Open rt17603 opened 3 years ago
Original report by Anita Ganesan (Bitbucket: [Anita Ganesan](https://bitbucket.org/Anita Ganesan), GitHub: ag12733).
Currently model error is the only hyper-parameter but there is no option to fix it (i.e. set the stepsize for the model error to 0). This is needed to be able to replicate other papers or to simplify the inversions for testing things are working.
Original comment by Rachel Tunnicliffe (Bitbucket: rt17603, GitHub: rt17603).
@{5f6329c3a1048d0069fa099c} to check if this has been completed with a PR
Original report by Anita Ganesan (Bitbucket: [Anita Ganesan](https://bitbucket.org/Anita Ganesan), GitHub: ag12733).
Currently model error is the only hyper-parameter but there is no option to fix it (i.e. set the stepsize for the model error to 0). This is needed to be able to replicate other papers or to simplify the inversions for testing things are working.