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
Currently, if we have a distribution and want to find out its parameters programmatically we can do:
dist.params, a tuple with the values
dist.param_names a tuple with the names
For distributions with more than one parameterization, these two functions will return the values and names of the default/canonical parameterization. Additionally, we can obtain the values of the parameters by name, even non-canonical names, for example, this works pz.Normal(mu=0, sigma=0.5).tau .
I think we should have a params_dict attribute or method that returns a dictionary with names:values. Not sure which one is better
For distributions with more than one parameterization, what should we return?
attribute. Just the canonical dict, a tuple of dicts with all parameterizations?
method we can pass an argument to ask for different parameterizations like "canonical" vs "alternative". For distributions with more than 2 parametrization we need some other keyword... not sure what would be a nice choice.
The text was updated successfully, but these errors were encountered:
Currently, if we have a distribution and want to find out its parameters programmatically we can do:
For distributions with more than one parameterization, these two functions will return the values and names of the default/canonical parameterization. Additionally, we can obtain the values of the parameters by name, even non-canonical names, for example, this works
pz.Normal(mu=0, sigma=0.5).tau
.I think we should have a
params_dict
attribute or method that returns a dictionary with names:values. Not sure which one is betterFor distributions with more than one parameterization, what should we return?
The text was updated successfully, but these errors were encountered: