Software Open Access
Cameron Davidson-Pilon; Jonas Kalderstam; Noah Jacobson; Sean Reed; Ben Kuhn; Paul Zivich; Mike Williamson; AbdealiJK; Deepyaman Datta; Andrew Fiore-Gartland; Alex Parij; Daniel WIlson; Gabriel; Luis Moneda; Arturo Moncada-Torres; Kyle Stark; Harsh Gadgil; Jona; JoseLlanes; Karthikeyan Singaravelan; Lilian Besson; Miguel Sancho Peña; Steven Anton; Andreas Klintberg; GrowthJeff; Javad Noorbakhsh; Matthew Begun; Ravin Kumar; Sean Hussey; Skipper Seabold
0.26.0 - 2021-05-26 New features
.BIC_is now present on fitted models.
CoxPHFitterwith spline baseline can accept pre-computed knot locations.
predictalways predicts the survival function (as does every other model),
confidence_interval_is always the CI for the survival function (as does every other model), and so on. In summary: the API for estimates doesn't change depending on what your censoring your dataset is.
find_best_parametric_modelwhere the wrong BIC value was being computed.