Null deviance statsmodels GLM python pseudo R2

Question

To compute the pseudo R2 of my GLM model, I need the null deviance.

formula: 1-(deviance/null.deviance)

The model deviance is given in the model.summary() output, but how do I compute/access the null deviance?

My code example is:

import statsmodels.formula.api as smf 
import statsmodels.api as sm

# generate random dataframe
df = pd.DataFrame(np.random.randint(0,100,size=(100, 3)), columns=list('ABC'))

# negative binomial regression 
model = smf.glm(formula = "c ~ a + b", data=df, family=sm.families.NegativeBinomial()).fit()
model.summary()

Answer 1

Actually, figured it out:

model.null_deviance 
model.deviance