Changelog

3.4.1 - 2026-05-06

Bug fix:

• Fixed monotonic_constraints raising ValueError with drop_first=True on categorical factors.

• Fixed pickle compatibility for models saved before monotonic_constraints was added.

3.4.0 - 2026-04-27

New feature:

• Added monotonic_constraints parameter to GeneralizedLinearRegressor and GeneralizedLinearRegressorCV for enforcing monotonic effects on splines, ordered categoricals, and interactions via the formula interface. Uses a new irls-ls-monotonic solver (5-20× faster than trust-constr on insurance benchmark).

3.3.0 - 2026-04-14

New feature:

• GeneralizedLinearRegressorCV now exposes train_deviance_path_, an array of shape (n_folds, n_l1_ratios, n_alphas) with the training-set deviance.

Other changes:

• Replaced the hess="2-point" finite-difference Hessian in the trust-constr solver with SR1() (quasi-Newton), significantly speeding up convergence when the number of features is large relative to the number of rows.

3.2.3 - 2026-03-18

Bug fix:

• Fixed incorrect call in log_likelihood(). The previous implementation always returned NaN.

3.2.2 - 2026-03-17

Bug fix:

• Fixed incorrect formula in inverse_derivative2(). This affected observed information matrix computation and robust/clustered standard errors for models using the complementary log-log link.

3.2.1 - 2026-03-16

Bug fix:

• Fixed an error when predicting at a specific alpha with categorical features.

Other changes:

• Downgraded log messages in align_df_categories and add_missing_categories from INFO to DEBUG, and deduplicated them so they are emitted only once per column per fitted model.

3.2.0 - 2026-03-10

New features:

• Add solver="closed-form" for Gaussian identity-link models, using an analytical normal-equations solution for ridge/OLS, auto-selecting it under solver="auto" for unconstrained no-L1 cases, and falling back to least-squares for singular or ill-conditioned systems.

• GeneralizedLinearRegressor and GeneralizedLinearRegressorCV now accept Polars DataFrames as input, in addition to pandas DataFrames and numpy arrays.

Bug fixes:

• Fixed predict(X, alpha_index=...) and predict(X, alpha=...) on GeneralizedLinearRegressorCV, which previously raised an error. The CV estimator now refits on the full data over the entire alpha path for the best l1_ratio_.

• Fixed alpha path computation in GeneralizedLinearRegressorCV: the alpha grid is now computed from the properly standardized feature matrix and per-feature P1 penalties (matching the base class), and accounts for offset when present.

3.1.3 - 2025-02-18

Bug fixes:

• Fixed deviance_path_ in GeneralizedLinearRegressorCV being scaled down by a factor of n_folds because test fold weights were not normalized to sum to 1.

• Fixed NegativeBinomialDistribution theta setter rejecting np.number types, causing dist.theta = dist.theta to raise a TypeError.

Other changes:

• We disabled fast math to avoid invalid results (e.g., when dividing by zero).

• The default alpha_max for the regularization path in the pure ridge case (l1_ratio=0) is now computed from the null model gradient, matching glmnet’s behavior. Previously it was hardcoded to 10.

• When min_alpha_ratio is not set and n_samples < n_features, the default is now 1e-2 instead of 1e-6, matching glmnet’s convention.

3.1.2 - 2025-01-30

Other changes:

• Linux-aarch64 wheels are now uploaded to PyPI.

3.1.1 - 2025-01-13

Bug fixes:

• Fixed a bug where _rowwise_gradient_hessian() and _eta_mu_deviance() would call functions with wrong arguments in the p = 3 case.

• Fixed glum.InverseGaussianDistribution not using the optimized gradient, Hessian and deviance implementations, as well as those derivatives having the wrong sign.

Other changes:

• Build and test with Python 3.13 in CI.

3.1.0 - 2024-11-11

New features:

• New argument max_inner_iter for classes GeneralizedLinearRegressor and GeneralizedLinearRegressorCV to control the maximum number of iterations of the inner solver in the IRLS-CD algorithm.

• New fitted attributes col_means_ and col_stds_ for classes GeneralizedLinearRegressor and GeneralizedLinearRegressorCV.

• GeneralizedLinearRegressor now prints more informative logs when fitting with alpha_search=True and verbose=True.

Bug fixes:

• Fixed a bug where glum.GeneralizedLinearRegressor.fit() would raise a dtype mismatch error if fit with alpha_search=True.

• Use data type (float64 or float32) dependent precision in solvers.

