Hi, the Ostrich warm start does not seem to work.
When doing a warm start, Ostrich reads in the objective function values in the output file as parameter values, and therefore doesn't restart the model from the best parameter set.
Moreover, I'm wondering if a calibration process could actually be resumed, even if the above issue was fixed. Ostrich does not seem to read parameter values for previous simulation trials in order to benefit from previous searches' information regarding parameter sensitivity and preferred values, or to read in how many simulation trials are left based on runs previously done and the max. number of runs specified in Ostin.txt.
Thanks,
Etienne Gaborit