pyproject.toml

[build-system]
requires = ["hatchling>=1.8.0"]
build-backend = "hatchling.build"

[project]
name = "temporalscope"
dynamic = ["version"]
description = "TemporalScope: Model-Agnostic Temporal Feature Importance Analysis."
authors = [
  { name = "Philip Ndikum", email = "philip-ndikum@users.noreply.github.com" },
  { name = "Serge Ndikum" },
  { name = "Kane Norman", email = "kanenorman@fas.harvard.edu" },
]
license = "Apache-2.0"
readme = "README.md"
requires-python = ">=3.10,<3.12"
dependencies = [
  # Core Explainability Libraries
  "shap>=0.46.0",       # SHAP (SHapley Additive exPlanations): For model interpretability and feature importance in machine learning.
  "borutashap>=1.0.17", # Boruta-SHAP: Combines SHAP with Boruta for feature selection, helping identify the most important features.
  "lime>=0.2.0.1",      # LIME (Local Interpretable Model-agnostic Explanations): Enables model-agnostic interpretability, focusing on local, instance-specific explanations.

  # General Machine Learning and Statistics
  "scikit-learn>=1.5.1", # Scikit-Learn: Essential machine learning library for classification, regression, and clustering.
  "lightgbm>=4.5.0",     # LightGBM: Gradient-boosted decision tree library that's efficient for high-dimensional data.
  "statsmodels>=0.14.2", # Statsmodels: Provides statistical models and hypothesis tests, useful for traditional time series analysis.

  # Data Manipulation and Storage
  "pandas>=1.5.0",           # Pandas: Core library for handling structured data in DataFrames.
  "modin[all]>=0.31.0",      # Modin: Parallelizes Pandas operations for large datasets with Dask/Ray backend.
  "pyarrow>=17.0.0",         # PyArrow: Enables efficient columnar data format (Arrow) for fast data processing and storage, commonly used with Parquet files.
  "polars>=1.5.0",           # Polars: High-performance DataFrame library, ideal for handling large datasets in memory.
  "dask[dataframe]>=2024.7", # Dask DataFrame: Handles parallelized computations for very large datasets.

  # Specialized Computation
  "flax>=0.8.5", # Flax: Neural network library for JAX, useful for building deep learning models.
  "jax>=0.4.31", # JAX: Accelerated computation for machine learning research, supporting CPU/GPU/TPU backends.

  # Backend-Agnostic Processing
  "narwhals>=0.2.3", # Narwhals: Ensures backend compatibility across Pandas, Polars, Modin, and other dataframes, optimizing performance.

  # Environment Configuration
  "python-dotenv>=1.0.1", # Python-Dotenv: Manages environment variables for secure and flexible configuration.

  # Markdown Table Display
  "tabulate>=0.9.0", # Tabulate: Allows Markdown-friendly table formatting for DataFrames in Narwhals.
]
classifiers = [
  "Development Status :: 2 - Pre-Alpha",
  "Intended Audience :: Science/Research",
  "License :: OSI Approved :: Apache Software License",
  "Operating System :: OS Independent",
  "Programming Language :: Python",
  "Programming Language :: Python :: 3",
  "Programming Language :: Python :: 3.10",
  "Programming Language :: Python :: 3.11",
  "Programming Language :: Python :: 3.12",
  "Topic :: Scientific/Engineering :: Artificial Intelligence",
]
keywords = [
  "Shap",
  "Borutashap",
  "Feature-Importance",
  "Temporal-Feature-Importance",
  "Temporal-Feature-Analysis",
  "XAI",
  "ML",
  "AI",
  "Machine-Learning",
  "Artificial-Intelligence",
  "TemporalScope",
  "Time-Series",
]

[project.optional-dependencies]
docs = [
  "mkdocs",
  "mkdocstrings-python",
  "mkdocs-material",
  "mkdocs-gen-files",
  "mkdocs-literate-nav",
  "mkdocs-section-index",
  "markdown-callouts",
  "black",
]

[project.urls]
"Source code" = "https://github.com/philip-ndikum/TemporalScope"
Documentation = "https://temporalscope.readthedocs.io/en/latest/"

[tool.hatch.envs.default]
dependencies = [
  "pre-commit",
  "ruff",
  "jupyterlab",
  "notebook",
  "commitizen",
  "types-tabulate>=0.9.0",
]

[tool.hatch.envs.docs]
features = ["docs"]

