Cure Models not working with the given example

[andreagiussani]

Hello,

I am having issues with one of the custom regression models' example - the cure model one. I have installed the latest version of lifelines, and the snippet is the following one.

from lifelines.fitters import ParametricRegressionFitter
from autograd.scipy.special import expit
from lifelines.datasets import load_rossi
class CureModel(ParametricRegressionFitter):
    _scipy_fit_method = "SLSQP"
    _scipy_fit_options = {"ftol": 1e-10, "maxiter": 200}

    _fitted_parameter_names = ["lambda_", "beta_", "rho_"]

    def _cumulative_hazard(self, params, T, Xs):
        c = expit(np.dot(Xs["beta_"], params["beta_"]))

        lambda_ = np.exp(np.dot(Xs["lambda_"], params["lambda_"]))
        rho_ = np.exp(np.dot(Xs["rho_"], params["rho_"]))
        sf = np.exp(-(T / lambda_) ** rho_)

        return -np.log((1 - c) + c * sf)


cm = CureModel(penalizer=0.0)

rossi = load_rossi()

covariates = {"lambda_": rossi.columns.difference(['week', 'arrest']), "rho_": "1", "beta_": 'fin + 1'}

cm.fit(rossi, "week", event_col="arrest", regressors=covariates, timeline=np.arange(250))
cm.print_summary(2)

During the fitting phase, I get the following error: TypeError: ufunc 'expit' not supported for the input types, and the inputs could not be safely coerced to any supported types according to the casting rule ''safe''. I cannot understand why, since I pass correctly the data inside the fit call.

Any help is appreciated.

Thanks a lot

[CamDavidsonPilon]

Hi @andreagiussani, it's a non-obvious problem: you need to add this to the imports:

from autograd import numpy as np

[andreagiussani]

it works - thanks a lot