summary#

SuperGLM.summary(
alpha: float = 0.05,
detail: str = 'compact',
level_display: str = 'expanded',
)#

Rich model summary with coefficient table (statsmodels-style).

level_display="expanded" shows exact original categorical levels. Use "grouped" for one row per fitted group plus a membership legend.

Coefficient rows use fitted-family covariance: known-scale families retain phi=1 and estimated-scale families use fitted phi. Pearson quasi-likelihood diagnostics are available separately from metrics() through coefficient_se, intercept_se, and feature_se. The dict-like standard_errors payload labels its coefficient-SE scale explicitly as "fitted-family".