ProbabilityDict class

This class is similar to the SamplesDict class but instead of storing a collection of posterior samples for each parameter, it stores the analytic PDF for each parameter. The analytic PDF is stored as a DiscretePDF object.

class pesummary.utils.probability_dict.ProbabilityDict(*args, logger_warn='warn', autoscale=False)[source]

Class to store integers with discrete probabilities for multiple parameters.

Parameters

args (dict) – dictionary containing the discrete probabilities for each parameter. Key should be the parameter name and value should be a 2d array of length 2. First element integers and second element probabilities corresponding to integers. See pesummary.utils.pdf.DiscretePDF for more details

rvs:

randomly draw samples for each distribution in ProbabilityDict

Examples

>>> from pesummary.utils.probability_dict import ProbabilityDict
>>> numbers = [1,2,3,4]
>>> probabilities = [0.1, 0.2, 0.3, 0.4]
>>> pdf = ProbabilityDict({"parameter": [numbers, probabilities]})
>>> print(type(pdf["parameter"]))
<class 'pesummary.utils.pdf.DiscretePDF'>
>>> print(pdf["parameter"].probs)
[0.1, 0.2, 0.3, 0.4]