search

brinckmann / montepython_public

Public repository for the Monte Python Code
MIT License
93 stars 77 forks source link

conflict between Planck and Pantheon #293

Closed ming-jian closed 1 year ago

ming-jian commented 2 years ago

hi, everyone I use the montepython to constrain the CLP model.

in base2018TTTEEE.param: data.experiments=['Planck_highl_TTTEEE', 'Planck_lowl_EE', 'bao_boss_dr12','bao_smallz_2014'] data.over_sampling=[1, 5] ...... data.parameters['H0'] = [0, None, None, 0, 1, 'derived'] .... It works well.

in Pantheon.param data.experiments=['Pantheon'] ...... data.parameters['H0'] = [0, None, None, 0, 1, 'cosmo'] data.parameters['M'] = [ - 19.02, None, None, 0.004, 1, 'nuisance'] .... It also works well.

However, it cannot work in their combination. in base2018TTTEEE.param: data.experiments=['Planck_highl_TTTEEE', 'Planck_lowl_EE', 'bao_boss_dr12','bao_smallz_2014','Pantheon'] data.over_sampling=[1, 5] ...... data.parameters['H0'] = [0, None, None, 0, 1, 'derived'] data.parameters['M'] = [ - 19.02, None, None, 0.004, 1, 'nuisance'] ....

It shows: The initialization failed.

I donot know how to deal with it.

HoisW commented 1 year ago

set data.over_sampling=[1, 5,5]

brinckmann commented 1 year ago

You have three sets of sampled parameters, cosmological parameters, Planck nuisnace parameters, and Pantheon nuisance parameters, so your over sampling field needs three entries as HoisW says, e.g. data.over_sampling=[1, 5, 5]

Best, Thejs

ming-jian commented 1 year ago

set data.over_sampling=[1, 5,5]

Thanks very much Prof. HoisW and Prof. Thejs ! It is ok !