Open StephanSchaller opened 2 years ago
https://computing.llnl.gov/projects/sundials/cvodes
No, I don't think so (At least Juri said not). So this is the adjoint sensitivities for use in the PI
Yep, it's not done for the sensitivity calculations yet.
Here we could provide the analytical solution not only for dRHS/dY
(which is calculated and used also for the ODE solving without sensitivites) but also dRHS/dP
- which is not done yet.
https://sundials.readthedocs.io/en/latest/cvodes/Mathematics_link.html#forward-sensitivity-analysis
Isn't it the case already?