[tool.hatch.envs.test]
extra-dependencies = [
  "pytest",
  "pytest-cov",
  "pytest-custom_exit_code",
  "pytest-mock",
  "papermill>=2.5.0",
]

[tool.hatch.envs.docs.scripts]
build = "mkdocs build --strict"
serve = "mkdocs serve --strict"

[tool.hatch.envs.test.scripts]
unit = 'pytest --cov-report=lcov --cov-report=term --cov="temporalscope" -m "not integration and not notebook" {args:test}'
notebook = 'pytest -m "notebook" {args:test}'
integration = 'pytest -m "integration" {args:test}'

[tool.pytest.ini_options]
testpaths = ["tests"]
addopts = "-v -ra -q"
log_cli = true
log_cli_level = "INFO"
log_format = "[%(asctime)s][%(levelname)s] %(message)s"
log_date_format = "%Y-%m-%d %H:%M:%S"
minversion = "6.0"
filterwarnings = "ignore"
markers = [
  "notebook: marks tests that verify notebook execution",
  "integration: marks tests that require external resources",
]

[tool.ruff.format]
quote-style = "double"
indent-style = "space"
docstring-code-format = true

[tool.ruff]
extend-exclude = ["*.pyc", "tutorial_notebooks/*"]
target-version = "py310"
line-length = 120

[tool.ruff.lint]
select = [
  "A001", # builtin-variable-shadowing
  "A002", # builtin-argument-shadowing
  "A003", # builtin-attribute-shadowing
  "C4",   # flake8-comprehensions
  "C90",  # McCabe cyclomatic complexity
  "D",    # flake8-docstrings
  "E501", # Long lines
  "EXE",  # flake8-executable
  "F",    # Pyflakes
  "I",    # isort
  "INT",  # flake8-gettext
  "PERF", # Perflint
  "PL",   # Pylint
  "Q",    # flake8-quotes
  "SIM",  # flake8-simplify
  "SLOT", # flake8-slots
  "T10",  # flake8-debugger
  "W",    # pycodestyle
  "YTT",  # flake8-2020
]

ignore = [
  "C901",    # temporary
  "D100",    # Missing docstring in public module
  "D104",    # Missing docstring in public package
  "D203",    # 1 blank line required before class docstring
  "D211",    # No blank lines allowed before class docstring
  "D213",    # Multi-line docstring summary should start at the second line
  "D400",    # Ignore "First line should end with a period" for docstrings.
  "D401",    # Ignore "First line should be in imperative mood" for docstrings.
  "D415",    # Ignore "First line should end with a period, question mark, or exclamation point."
  "E501",    # Ignore "Line too long" in docstrings/comments for exceeding 120 characters.
  "PERF203", # `try`-`except` within a loop incurs performance overhead
  "PERF401", # Use a list comprehension to create a transformed list
  "PLR0913", # too many arguments
  "PLR1714", # repeated-equality-comparison
  "PLR5501", # collapsible-else-if
  "PLW2901", # redefined-loop-name
  "SIM102",  # temporary
  "SIM108",  # if-else-block-instead-of-if-exp
]

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

[tool.ruff.lint.per-file-ignores]
"docs/conf.py" = [
  "A001", # builtin-variable-shadowing
  "D103", # missing docstring in public function
]
"test/*" = ["PLR2004"] # Ignore magic number warnings in test files

[tool.mypy]
files = "src/temporalscope"
python_version = "3.10"
ignore_missing_imports = true
warn_unreachable = true
exclude = 'test/*'
warn_return_any = false       # Turn off MyPy warnings for missing return types

[tool.bandit]
exclude_dirs = ["test"]

[tool.hatch.build.targets.wheel]
packages = ["src/temporalscope"]

[tool.hatch.build.targets.sdist]
include = ["/src", "/VERSION.txt"]

[tool.hatch.version]
path = "VERSION.txt"
pattern = "(?P<version>.+)"

[tool.hatch.envs.default.scripts]
check = "ruff check {args}"
fix = "ruff check --fix"
format = "ruff format {args}"
format-check = "ruff format --check {args}"
clear-coverage = "coverage erase"
generate-kernel = """
python -m ipykernel install --user --name temporalscope-kernel --display-name "TemporalScope"
echo "Jupyter kernel 'TemporalScope' created. You can now use it in Jupyter notebooks."
"""

[tool.commitizen]
version = "0.1.0"
update_changelog_on_bump = true
tag_format = "v$version"
version_files = ["pyproject.toml:version", "VERSION.txt"]

[tool.hatch.metadata]
allow-direct-references = true