v0.24.4

0.24.4 - 2020-04-13

Bug fixes

• Improved stability of interval censoring in parametric models.
• setting a dataframe in ancillary_df works for interval censoring
• .score works for interval censored models

v0.24.3

0.24.3 - 2020-03-25

New features

• new logx kwarg in plotting curves
• PH models have compute_followup_hazard_ratios for simulating what the hazard ratio would be at previous times. This is useful because the final hazard ratio is some weighted average of these.

Bug fixes

• Fixed error in HTML printer that was hiding concordance index information.

v0.24.2

0.24.2 - 2020-03-15

Bug fixes

• Fixed bug when no covariates were passed into CoxPHFitter. See #975
• Fixed error in StatisticalResult where the test name was not displayed correctly.
• Fixed a keyword bug in plot_covariate_groups for parametric models.

v0.24.1

0.24.1 - 2020-03-05

New features

• Stability improvements for GeneralizedGammaRegressionFitter and CoxPHFitter with spline estimation.

Bug fixes

• Fixed bug with plotting hazards in NelsonAalenFitter.

v0.24.0

0.24.0 - 2020-02-20

This version and future versions of lifelines no longer support py35. Pandas 1.0 is fully supported, along with previous version. Minimum Scipy has been bumped to 1.2.0

New features

• CoxPHFitter and CoxTimeVaryingFitter has support for an elastic net penalty, which includes L1 and L2 regression.
• CoxPHFitter has new baseline survival estimation methods. Specifically, spline now estimates the coefficients and baseline survival using splines. The traditional method, breslow, is still the default however.
• Regression models have a new score method that will score your model against a dataset (ex: a testing or validation dataset). The default is to evaluate the log-likelihood, but also the concordance index can be chose.
• New MixtureCureFitter for quickly creating univariate mixture models.
• Univariate parametric models have a plot_density, density_at_times, and property density_ that computes the probability density function estimates.
• new dataset for interval regression involving C. Botulinum.
• new lifelines.fitters.mixins.ProportionalHazardMixin that implements proportional hazard checks.

API Changes

• Models' prediction method that return a single array now return a Series (use to return a DataFrame). This includes predict_median, predict_percentile, predict_expectation, predict_log_partial_hazard, and possibly others.
• The penalty in Cox models is now scaled by the number of observations. This makes it invariant to changing sample sizes. This change also make the penalty magnitude behave the same as any parametric regression model.
• score_ on models has been renamed concordance_index_
• models' .variance_matrix_ is now a DataFrame.
• CoxTimeVaryingFitter no longer requires an id_col. It's optional, and some checks may be done for integrity if provided.
• Significant changes to utils.k_fold_cross_validation.
• removed automatically adding inf from PiecewiseExponentialRegressionFitter.breakpoints and PiecewiseExponentialFitter.breakpoints
• tie_method was dropped from Cox models (it was always Efron anyways...)
• Mixins are moved to lifelines.fitters.mixins
• find_best_parametric_model evaluation kwarg has been changed to scoring_method.
• removed _score_ and path from Cox model.

Bug fixes

• Fixed show_censors with KaplanMeierFitter.plot_cumulative_density see issue #940.
• Fixed error in "BIC" code path in find_best_parametric_model
• Fixed a bug where left censoring in AFT models was not converging well
• Cox models now incorporate any penalizers in their log_likelihood_

v0.23.9

0.23.9 - 2020-01-28

Bug fixes

• fixed important error when a parametric regression model would not assign the correct labels to fitted
  parameters' variances. See more here: #931. Users of GeneralizedGammaRegressionFitter and any custom regression models should update their code as soon as possible.

v0.23.8

Bug fixes

• fixed important error when a parametric regression model would not assign the correct labels to fitted parameters. See more here: #931. Users of GeneralizedGammaRegressionFitter and any custom regression models should update their code as soon as possible.

v0.23.7

Bug fixes for py3.5. This will be the last version of lifelines that supports Python 3.5.

v0.23.6

0.23.6 - 2020-01-07

New features

• New univariate model, SplineFitter, that uses cubic splines to model the cumulative hazard.
• To aid users with selecting the best parametric model, there is a new lifelines.utils.find_best_parametric_model function that will iterate through the models and return the model with the lowest AIC (by default).
• custom parametric regression models can now do left and interval censoring.

v0.23.5

0.23.5 - 2020-01-05

New features

• New predict_hazard for parametric regression models.
• New lymph node cancer dataset, originally from H.F. for the German Breast Cancer Study Group (GBSG) (1994)

Bug fixes

• fixes error thrown when converge of regression models fails.
• kwargs is now used in plot_covariate_groups
• fixed bug where large exponential numbers in print_summary were not being suppressed correctly.