[build-system]

requires = ["setuptools>=77.0.3", "setuptools_scm[toml]>=8"]

build-backend = "setuptools.build_meta"

[project]

name = "skrub"

dynamic = ["version"]

description = "Machine learning with dataframes"

readme = "README.rst"

authors = [

{ name = "Riccardo Cappuzzo" },

{ name = "Jerome Dockes" },

{ name = "Guillaume Lemaitre" },

{ name = "Vincent Maladiere" },

{ name = "Gael Varoquaux" },

]

license = "BSD-3-Clause"

license-files = ["LICENSE.txt"]

classifiers = [

"Development Status :: 5 - Production/Stable",

"Environment :: Console",

"Intended Audience :: Science/Research",

"Operating System :: OS Independent",

"Programming Language :: Python :: 3.10",

"Programming Language :: Python :: 3.11",

"Programming Language :: Python :: 3.12",

"Programming Language :: Python :: 3.13",

"Programming Language :: Python :: 3.14",

"Topic :: Scientific/Engineering",

"Topic :: Software Development :: Libraries",

]

requires-python = ">=3.10"

dependencies = [

"numpy>=1.23.5",

"pandas>=1.5.3",

"scikit-learn>=1.4.2",

"scipy>=1.9.3",

"jinja2>=3.1.2",

"matplotlib>=3.4.3",

"requests>=2.27.1",

"pydot",

]

[tool.setuptools.dynamic]

version = { file = "skrub/VERSION.txt" }

[project.optional-dependencies]

dev = [

"ipykernel",

"ipython",

"jupyterlab",

# doc

"jupyterlite-sphinx",

"jupyterlite-pyodide-kernel",

"numpydoc",

"pydata-sphinx-theme",

"seaborn",

# TODO: unpin sphinx when incmpatibility with autosummary-accessors is fixed:

# https://github.com/xarray-contrib/sphinx-autosummary-accessors/issues/165

"sphinx<9",

"sphinx-copybutton",

"sphinx-gallery",

"sphinxext-opengraph",

"sphinx-autosummary-accessors",

"statsmodels",

# lint

"black==23.3.0",

"ruff==0.14.2",

"pre-commit",

# test

"pytest",

"pytest-cov",

"pytest-xdist",

# optional

"pyarrow",

"polars",

"plotly",

"optuna",

]

transformers = ["sentence-transformers"]

[project.urls]

Homepage = "https://skrub-data.org/"

Source = "https://github.com/skrub-data/skrub"

Issues = "https://github.com/skrub-data/skrub/issues"

[tool.setuptools]

packages = ["skrub"]

[tool.pixi.workspace]

channels = ["conda-forge", "pytorch"]

platforms = ["linux-64", "osx-arm64", "osx-64", "win-64"]

[tool.pixi.dependencies]

numpy = ">=1.23.5"

pandas = ">=1.5.3"

scikit-learn = ">=1.4.2"

scipy = ">=1.9.3"

jinja2 = ">=3.1.2"

matplotlib = ">=3.4.3"

pydot = "*"

requests = ">=2.27.1"

[tool.pixi.feature.doc.dependencies]

jupyterlite-sphinx = "*"

jupyterlite-pyodide-kernel = "*"

matplotlib = "*"

pydot = "*"

numpydoc = "*"

pydata-sphinx-theme = "*"

plotly = "*"

seaborn = "*"

# TODO: unpin sphinx when incmpatibility with autosummary-accessors is fixed:

# https://github.com/xarray-contrib/sphinx-autosummary-accessors/issues/165

sphinx = "<9"

sphinx-copybutton = "*"

sphinx-gallery = "*"

sphinxext-opengraph = "*"

sphinx-autosummary-accessors = ">=2025.3.1,<2026"

sphinx-sitemap = "*"

statsmodels = "*"

optuna = "*"

[tool.pixi.feature.lint.dependencies]

black = "==23.3.0"

ruff = "==0.14.2"

pre-commit = "*"

