Adding Thresholder and RejectOptionClassifier

docs/refs.bib

@@ -28,6 +28,16 @@ @inproceedings{agarwal2019fair
   url       = {http://proceedings.mlr.press/v97/agarwal19d.html}
 }
 
+@inproceedings{kamiran2012rejectoptionclassifier,
+  author={Kamiran, Faisal and Karim, Asim and Zhang, Xiangliang},
+  booktitle={2012 IEEE 12th International Conference on Data Mining}, 
+  title={Decision Theory for Discrimination-Aware Classification}, 
+  year={2012},
+  volume={},
+  number={},
+  pages={924-929},
+  doi={10.1109/ICDM.2012.45}}
+
 @inproceedings{hardt2016equality,
   author    = {Moritz Hardt and
                Eric Price and

docs/user_guide/datasets/boston_housing_data.rst

@@ -4,7 +4,6 @@ Revisiting the Boston Housing Dataset
 
 Introduction
 ^^^^^^^^^^^^^^^^^
-
 The Boston Housing dataset is one of the datasets currently callable in :mod:`fairlearn.datasets` module.
 In the past, it has commonly been used for benchmarking in popular machine learning libraries, 
 including `scikit-learn <https://scikit-learn.org/>`_ and `OpenML <https://www.openml.org/>`_. 
@@ -447,3 +446,16 @@ you pause about using it in the future.
 
   .. [#11] Kinmberlé Crenshaw, Mapping the margins: Intersectionality, identity politics, and violence against women of color, 
       Stanford Law Review, 1993, 43(6), 1241-1299.
+
+.. _hospital_readmissions_dataset:
+
+Hospital readmissions dataset
+------------------------------
+This is a clincial dataset of hospital readmissions over a ten-year period (1998-2008)
+for diabetic patients across 130 different hospitals in the US. Each record
+represents the hospital admission records for a patient diagnosed with
+diabetes whose stay lasted one to fourteen days. We would like to develop a
+classification model, which decides whether the patients should be suggested
+to their primary care physicians for an enrollment into the high-risk care
+management program. The positive prediction will mean recommendation into the
+care program.

docs/user_guide/index.rst

@@ -11,4 +11,4 @@ User Guide
    mitigation
    datasets/index
    installation_and_version_guide/index
-   further_resources
+   further_resources

docs/user_guide/installation_and_version_guide/v0.7.1.rst

@@ -0,0 +1,25 @@
+v0.7.1
+======
+
+.. note::
+
+  v0.7.1 is not yet released. This page reflects changes on the current
+  :code:`main` branch that will eventually be a part of v0.7.1.
+
+* Relaxed checks made on :code:`X` in :code:`_validate_and_reformat_input()`
+  since that is the concern of the underlying estimator and not Fairlearn.
+* Add support for Python 3.9
+* Added error handling in :code:`MetricFrame`. Methods :code:`group_max`, :code:`group_min`,
+  :code:`difference` and :code:`ratio` now accept :code:`errors` as a parameter,
+  which could either be :code:`raise` or :code:`coerce`.
+* Fixed a bug whereby passing a custom :code:`grid` object to a :code:`GridSearch`
+  reduction would result in a :code:`KeyError` if the column names were not ordered
+  integers.
+* :class:`~fairlearn.preprocessing.CorrelationRemover` now exposes
+  :code:`n_features_in_` and :code:`feature_names_in_`.
+* Added the ACSIncome dataset and corresponding documentation.
+* Added :class:`~fairlearn.postprocessing.Thresholder`, including the corresponding
+  documentation and a user guide.
+* Added :func:`~fairlearn.postprocessing.plot_histograms_per_group`,
+  :func:`~fairlearn.postprocessing.plot_positive_predictions`,
+  :func:`~fairlearn.postprocessing.plot_proba_distribution` to :code:`fairlearn.postprocessing`.