v0.3.0a9

Bug fix!

Previously there was an error when calculating the uncertainties on the binned datapoints in WiseDataByVisit!
The led to overestimated errors by a factor of $\sqrt{N_{singleexposure}}$

Error factor

WiseBigdataDESYCluster can now make plots of the $\ chi^2$ distribution of the lightcurves compared to the median of each lightcurve. If the sample is dominated by non-varying lightcurves this is expected to follow an F-Distribution (not a simple $\chi^2$ because the data in the bins are sometimes below 10!). Doing this analysis I realised that the errors were underestimated. The errors are now multiplied by the corresponding factor (depending on the band).

What's Changed

• use correct weights when binning lightcurve by @JannisNe in #52
• Make plots of chi2 distributions of light curves by @JannisNe in #50
• Removing outliers in bins during binning by @JannisNe in #53
• Better Chi2 plots by @JannisNe in #56
• Bump virtualenv from 20.16.6 to 20.17.0 by @dependabot in #54
• Bugfix Error calculation by @JannisNe in #55
• Error factor by @JannisNe in #57

Full Changelog: v0.3.0a8...v0.3.0a9