Add `loss` selection for fitting(in the scaling step)

scipp / essnmx

Data reduction for NMX at the European Spallation Source

https://scipp.github.io/essnmx/

BSD 3-Clause "New" or "Revised" License

0 stars 1 forks source link

Add `loss` selection for fitting(in the scaling step) #64

Open YooSunYoung opened 3 months ago

YooSunYoung commented 3 months ago

@jokasimr suggested trying different loss functions for fitting. Maybe it'll be convenient if a user can choose loss functions as well.

jokasimr commented 3 months ago

If the data is noisy it makes sense to have something less sensitive to outliers. See the options here.

Imo it's better if we decide what loss is best for the application, based on what is reasonable in the context, otherwise the users might get too many knobs to turn.

YooSunYoung commented 3 months ago

It'll be mainly for IDS users and advanced users though. We can just set a default one for non-expert users.

At least until the workflow is settled. It's still being developed by IDS members.