You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Describe the bug
Calling plot_2d without chosing which params to plot generically seems to error as it includes logL, nlive and logLbirth, which typically don't play nicely with 2d KDEs
To Reproduce
Using twin_gaussian example from PCLite repo chains (using anesthetic 2.3.0 installed via pip)
importanestheticasnsfrommatplotlibimportpyplotaspltsamples=ns.read_chains("chains/twin_gaussian")
#fine because you force it to be scatterssamples.plot_2d(kinds=dict(upper="scatter_2d",lower='scatter_2d'))
#errors due to triangulation errorssamples.plot_2d()
Expected behavior
Default behavior should be all params but not nlive, logl and logl_birth.
Additional context
Might be related to the previous change wrt integer axis names
The text was updated successfully, but these errors were encountered:
Very much related to #324. Can we move the discussion to there, in particular whether we think that default behaviour should be to avoid the nlive logl and logl_birth columns.
Describe the bug
Calling plot_2d without chosing which params to plot generically seems to error as it includes logL, nlive and logLbirth, which typically don't play nicely with 2d KDEs
To Reproduce
Using twin_gaussian example from PCLite repo chains (using anesthetic 2.3.0 installed via pip)
Expected behavior
Default behavior should be all params but not nlive, logl and logl_birth.
Additional context
Might be related to the previous change wrt integer axis names
The text was updated successfully, but these errors were encountered: