pyrtid.inverse.get_theoretical_noise_level#

pyrtid.inverse.get_theoretical_noise_level(observables: Sequence[Observable], n_std: float = 5) float[source]#

Get the theoretical noise level in the solution.

If the obersations uncertaintes are well mastered, it can be used as a threshold for the loss function ad defined by Gao et al. 2006.

It is propotionnal to the number of observations. And assume that the observation noise is well known.