**Other changes:

• Lower absolute tolerance when matching the alpha argument in glum.GeneralizedLinearRegressor.predict() to the alphas used in training when alpha_search=True.

3.0.2 - 2024-06-25

Bug fix:

• Fixed wald_test() when using terms and no intercept.

Other changes:

• Moved the development infrastructure to pixi.

• Moved the linting and formatting to ruff.

• Removed libblas MKL from the development environment.

• Replaced deprecated ‘oldest-supported-numpy’ dependency with ‘numpy’ to support 2.0 release.

3.0.1 - 2024-05-23

Bug fix:

• We now support scikit-learn 1.5.

3.0.0 - 2024-04-27

Breaking changes:

• All arguments to GeneralizedLinearRegressorBase, GeneralizedLinearRegressor and GeneralizedLinearRegressorCV are now keyword-only.

• All arguments to public methods of GeneralizedLinearRegressorBase, GeneralizedLinearRegressor or GeneralizedLinearRegressorCV except X, y, sample_weight and offset are now keyword-only.

• GeneralizedLinearRegressor’s default value for alpha is now 0, i.e. no regularization.

• GammaDistribution, InverseGaussianDistribution, NormalDistribution and PoissonDistribution no longer inherit from TweedieDistribution.

• The power parameter of TweedieLink has been renamed from p to power, in line with TweedieDistribution.

• TweedieLink no longer instantiates IdentityLink or LogLink for power=0 and power=1, respectively. On the other hand, TweedieLink is now compatible with power=0 and power=1.

New features:

• Added a formula interface for specifying models.

• Improved feature name handling. Feature names are now created for non-pandas input matrices too. Furthermore, the format of categorical features can be specified by the user.

• Term names are now stored in the model’s attributes. This is useful for categorical features, where they refer to the whole variable, not just single levels.

• Added more options for treating missing values in categorical columns. They can either raise a ValueError ("fail"), be treated as all-zero indicators ("zero") or represented as a new category ("convert").

• meth:GeneralizedLinearRegressor.wald_test can now perform tests based on a formula string and term names.

• InverseGaussianDistribution gains a log_likelihood() method.

2.7.0 - 2024-02-19

Bug fix:

• Added cython compiler directive legacy_implicit_noexcept = True to fix performance regression with cython 3.

Other changes:

• Require Python>=3.9 in line with NEP 29 <https://numpy.org/neps/nep-0029-deprecation_policy.html#support-table>.

• Build and test with Python 3.12 in CI.

• Added line search stopping criterion for tiny loss improvements based on gradient information.

• Added warnings about breaking changes in future versions.

2.6.0 - 2023-09-05

New features:

• Added the complementary log-log (cloglog) link function.

• Added the option to store the covariance matrix after estimating it. In this case, the covariance matrix does not have to be recomputed when calling inference methods.

• Added methods for performing Wald tests based on a restriction matrix, feature names or term names.

• Added a method for creating a coefficient table with confidence intervals and p-values.

Bug fix:

• Fixed covariance_matrix() mutating feature names when called with a data frame. See here.

Other changes:

• When computing the covariance matrix, check whether the design matrix is ill-conditioned for all types of input. Furthermore, do it in a more efficient way.

• Pin tabmat<4.0.0 (the new release will bring breaking changes).

2.5.2 - 2023-06-02

Bug fix

• Fix the glm_benchmarks_analyze command line tool. See here.

• Fixed a bug in GeneralizedLinearRegressor when fit on a data set with a constant column and warm_start=True. See here.

Other changes:

• Remove dev dependency on dask_ml.

• We now pin llvm-openmp=11 when creating the wheel for macOS in line with what scikit-learn does.

2.5.1 - 2023-05-19

Other changes:

• Better error message when the number of input features is different between the fit and predict methods.

Bug fix:

• We fixed a bug in the computation of log_likelihood(). Previously, this method just returned None.

2.5.0 - 2023-04-28

New feature:

• Added Negative Binomial distribution by setting the 'family' parameter of GeneralizedLinearRegressor and GeneralizedLinearRegressorCV to 'negative.binomial'.

2.4.1 - 2023-03-14

Bug fixes:

• Fixed an issue with _score_matrix() which failed when called with a tabmat matrix input.

Other changes:

• Removes unused scikit-learn cython imports.

2.4.0 - 2023-01-31

Other changes:

• LogitLink has been made public.

• Apple Silicon wheels are now uploaded to PyPI.

2.3.0 - 2023-01-06

Bug fixes:

• A data frame with dense and sparse columns was transformed to a dense matrix instead of a split matrix by _set_up_and_check_fit_args(). Fixed by calling tabmat.from_pandas on any data frame.

