Phase plot containing unlikely models

Hi,

When I plot results from 'allesfitter' I keep getting poorly-fitting models included in the plot. Both RV and photometry data is being fit together. From my understanding, it is incredibly unlikely to see these models in the plots as they should not be included in the posterior samples. Am I missing a step when plotting in order to make sure I am vetting these models?

I am running 'allesfitter' version 1.2.8 on a target. I am plotting the output from 'allesfitter' from sampling), and even with a low number of 'Nsamples' (e.g. 20 or 200) I find multiple models that should have low probability.

I have tried my own plotting routine as well as the one in the example below, but I get the same result

#------------------------------------------------------------------------------
# Photometry
#------------------------------------------------------------------------------
#::: iterate over all plot styles
for style in ['full']:

    #::: set up the figure
    fig, axes = plt.subplots(2, 1, figsize=(6,6), gridspec_kw={'height_ratios': [3,1]}, sharex=True, Nsamples=200)
    fig.subplots_adjust(hspace=0)

    #::: alles.plot(...) data and model
    alles.plot('Leonardo','b',style,ax=axes[0])
    axes[0].set_title('Leonardo, '+style)

    #::: alles.plot(...) residuals
    alles.plot('Leonardo','b',style+'_residuals',ax=axes[1])
    axes[1].set_title('')

    fig.savefig('Leonardo_'+style+'.pdf', bbox_inches='tight')

github_issue_plot.pdf

MNGuenther / allesfitter

Phase plot containing unlikely models #79