[tool.pixi.feature.optional.dependencies]

pyarrow = "*"

polars = "*"

plotly = "*"

optuna = "*"

[tool.pixi.feature.transformers]

# pytorch is not anymore released on osx-64 platform

platforms = ["linux-64", "osx-arm64", "win-64"]

[tool.pixi.feature.transformers.dependencies]

pytorch = { version = ">=2.4.1,<3" }

transformers = "*"

datasets = "*"

sentence-transformers = ">=3.0,<6"

tokenizers = ">=0.22.0,<=0.23.0"

pyarrow = "<21"

[tool.pixi.feature.test.dependencies]

numpydoc = "*"

pytest = "*"

pytest-cov = "*"

pytest-xdist = "*"

[tool.pixi.feature.dev.dependencies]

ipykernel = "*"

ipython = "*"

jupyterlab = "*"

[tool.pixi.pypi-dependencies]

skrub = { path = ".", editable = true }

[tool.pixi.feature.min-dependencies.dependencies]

numpy = "==1.23.5"

pandas = "==1.5.3"

scikit-learn = "==1.4.2"

scipy = "==1.9.3"

jinja2 = "==3.1.2"

matplotlib = "==3.4.3"

pydot = "*"

requests = "==2.27.1"

# pinning pyparsing because it raises deprecation warnings in matplotlib

# TODO: unpin when it gets fixed

pyparsing = "<3.3"

[tool.pixi.feature.min-optional-dependencies.dependencies]

pyarrow = "==15.0.2"

polars = "==0.20.15"

plotly = "*"

optuna = "*"

[tool.pixi.feature.nightly-dependencies.pypi-options]

# Get the nightly of as many dependencies as possible

extra-index-urls = [

# scikit-learn, numpy, pandas...

"https://pypi.anaconda.org/scientific-python-nightly-wheels/simple",

# pyarrow

"https://pypi.fury.io/arrow-nightlies",

]

[tool.pixi.feature.nightly-dependencies.pypi-dependencies]

# Have to repeat dependencies here because otherwise pixi pulls from conda channels

# rather than PyPI indexes

skrub = { path = ".", editable = true }

numpy = "*"

pandas = "*"

scikit-learn = "*"

scipy = "*"

jinja2 = "*"

matplotlib = "*"

pydot = "*"

requests = "*"

pyarrow = "*"

polars = "*"

plotly = "*"

[tool.pixi.feature.polars-without-pyarrow.dependencies]

polars = "*"

plotly = "*"

optuna = "*"

[tool.pixi.feature.py310.dependencies]

python = "~=3.10.0"

[tool.pixi.feature.py311.dependencies]

python = "~=3.11.0"

[tool.pixi.feature.py314.dependencies]

python = "~=3.14.0"

[tool.pixi.feature.doc.tasks]

build-doc = { cmd = "make html", cwd = "doc" }

build-doc-quick = { cmd = "make html-noplot", cwd = "doc" }

clean-doc = { cmd = "rm -rf _build/ && rm -rf auto_examples/ && rm -rf reference/generated/", cwd = "doc" }

linkcheck = { cmd = "make linkcheck", cwd = "doc" }

linkcheck-quick = { cmd = "make linkcheck-noplot", cwd = "doc" }

[tool.pixi.feature.lint.tasks]

lint = { cmd = "pre-commit install && pre-commit run -v --all-files --show-diff-on-failure" }

[tool.pixi.feature.check-pyi-diff.tasks]

# Note: this only works on Unix

check = { cmd = "python build_tools/generate_data_ops_stub.py | diff - skrub/_data_ops/_data_ops.pyi" }

