pyproject.toml

[tool.poetry]
name = "mltb2"
version = "1.0.3rc1"
description = "Machine Learning Toolbox 2"
authors = ["PhilipMay <philip@may.la>"]
readme = "README.md"
homepage = "https://github.com/telekom/mltb2"
documentation = "https://telekom.github.io/mltb2/"
include = ["LICENSE", "CONTRIBUTING.md", "Makefile"]
keywords = [
  "optuna",
  "deep learning",
  "ml",
  "ai",
  "machine learning",
  "hyperparameter optimization",
  "hyperparameter tuning",
  "nlp",
  "natural language processing",
  "transformers",
  "llm",
  "large-language-models",
  "openai",
  "plot",
]
# all classifiers see https://pypi.org/classifiers/
classifiers = [
  "Development Status :: 3 - Alpha",
  # "Development Status :: 4 - Beta",
  # "Development Status :: 5 - Production/Stable",
  "Intended Audience :: Science/Research",
  "Intended Audience :: Developers",
  "License :: OSI Approved :: MIT License",
  "Natural Language :: German",
  "Natural Language :: English",
  # "Programming Language :: Python :: 3.8",
  "Programming Language :: Python :: 3.9",
  "Programming Language :: Python :: 3.10",
  "Programming Language :: Python :: 3.11",
  "Programming Language :: Python :: 3.12",
  "Programming Language :: Python :: 3 :: Only",
  "Topic :: Scientific/Engineering :: Mathematics",
  "Topic :: Scientific/Engineering :: Artificial Intelligence",
  "Topic :: Software Development :: Libraries :: Python Modules",
  "Operating System :: OS Independent",
]

[build-system]
requires = ["poetry-core"]
build-backend = "poetry.core.masonry.api"

[tool.poetry.scripts]
arango-col-backup = 'mltb2.arangodb:arango_collection_backup'

[tool.poetry.urls]
"Bug Tracker" = "https://github.com/telekom/mltb2/issues"

[tool.poetry.dependencies]
python = "^3.9"
numpy = "^1"  # restriction by fasttext
scipy = "*"
tqdm = "*"
platformdirs = {version = "*", optional = true}
scikit-learn = {version = "*", optional = true}
fasttext-wheel = {version = "*", optional = true}
optuna = {version = "*", optional = true}
matplotlib = {version = "*", optional = true}
SoMaJo = {version = ">=2.4.1", optional = true}

# some versions have poetry issues
# 2.3.0 does not work with Intel Mac
torch = {version = "!=2.0.1,!=2.1.0,!=2.3.0", optional = true}

transformers = {version = "*", optional = true}
tiktoken = {version = "*", optional = true}
safetensors = {version = "!=0.3.2", optional = true}  # version 0.3.2 has poetry issues
openai = {version = "^1", optional = true}
pyyaml = {version = "*", optional = true}
pandas = {version = "*", optional = true}
beautifulsoup4 = {version = "*", optional = true}
joblib = {version = "*", optional = true}
python-dotenv = {version = "*", optional = true}
python-arango = {version = "*", optional = true}
jsonlines = {version = "*", optional = true}
markdownify = {version = "*", optional = true}
mdformat = {version = "*", optional = true}

[tool.poetry.extras]
files = ["platformdirs", "scikit-learn", "joblib"]
fasttext = ["fasttext-wheel", "platformdirs", "scikit-learn"]
data = ["platformdirs", "scikit-learn", "pandas", "beautifulsoup4", "joblib"]
optuna = ["optuna"]
plot = ["matplotlib"]
somajo = ["SoMaJo"]
transformers = ["scikit-learn", "torch", "transformers", "safetensors"]
md = ["scikit-learn", "torch", "transformers", "safetensors"]
somajo_transformers = ["SoMaJo", "scikit-learn", "torch", "transformers", "safetensors"]
openai = ["tiktoken", "openai", "pyyaml"]
arangodb = ["python-dotenv", "python-arango", "jsonlines"]
bs = ["beautifulsoup4", "markdownify", "mdformat"]