New features:

• The following classes and functions have been made public: BinomialDistribution, ExponentialDispersionModel, GammaDistribution, GeneralizedHyperbolicSecant, InverseGaussianDistribution, NormalDistribution, PoissonDistribution, IdentityLink, Link, LogLink, TweedieLink, get_family() and get_link().

• The distribution and link classes now feature a more lenient equality check instead of the default identity check, so that, e.g., TweedieDistribution(1) == TweedieDistribution(1) now returns True.

2.2.1 - 2022-11-25

Other changes:

• Fixing pypi upload issue. Version 2.2.0 will not be available through the standard distribution channels.

2.2.0 - 2022-11-25

New features:

• Add an argument to GeneralizedLinearRegressorBase to drop the first category in a Categorical column using [implementation in tabmat](https://github.com/Quantco/tabmat/pull/168)

• One may now request the Tweedie loss by setting the 'family' parameter of GeneralizedLinearRegressor and GeneralizedLinearRegressorCV to 'tweedie'.

Bug fixes:

• Setting bounds for constant columns was not working (bounds were internally modified to 0). A similar issue was preventing inequalities from working with constant columns. This is now fixed.

Other changes:

• No more builds for 32-bit systems with python >= 3.8. This is due to scipy not supporting it anymore.

2.1.2 - 2022-07-01

Other changes:

• Next attempt to build wheel for PyPI without --march=native.

2.1.1 - 2022-07-01

Other changes:

• We are now building the wheel for PyPI without --march=native to make it more portable across architectures.

2.1.0 - 2022-06-27

New features:

• Added aic(), aicc() and bic() attributes to the GeneralizedLinearRegressor. These attributes provide the information criteria based on the training data and the effective degrees of freedom of the maximum likelihood estimate for the model’s parameters.

• std_errors() and covariance_matrix() of GeneralizedLinearRegressor now accept data frames with categorical data.

Bug fixes:

• The score() method of GeneralizedLinearRegressor and GeneralizedLinearRegressorCV now accepts offsets.

• Fixed the calculation of the information matrix for the Binomial distribution with logit link, which affected nonrobust standard errors.

Other:

• The CI now runs daily unit tests against the nightly builds of numpy, pandas and scikit-learn.

• The minimally required version of tabmat is now 3.1.0.

2.0.3 - 2021-11-05

Other:

• We are now specifying the run time dependencies in setup.py, so that missing dependencies are automatically installed from PyPI when installing glum via pip.

2.0.2 - 2021-11-03

Bug fix:

• Fixed the sign of the log likelihood of the Gaussian distribution (not used for fitting coefficients).

• Fixed the wide benchmarks which had duplicated columns (categorical and numerical).

Other:

• The CI now builds the wheels and upload to pypi with every new release.

• Renamed functions checking for qc.matrix compliance to refer to tabmat.

2.0.1 - 2021-10-11

Bug fix:

• Fixed pyproject.toml. We now support installing through pip and pep517.

2.0.0 - 2021-10-08

Breaking changes:

• Renamed the package to glum!! Hurray! Celebration.

• GeneralizedLinearRegressor and GeneralizedLinearRegressorCV lose the fit_dispersion parameter. Please use the dispersion() method of the appropriate family instance instead.

• All functions now use sample_weight as a keyword instead of weights, in line with scikit-learn.

• All functions now use dispersion as a keyword instead of phi.

• Several methods GeneralizedLinearRegressor and GeneralizedLinearRegressorCV that should have been private have had an underscore prefixed on their names: tear_down_from_fit(), _set_up_for_fit(), _set_up_and_check_fit_args(), _get_start_coef(), _solve() and _solve_regularization_path().

• glum.GeneralizedLinearRegressor.report_diagnostics() and glum.GeneralizedLinearRegressor.get_formatted_diagnostics() are now public.

New features:

• P1 and P2 now accepts 1d array with the same number of elements as the unexpanded design matrix. In this case, the penalty associated with a categorical feature will be expanded to as many elements as there are levels, all with the same value.

• ExponentialDispersionModel gains a dispersion() method.

• BinomialDistribution and TweedieDistribution gain a log_likelihood() method.

• The fit() method of GeneralizedLinearRegressor and GeneralizedLinearRegressorCV now saves the column types of pandas data frames.

• GeneralizedLinearRegressor and GeneralizedLinearRegressorCV gain two properties: family_instance and link_instance.

