Open lbittarello opened 3 years ago
What should the interface look like? fit
is pretty straightforward, but I'm not sure about predict
.
I found a way to phrase the obvious, which worked for me using the formula interface:
formula = "Y~ {np.ones_like(Y)}", fit_intercept=True
The more obvious approach "Y~1" throws errors with "fit_intercept=True" and with "fit_intercept=False"
It'd be nice to be able to fit GLMs without covariates. It's usually trivial, except when offsets are involved...