Model-diagnostics

What's Changed
* ENH change default bin_method to "sturges" by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/199
* REL increase to version 1.4.2 by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/200


**Full Changelog**: https://github.com/lorentzenchr/model-diagnostics/compare/v1.4.1...v1.4.2

What's Changed
* DOC update examples with new release by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/196
* ENH rename collapsed "rest-n" to "other n" in binned strings by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/197
* REL increase to version 1.4.1 by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/198


**Full Changelog**: https://github.com/lorentzenchr/model-diagnostics/compare/v1.4.0...v1.4.1

Highlights
The binning methods of [`numpy.histogram_bin_edges`](https://numpy.org/doc/stable/reference/generated/numpy.histogram_bin_edges.html) are now available in `compute_bias`, `compute_marginal`, `plot_bias` and `plot_marginal`.
The default argument of `bin_method` of these functions changed to `"auto"`.
Additionally, strings/categorical/enum features now display all the remaining (unique) values of a feature as `"rest-n"` where `n` indicates the number of unique values. This way, `n_bins` is strictly adhered to.

The marginal plot from the example [Regression on Workers' Compensation Dataset](https://lorentzenchr.github.io/model-diagnostics/examples/regression_on_workers_compensation/#33-marginal-plot-a-model-overview-per-feature) now looks like this:


What's Changed
* MNT hatchling >= 1.26.3 by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/192
* DOC improve docstring of add_marginal_subplot by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/193
* ENH add numpy binning options by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/194
* REL increase to version 1.4.0 by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/195


**Full Changelog**: https://github.com/lorentzenchr/model-diagnostics/compare/v1.3.0...v1.4.0

Highlights
`plot_marginal` got a new argument `show_lines` and a helper function `add_marginal_subplot` to simplify subplots with plotly.
<img width="918" alt="image" src="https://github.com/user-attachments/assets/eba675c7-266b-4229-808b-30da6891ff57">

What's Changed
* DOC fix mkdocs-jupyter==0.24.7 by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/174
* MNT remove polyfill, javascipt updates and mkdocs-material version 9.5.31 by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/176
* MNT remove old fix for polars version 0.20 by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/182
* MNT simplify internal bin_feature a bit by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/183
* MNT remove old fix for polars version 0.20 part 2 by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/184
* CI switch back from macos-14 to macos-latest by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/185
* ENH add show_lines to plot_marginal by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/186
* FIX bin_edges with Null values in compute_marginal by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/188
* ENH add add_marginal_subplot by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/187
* CI use codecov token for upload by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/189
* ENH smaller confidence intervals in plot_reliability_diagram by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/190
* REL increase to version 1.3.0 by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/191


**Full Changelog**: https://github.com/lorentzenchr/model-diagnostics/compare/v1.2.0...v1.3.0

Highlights
The most important changes in this release are the following:
* The minimal required version of polars is now 1.0.0. This simplified a lot of code as polars had quite some deprecations in the releases before 1.0.0.
* Numpy 2.0.0 and higher is supported (and tested).
* 2 new functions for inspection of model calibration and model effects:
* `model_diagnostics.calibration.compute_marginal`
* `model_diagnostics.calibration.plot_marginal` which can look like this


What's Changed
* MNT Development Status :: 5 - Production/Stable by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/150
* MNT hatchling >= 1.24.2 by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/151
* MNT DOC mkdocs version and automatic dark-light mode by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/152
* DOC code block for pip install in README by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/153
* MNT ruff update, remove black by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/154
* MNT binning utilities by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/156
* DOC more general add unified docstring for y_pred by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/157
* MNT numpy 2.0.0 by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/158
* DOC add hatch config in development guide by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/159
* DOC make _config part of top level in API reference by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/160
* DOC add some more development commands for testing by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/161
* ENH compute_marginal and plot_marginal by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/162
* FIX drop in test_compute_marginal_multiple_predictions by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/164
* CI update test.yml and step order by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/163
* FIX typo in error message by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/166
* MNT increase minimum polars version to 1.0.0 by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/167
* DOC add more note links and minor doc improvements by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/168
* MNT remove polars_version by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/169
* CI add pyathon 3.12 and fix pandas without pyarrow installed by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/170
* DOC include inherited members in API reference by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/172
* REL increase to version 1.2.0 by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/171


**Full Changelog**: https://github.com/lorentzenchr/model-diagnostics/compare/v1.1.1...v1.2.0

This is a bugfix release.
What's Changed
* FIX polars 0.20.20 lazy neglects head(0) by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/148
* FIX remove argument plot_backend from plot_bias by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/147
* REL increase to version 1.1.1 by lorentzenchr in https://github.com/lorentzenchr/model-diagnostics/pull/149


**Full Changelog**: https://github.com/lorentzenchr/model-diagnostics/compare/v1.1.0...v1.1.1