To derive a stress-strain profile, I utilized "StressStrain.pl" in example directory. This code applys a certain amount of stress by Souza-Martins barostats and get the resultant measured stress by averaging the result of \\\$numProductionStepsPerStressValue. 

The main problem of my run is discrepancy between applied stress and measured stress. The stress fluctuation of the measured stress is so huge to calculate exact stress state. The result does not seem to stress-strain profile. How can I get a clean stress-strain profile?