pyproject.toml

[tool.poetry]
authors = ["kstathou"]
description = "Experiments with Large Language Models (LLMs). Fine-tune Mistral."
homepage = "https://github.com/kstathou/llm-stack"
keywords = ['package', 'python', 'machine-learning'] # noqa
name = "llm-stack"
packages = [
  {include = "llm_stack", from = "src"},
]
readme = "README.md"
repository = "https://github.com/kstathou/llm-stack"
version = "0.1.0"

[tool.poetry.build]
generate-setup-file = false

[tool.poetry.dependencies]
python = "3.11.5"
datasets = "^2.15.0"
transformers = "^4.35.2"
modal = "^0.55.4147"
peft = "^0.6.2"
evaluate = "^0.4.1"
bitsandbytes = "^0.41.2.post2"
wandb = "^0.16.0"
accelerate = "^0.24.1"
torch = "2.0.1"
feedparser = "^6.0.10"
python-dotenv = "^1.0.0"
httpx = "^0.25.2"
openai = "^1.3.7"
tenacity = "^8.2.3"
tqdm = "^4.66.1"


[tool.poetry.group.test]
optional = true

[tool.poetry.group.test.dependencies]
pytest = "^7.1.1" # Allows for testing of the project
pytest-cov = "^4.0.0" # Allows to run coverage of the project

[tool.poetry.group.lint]
optional = true

[tool.poetry.group.lint.dependencies]
bandit = {version = "^1.7.4", extras = ["toml"]}# Security linter
ruff = "^0.1.6" # black + flake8 + isort
pre-commit = "^2.18.1" # Runs a pipeline before commiting code
yamllint = "^1.26.3" # Linter for YAML files
isort = "^5.10.1" # Sort imports

[tool.poetry.group.dev.dependencies]
ipykernel = "^6.23.3"

[build-system]
build-backend = "poetry.core.masonry.api"
requires = ["poetry-core>=1.2.0rc1"]

[tool.ruff]
# Exclude a variety of commonly ignored directories.
exclude = [
    ".bzr",
    ".direnv",
    ".eggs",
    ".git",
    ".git-rewrite",
    ".hg",
    ".mypy_cache",
    ".nox",
    ".pants.d",
    ".pytype",
    ".ruff_cache",
    ".svn",
    ".tox",
    ".venv",
    ".vscode",
    "__pypackages__",
    "_build",
    "buck-out",
    "build",
    "dist",
    "node_modules",
    "venv",
    ".env",
]

# Same as Black.
line-length = 119
indent-width = 4

# Assume Python 3.11
target-version = "py311"

[tool.ruff.lint]
# Enable Pyflakes (`F`) and a subset of the pycodestyle (`E`)  codes by default.
# Unlike Flake8, Ruff doesn't enable pycodestyle warnings (`W`) or
# McCabe complexity (`C901`) by default.
select = [
  "C", # McCabe complexity
  "E", # pep8 errors
  "W", # pep8 warnings
  "F", # pyflakes errors
  "N", # naming conventions
  "B", # bugbear errors
  "ANN", # flake8 annotations errors
  "T", # flake8 print errors
  "D", # flake8 doscstrings errors
  ]

ignore = [
  "D213", # Multi-line docstring summary should start at the first line
  "D203", # one-blank-line-before-class
  "E501", # Line too long  (using B950 instead, which has 10% tolerance)
  "D107", # Missing docstring in __init__
  "D202", # No blank lines allowed after function docstring
  "D400", # First line should end with a period
  "D100", # Missing docstring in public module
  "D104", # Missing docstring in public package
  "ANN003", # Missing type annotation for **kwargs
  "ANN002", # Missing type annotation for **args
  "ANN1", # Missing type annotation for self in method or cls method
  "ANN101", # Missing type annotation for class method
  ]

# Allow fix for all enabled rules (when `--fix`) is provided.
fixable = ["ALL"]
unfixable = []

# Allow unused variables when underscore-prefixed.
dummy-variable-rgx = "^(_+|(_+[a-zA-Z0-9_]*[a-zA-Z0-9]+?))$"

[tool.ruff.format]
# Like Black, use double quotes for strings.
quote-style = "double"

# Like Black, indent with spaces, rather than tabs.
indent-style = "space"

# Like Black, respect magic trailing commas.
skip-magic-trailing-comma = false

# Like Black, automatically detect the appropriate line ending.
line-ending = "auto"

[tool.isort]
combine_as_imports = true
force_single_line = true
line_length = 119
lines_after_imports = 2
lines_between_types = 1
profile = 'black'
src_paths = ["src", "tests"]

[tool.bandit]
exclude_dirs = [
  ".venv",
  "cdk.out",
]
skips = [
  # Choose the bandit errors to ignore globally
  "B101", # Use of assert detected. The enclosed code will be removed when compiling to optimised byte code.
  "B311", # Standard pseudo-random generators are not suitable for security/cryptographic purposes.
  "B608", # Possible SQL injection vector through string-based query construction.
]

[tool.pytest.ini_options]
addopts = """
  --verbose
  --color=yes
  --assert=plain
  --cov-report term
  --cov=src
  --last-failed
"""
minversion = "7.1"
testpaths = [
  "tests",
]

[tool.coverage.run]
omit = ["*/tests/*"] # Remove test files from coverage run.

[tool.coverage.report]
omit = ["*/tests/*"] # Remove test files from coverage report.