I recently use montepython to constraint parameters from gravity model (parameters_smg). There's no problem with the hiclass since I'm running alpha attractor, a model that has been written on the code hi_class_public/gravity_models.
and here's my problem:
I have no clue to set my uninformative prior to random number from uniform distribution. I tried to set 1-sigma to 0.01 (small number) and min-max to my range, ex. parameters_smg__2 [ 2, 0, 3, 0.01, 1, 'cosmo' ]. Then I just read #73 said "set 1-sigma to 10% of the prior range". But I don't get it, why could this be the way to get uniform distribution for a prior?
I have parameter 'c' (parameters_smg__4 in the code) which should be determined by closure relation, while it must set to cosmo, so I make it data.parameters['parameters_smg__4'=[0, None, None, 0, 1, 'cosmo'] and add data.cosmo_arguments['tuning_index_smg']=3. But it turns to error message "Cosmological Module Error. Error in Class: Class did not read input parameter: tuning_index_smg" and the program runs nothing.
Then I check hiclass. Seems that 'c' is automatically derived from closure relation even without put data.cosmo_arguments['tuning_index_smg']=3, I see the value would be printed on terminal while running hiclass.
But If I just write data.parameters['parameters_smg__4'=[0, None, None, 0, 1, 'cosmo'], 'c' would not be written in the result files (.paramnames and .txt). However I need to show this 'c' in my posterior. Does anyone know how to solve this problem?
Hello,
I recently use montepython to constraint parameters from gravity model (parameters_smg). There's no problem with the hiclass since I'm running alpha attractor, a model that has been written on the code hi_class_public/gravity_models.
and here's my problem:
I have no clue to set my uninformative prior to random number from uniform distribution. I tried to set 1-sigma to 0.01 (small number) and min-max to my range, ex.
parameters_smg__2 [ 2, 0, 3, 0.01, 1, 'cosmo' ]. Then I just read #73 said "set 1-sigma to 10% of the prior range". But I don't get it, why could this be the way to get uniform distribution for a prior?I have parameter 'c' (
parameters_smg__4in the code) which should be determined by closure relation, while it must set tocosmo, so I make itdata.parameters['parameters_smg__4'=[0, None, None, 0, 1, 'cosmo']and adddata.cosmo_arguments['tuning_index_smg']=3. But it turns to error message "Cosmological Module Error. Error in Class: Class did not read input parameter: tuning_index_smg" and the program runs nothing.Then I check hiclass. Seems that 'c' is automatically derived from closure relation even without put
data.cosmo_arguments['tuning_index_smg']=3, I see the value would be printed on terminal while running hiclass.But If I just write
data.parameters['parameters_smg__4'=[0, None, None, 0, 1, 'cosmo'], 'c' would not be written in the result files (.paramnames and .txt). However I need to show this 'c' in my posterior. Does anyone know how to solve this problem?Thank you 🙏🏿