Glum

------------------

**New features:**

- Added :meth:`aic`, :meth:`aicc` and :meth:`bic` attributes to the :class:`~glum.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.
- :meth:`~glum.GeneralizedLinearRegressor.std_errors` and :meth:`~glum.GeneralizedLinearRegressor.covariance_matrix`
of :class:`~glum.GeneralizedLinearRegressor` now accept data frames with categorical data.

**Bug fixes:**

- The :meth:`score` method of :class:`~glum.GeneralizedLinearRegressor` and :class:`~glum.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.

------------------

**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.

------------------

**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.

------------------

**Bug fix:**

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

------------------

**Breaking changes:**

- Renamed the package to ``glum``!! Hurray! Celebration.
- :class:`~glum.GeneralizedLinearRegressor` and :class:`~glum.GeneralizedLinearRegressorCV` lose the ``fit_dispersion`` parameter.
Please use the :meth:`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 :class:`~glum.GeneralizedLinearRegressor` and :class:`~glum.GeneralizedLinearRegressorCV` that should have been private have had an underscore prefixed on their names: :meth:`tear_down_from_fit`, :meth:`_set_up_for_fit`, :meth:`_set_up_and_check_fit_args`, :meth:`_get_start_coef`, :meth:`_solve` and :meth:`_solve_regularization_path`.
- :meth:`glum.GeneralizedLinearRegressor.report_diagnostics` and :meth:`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.
- :class:`ExponentialDispersionModel` gains a :meth:`dispersion` method.
- :class:`BinomialDistribution` and :class:`TweedieDistribution` gain a :meth:`log_likelihood` method.
- The :meth:`fit` method of :class:`~glum.GeneralizedLinearRegressor` and :class:`~glum.GeneralizedLinearRegressorCV`
now saves the column types of pandas data frames.
- :class:`~glum.GeneralizedLinearRegressor` and :class:`~glum.GeneralizedLinearRegressorCV` gain two properties: ``family_instance`` and ``link_instance``.
- :meth:`~glum.GeneralizedLinearRegressor.std_errors` and :meth:`~glum.GeneralizedLinearRegressor.covariance_matrix` have been added and support non-robust, robust (HC-1), and clustered
covariance matrices.
- :class:`~glum.GeneralizedLinearRegressor` and :class:`~glum.GeneralizedLinearRegressorCV` now accept ``family='gaussian'`` as an alternative to ``family='normal'``.

**Bug fix:**

- The :meth:`score` method of :class:`~glum.GeneralizedLinearRegressor` and :class:`~glum.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 <https://github.com/Quantco/glum/pull/376>`_.
- ``glm_benchmarks_analyze`` now includes ``offset`` in the index. See `here <https://github.com/Quantco/glum/issues/346>`_.
- ``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 <https://github.com/Quantco/glum/pull/424>`_.

------------------

**Bug fix:**

* Have the :meth:`linear_predictor` and :meth:`predict` methods of :class:`~glum.GeneralizedLinearRegressor` and :class:`~glum.GeneralizedLinearRegressorCV`
honor the offset when ``alpha`` is ``None``.