Change the definition of the DL2 parameter uncertainty

[YoshikiOhtani]

It is still a draft pull request related to the issue #69.

[YoshikiOhtani]

So now I opened this pull request for reviews. In short summary, it includes:

• the calculation of the reco energy uncertainty with log10(Ereco) instead of Ereco
• the calculation of the gammaness uncertainty with binomial variance
• an option to select the type of weights, "none" None, "uncert" and "intensity" (default: "uncert" None)

I assigned @jsitarek for a reviewer and so it would be appreciated if you could give me comments or approval if it is fine. Thanks!

[YoshikiOhtani]

Sorry for pushing again some commits, but I have also updated the scripts of creating the IRFs and DL3 files so that one can select the type of DL2 weight. Also, I changed the default weight to None (means weight = 1) for a safe side.

[jsitarek]

Hi @YoshikiOhtani
Thanks for the change.
I suggested to rename one of the variables
And I also have a side comment concerning the lat/lon averaging.Other than this it is fine with it.

[YoshikiOhtani]

Hi @jsitarek, thank you very much for your reviews, I pushed the change of the variable names and also the transformation of the mean Alt/Az to the RA/Dec coordinate instead of computing the mean RA/Dec coordinate. Would be great if you could give an approval in case the changes are fine for you. Thanks!

[YoshikiOhtani]

I'm sorry but I mistakenly/accidentally merged this branch to the master... please let me know if you find some problems...!