[tool.pixi.feature.test.tasks]

test = { cmd = "pytest -vsl --cov=skrub --cov-report=xml skrub" }

test-user-guide = { cmd = "pytest doc/**/*.rst" }

[tool.pixi.environments]

lint = ["lint"]

check-pyi-diff = ["check-pyi-diff"]

doc = ["optional", "doc", "transformers"]

test = ["optional", "test"]

dev = ["dev", "optional", "doc", "lint", "test", "transformers"]

ci-py310-min-deps = ["py310", "min-dependencies", "test"]

ci-py310-min-optional-deps = [

"py310",

"min-dependencies",

"min-optional-dependencies",

"test",

]

ci-py311-transformers = ["py311", "transformers", "test"]

ci-py314-latest-deps = ["py314", "test"]

ci-py314-polars-without-pyarrow = ["py314", "test", "polars-without-pyarrow"]

ci-py314-latest-optional-deps = ["py314", "optional", "test"]

# no-default-feature is necessary for pixi to download from nightly PyPI indexes rather than conda channels

ci-nightly-deps = { no-default-feature = true, features = [ "py314", "nightly-dependencies", "test" ] }

[tool.ruff]

line-length = 88

exclude = [

".git",

"__pycache__",

"dist",

"doc/_build",

"doc/auto_examples",

"build",

"pixi.lock"

]

[tool.ruff.lint]

# all rules can be found here: https://beta.ruff.rs/docs/rules/

select = ["E", "F", "W", "C4", "I", "UP"]

ignore = [

# use `is` and `is not` for type comparisons

"E721",

# do not assign a lambda expression, use a def

"E731",

# do not use variables named 'l', 'O', or 'I'

"E741",

# unnecessary list comprehension (rewrite as a set comprehension)

"C403",

# unnecessary tuple literal (rewrite as a set literal)

"C405",

# unnecessary `dict()` call (rewrite as a literal)

"C408",

# unnecessary list literal passed to `tuple()` (rewrite as a tuple literal)

"C409",

]

[tool.ruff.lint.per-file-ignores]

# It's fine not to put the import at the top of the file in the examples

# folder.

"examples/*" = ["E402"]

"doc/conf.py" = ["E402"]

# Long exception messages in docstrings

"skrub/_clean_null_strings.py" = ["E501"]

[tool.pytest.ini_options]

filterwarnings = [

# Turn deprecation warnings into errors

"error::FutureWarning",

"error::DeprecationWarning",

# Ignore warning from np.in1d since the future behavior is already the desired

# behavior. TODO remove when numpy min version >= 1.25.

'ignore:elementwise\ comparison\ failed:FutureWarning',

# TODO remove when pandas min version >= 2.2.0

'ignore:.*utcfromtimestamp\(\) is deprecated.*:DeprecationWarning',

# TODO remove when joblib min version >= 1.4.0

'ignore:.*ast\.Num is deprecated.*:DeprecationWarning',

# TODO remove when joblib min version >= 1.4.0

'ignore:.*ast\.NameConstant is deprecated.*:DeprecationWarning',

# TODO remove when joblib min version >= 1.4.0

'ignore:.*Attribute n is deprecated.*:DeprecationWarning',

# FIXME we need to decided what to do with pyarrow that is required by pandas >= 3.0

'ignore:(?s).*Pyarrow will become a required dependency of pandas.*:DeprecationWarning',

# accessing .values on a pandas dataframe raises this warning after numpy 1.25;

# should be addressed in pandas

'ignore:np.find_common_type is deprecated.*:DeprecationWarning',

'ignore:.*distutils Version classes are deprecated.*',

'ignore:.*`resume_download` is deprecated.*',

'ignore:.*`clean_up_tokenization_spaces` was not set.*',

'ignore:The current process just got forked.*',

'ignore:.*`clean_up_tokenization_spaces` was not set.*',

]

addopts = "--doctest-modules"

doctest_optionflags = "NORMALIZE_WHITESPACE ELLIPSIS"

[tool.codespell]

# Ref: https://github.com/codespell-project/codespell#using-a-config-file

skip = '.git*,*.svg,package-lock.json,*.lock,*.css,*-min.*'

check-hidden = true

# ignore-regex = ''

ignore-words-list = 'ans,serie,fulfilment'