• std_errors() and covariance_matrix() have been added and support non-robust, robust (HC-1), and clustered covariance matrices.

• GeneralizedLinearRegressor and GeneralizedLinearRegressorCV now accept family='gaussian' as an alternative to family='normal'.

Bug fix:

• The score() method of GeneralizedLinearRegressor and GeneralizedLinearRegressorCV now accepts data frames.

• Upgraded the code to use tabmat 3.0.0.

Other:

• A major overhaul of the documentation. Everything is better!

• The methods of the link classes will now return scalars when given scalar inputs. Under certain circumstances, they’d return zero-dimensional arrays.

• There is a new benchmark available glm_benchmarks_run based on the Boston housing dataset. See here.

• glm_benchmarks_analyze now includes offset in the index. See here.

• glmnet_python was removed from the benchmarks suite.

• The innermost coordinate descent was optimized. This speeds up coordinate descent dominated problems like LASSO by about 1.5-2x. See here.

1.5.1 - 2021-07-22

Bug fix:

• Have the linear_predictor() and predict() methods of GeneralizedLinearRegressor and GeneralizedLinearRegressorCV honor the offset when alpha is None.

1.5.0 - 2021-07-15

New features:

• The linear_predictor() and predict() methods of GeneralizedLinearRegressor and GeneralizedLinearRegressorCV gain an alpha parameter (in complement to alpha_index). Moreover, they are now able to predict for multiple penalties.

Other:

• Methods of Link now consistently return NumPy arrays, whereas they used to preserve pandas series in special cases.

• Don’t list sparse_dot_mkl as a runtime requirement from the conda recipe.

• The minimal numpy pin should be dependent on the numpy version in host and not fixed to 1.16.

1.4.3 - 2021-06-25

Bug fix:

• copy_X = False will now raise a value error when X has dtype int32 or int64. Previously, it would only raise for dtype int64.

1.4.2 - 2021-06-15

Tutorials and documentation improvements:

• Adding tutorials to the documentation.

• Additional documentation improvements.

Bug fix:

• Verbose progress bar now working again.

Other:

• Small improvement in documentation for the alpha_index argument to predict().

• Pinned pre-commit hooks versions.

1.4.1 - 2021-05-01

We now have Windows builds!

1.4.0 - 2021-04-13

Deprecations:

• Fusing the alpha and alphas arguments for GeneralizedLinearRegressor. alpha now also accepts array like inputs. alphas is now deprecated but can still be used for backward compatibility. The alphas argument will be removed with the next major version.

Bug fix:

• We removed entry points to functions in glum_benchmarks from the conda package.

1.3.1 - 2021-04-12

Bug fix:

• glum._distribution.unit_variance_derivative() is evaluating a proper numexpr expression again (regression in 1.3.0).

1.3.0 - 2021-04-12

New features:

• We added a new solver based on scipy.optimize.minimize(method='trust-constr').

• We added support for linear inequality constraints of type A_ineq.dot(coef_) <= b_ineq.

1.2.0 - 2021-02-04

We removed glum_benchmarks from the conda package.

1.1.1 - 2021-01-11

Maintenance release to get a fresh build for OSX.

1.1.0 - 2020-11-23

New feature:

• Direct support for pandas categorical types in fit and predict. These will be converted into a CategoricalMatrix.

1.0.1 - 2020-11-12

This is a maintenance release to be compatible with tabmat>=1.0.0.

1.0.0 - 2020-11-11

Other:

• Renamed alpha_level attribute of GeneralizedLinearRegressor and GeneralizedLinearRegressorCV to alpha_index.

• Clarified behavior of scale_predictors.

0.0.15 - 2020-11-11

Other:

• Pin tabmat<1.0.0 as we are expecting a breaking change with version 1.0.0.

0.0.14 - 2020-08-06

New features:

• Add Tweedie Link.

• Allow infinite bounds.

Bug fixes:

• Unstandardize regularization path.

• No copying in predict.

Other:

• Various memory and performance improvements.

• Update pre-commit hooks.

0.0.13 - 2020-07-23

See git history.

0.0.12 - 2020-07-07

See git history.

0.0.11 - 2020-07-02

See git history.

0.0.10 - 2020-06-30

See git history.

0.0.9 - 2020-06-26

See git history.

0.0.8 - 2020-06-24

See git history.

0.0.7 - 2020-06-17

See git history.

0.0.6 - 2020-06-16

See git history.

0.0.5 - 2020-06-10

See git history.

0.0.4 - 2020-06-08

See git history.

0.0.3 - 2020-06-08

See git history.