design points and the average prediction.
At each step, the predictive accuracy of the markers was assessed by Pearson's correlation between the predicted values and the phenotypes.
Good agreement is observed between the predicted values and experiment results from other studies.
The deviations between the predicted values and CFD values can be less than 3%.
The percentages of error all fall within 1%, between the predicted values and the experimental values.
We then compute the Spearman correlation between the predicted values and the signature effectiveness.
From this, it is clearly seen that a good agreement is observed between the predicted values and experimental measurements.
The predicted results using RA and ANN model indicate a good agreement between the predicted values and experimental results.
The results show a considerably high agreement between the predicted values from the correlation and measured experimental observations.
