Skip to content

Data detector level

Oleksandr Zenaiev requested to merge data-detector-level into master

Implemented matrix times vector multiplication in theory expression

Possible things to discuss:

  1. now KFactor, KMatrix and ApplGrid can legally have the number of values which is different from the number of data points (for KMatrix it is kind of obvious). This allows them to be further convoluted (multiplied) and change number of values. But no check is done in theory expression -> it is responsibility of the user to make sure that the final expression matches the number of data points. Some better interface is desirable, but perhaps not with the present implementation of theory expression

  2. perhaps it was a bug in prep_corr.f: one data structure was declared with N^4 size, while only N^2 seems to be needed. There is some mess with NCorrMax: is it the max. number of entries in the correlation file, or the max. number of correlated bins? To resolve it quickly, alternative NCorrMaxBins was introduced

Merge request reports