[tool.poetry.group.lint.dependencies]
black = "*"
ruff = "*"
mypy = "*"
mdformat-gfm = "*"
mdformat-frontmatter = "*"
mdformat-footnote = "*"

[tool.poetry.group.test.dependencies]
pytest = "*"
pytest-rerunfailures = "*"
pytest-random-order = "*"
hypothesis = "*"

[tool.poetry.group.doc.dependencies]
sphinx = "*"
sphinx_rtd_theme = "*"
sphinx_copybutton = "*"

[tool.pytest.ini_options]
addopts = "--random-order-bucket=global"

[tool.black]
line-length = 119
target-version = ["py39", "py310", "py311", "py312"]

[tool.ruff]
line-length = 119
target-version = "py39"


[tool.ruff.lint]
select = ["ALL"]
fixable = ["I"]
ignore = [
  "DJ",  # flake8-django - https://docs.astral.sh/ruff/rules/#flake8-django-dj
  "ERA",  # eradicate - https://docs.astral.sh/ruff/rules/#eradicate-era
  "ANN",  # flake8-annotations - https://docs.astral.sh/ruff/rules/#flake8-annotations-ann
  "FA",  # flake8-future-annotations - https://docs.astral.sh/ruff/rules/#flake8-future-annotations-fa
  "EM",  # flake8-errmsg - https://docs.astral.sh/ruff/rules/#flake8-errmsg-em
  "PTH",  # flake8-use-pathlib - https://docs.astral.sh/ruff/rules/#flake8-use-pathlib-pth
  "FBT",  # flake8-boolean-trap - https://docs.astral.sh/ruff/rules/#flake8-boolean-trap-fbt
  "TD", # flake8-todos - https://docs.astral.sh/ruff/rules/#flake8-todos-td
  "SLF",  # flake8-self - https://docs.astral.sh/ruff/rules/#flake8-self-slf
  "D107",  # Missing docstring in __init__
  "D410",  # Missing blank line after section ("{name}")
  "D411",  # Missing blank line before section ("{name}")
  "PLR0913",  # Too many arguments to function call ({c_args} > {max_args})
  "S106",  # Possible hardcoded password assigned to argument: "{}"
  "COM812",  # Trailing comma missing
  "S101",  # Use of `assert` detected
  "PLR2004",  # Magic value used in comparison
  "B011",  # Do not `assert False`
  "RET505",  # Unnecessary `else` after `return` statement
  "TRY003",  # Avoid specifying long messages outside the exception class
  "RET504",  # Unnecessary assignment before `return` statement
  "T201",  # `print` found
  "RET507",  # Unnecessary `else` after `continue` statement
  "PT015",  # Assertion always fails, replace with `pytest.fail()`
  "UP015",  # Unnecessary open mode parameters
  "FIX002",  # Line contains TODO, consider resolving the issue
  "PT011",  # `pytest.raises(ValueError)` is too broad, set the `match` parameter or use a more specific exception
  "PT001",  # Use `@pytest.fixture()` over `@pytest.fixture`
  "RUF015",  # Prefer `next(iter(sentences))` over single element slice
]

[tool.ruff.lint.per-file-ignores]
"tests/**/test_*.py" = [
  "D100",  # Missing docstring in public module
  "D103",  # Missing docstring in public function
  "PLR2004",  # Magic value used in comparison, consider replacing {value} with a constant variable
  "S101",  # Use of assert detected
  "N802",  # Function name should be lowercase
]

[tool.ruff.lint.pydocstyle]
convention = "google"

[tool.ruff.lint.flake8-copyright]
notice-rgx = "(# Copyright \\(c\\) \\d{4}.*\\n)+# This software is distributed under the terms of the MIT license\\n# which is available at https://opensource.org/licenses/MIT\\n\\n"

[tool.mypy]
ignore_missing_imports = true
warn_unused_ignores = true
enable_error_code=["ignore-without-code"]