* update readme install section (#864)

* [DOC] Multiple small improvements (#867)

* Small doc improvements

* Fix typos in AggTarget and AggJoiner

* Fix bulleted lists in TableVectorizer doc

* Update skrub/_table_vectorizer.py

Co-authored-by: Jérôme Dockès <[email protected]>

* Change _capitals to _aux in Joiner docstring

* add whitespace so printed dataframe columns align

* running ci to reproduce the error

---------

Co-authored-by: Jérôme Dockès <[email protected]>

* MAINT make sure pandas doesn't insert a "key_0" column in merge result (#878)

* MAINT scikit-learn estimator check should not be run yet #879 

because refrenced pr has not been merged before 1.4

* MAINT compatibility with pandas >= 2.2 (#880)

* make pandas output consistent in tests

* bypass pyarrow warning

* ignore warning

---------

Co-authored-by: Guillaume Lemaitre <[email protected]>

* MAINT avoid tie breaking in joiner examples (#882)

* MAINT compatibility for scikit-learn 1.4 (#883)

* MAINT compatibility for scikit-learn 1.4

* iter

* use _safe_set_output

* iter

* use string column names

* avoid doctest failure due to sklearn 1.4 change

---------

Co-authored-by: Jerome Dockes <[email protected]>

* DOC add imports in examples to have self-contained snippet (#875)

Co-authored-by: Guillaume Lemaitre <[email protected]>

* DOC more consistent formatting in docstring (#881)

* Small doc improvements

* Fix typos in AggTarget and AggJoiner

* Fix bulleted lists in TableVectorizer doc

* Iter

* Uniformize keys

* Format docstrings

* Draft multiaggjoiner and multijoiner

* Revamp aggjoiner like joiner

* Make new aggjoiner work

* Iter

* Add multijoiner and multiaggjoiner to __init__.py

* Update assembling doc

* Draft multiaggjoiner checks

* Add multijoiner and multiaggjoiner to api reference

* Remove comments

* Docstrings

* Write example for multiaggjoiner

* Check case where aux_table is 'X'

* Change TypeError into ValueError

* Change 'list' into 'iterable' in joiner docstring following PR 742 discussion

* Update aggjoiner tests

* Implement multiaggjoiner transform

* Implement multiaggjoiner transform, fix fit

* Don't check duplicate columns when aggregating on the main table

* Test that aggjoiner doesn't work with multiple tables

* Iter tests

* Check column names after aggjoiner aggregation

* Check that all cols are present in aux_table

* Fix typo aux_key into _aux_key

* Check suffix in aggjoiner, pass all tests

* Transform test df into a fixture, support polars in aggjoiner test

* Format aggjoiner using utils

* Iter multiaggjoiner

* Add new estimators to numpydoc xref

* Update public vs private attributes for aggjoiner

* Check operations, check missing columns in multiaggjoiner

* Update docstrings

* Iter check operations

* Draft tests for cols and operations check in multiaggjoiner

* Iter

* Iter checks multiaggjoiner

* Implement checks in aggjoiner transform, format joiner

* Check dataframes in multiaggjoiner

* Iter _check_keys

* Iter _check_keys, _check_cols and _check_operations

* Make multiaggjoiner work

* Split aggjoiner tests into smaller chunks

* Test cols, operations, suffixes in multiaggjoiner

* Small doc changes

* Uniformize keys order

* Update skrub/_dataframe/_pandas.py

Co-authored-by: Guillaume Lemaitre <[email protected]>

* Update skrub/_dataframe/_polars.py

Co-authored-by: Guillaume Lemaitre <[email protected]>

* Update skrub/_dataframe/_polars.py

Co-authored-by: Guillaume Lemaitre <[email protected]>

* Update skrub/_join_utils.py

Co-authored-by: Guillaume Lemaitre <[email protected]>

* Update _join_utils.py

* Update _pandas.py

---------

Co-authored-by: Guillaume Lemaitre <[email protected]>

* MAINT ignore deprecation warning arising from pandas after numpy 1.25 (#887)

* import parse_version from sklearn.utils.fixes (#892)

* temporarily skip ken embedding download tests (#901)

* changelog

---------

Co-authored-by: Vincent M <[email protected]>
Co-authored-by: Théo Jolivet <[email protected]>
Co-authored-by: Guillaume Lemaitre <[email protected]>