Comparing fitting with EasyCore and scipp

[henrikjacobsenfys]

Disclaimer: I am a novice with both packages, so it could very well be that I am not making proper use of EasyCore.

I use scipp to load and handle the data. This is convenient for several reasons. One is that it automatically handles units, and binning, combining, adding and subtracting data is easy. Fitting with scipp is a natural extension of this. It keeps the data in neat DataArrays, which makes visualisation easy. It also handles the units, and since it's based on scipy, it is easy to use functions from e.g. QENSlibrary. Extensions to multi-dimensional data (e.g. fitting multiple data sets to a single model) seem straightforward.

In EasyCore, it seems functions take only x as input parameter, and all variables seem to be global. If this is true, then I cannot easily use e.g. QENSlibrary functions.

It seems that I have to extract the values from my scipp DataArrays to pass them as numpy arrays to the EasyCore fitter. This loses the advantages of using scipp to handle the data in the first place, as I suddenly only have data as x, y and e. I do not know how easy it would be to fit multiple data sets at once.