Python-erros when clustering standard errors in paneldata with statsmodels

Question

I am working with Paneldata with earnings as the endogenous variable, and sex, age, education and dummies for states and years as the exogenous. My main problem seems to be