Batch Runner producing non-feasible outputs

It may be helpful to have a method within batch runner that removes any unconverged and physiologically infeasible simulations. It would be good to have the option of using all simulations outputted from the sampler OR only the feasible ones.

Keeping track of discarded ones is also important - I think.

When only the feasible simulations are taken, the input parameter array will also need to be adjusted so that it corresponds with the outputs used.

For example, if you sample the input space 10 times and the resulting outputs fail to converge or produce sensible results at input 2 and 5, then the input matrix will need row 2 and 5 removing.

Hope that makes sense,.

Thanks.

Start by locating the batch runner and sampler flow, then determine how unconverged or physiologically infeasible outputs are represented. Done should provide an option to retain all simulations or filter to feasible ones, keep discarded simulations identifiable, and remove the matching rows from the input parameter array.

Written by the indexing model from the issue text.