pyproject.toml

[project]
name = "LexaLCM_Pre0_288M"
readme = "README.md"
requires-python = ">=3.10"
version = "0.0.0_pre0"
description = "A pre-trained LCM model with 288M parameters based on Meta FAIR's LCM architecture."
dependencies = [
  "dacite>=1.8.1",
  "fire>=0.7.0",
  "hydra-core>=1.3.2",
  "importlib-resources~=6.4",
  "numpy>=1.21",
  "polars>=1.16.0",
  "pyarrow>=16.1.0",
  "retrying>=1.3.4",
  "safetensors>=0.5.3",
  "sentence-splitter>=1.4",
  "sonar-space>=0.3.2",
  "stopes[mono]>=2.2.0",
  "tensorboard>=2.18.0",
  "wandb>=0.19.11",
]

classifiers = [
  "License :: OSI Approved :: MIT License",
  "Topic :: Scientific/Engineering",
  "Development Status :: 4 - Beta",
]

[build-system]
requires = ["flit_core >=3.2,<4", "setuptools < 74"]
build-backend = "flit_core.buildapi"

[tool.flit.module]
name = "lcm" # TODO change module name

[project.optional-dependencies]
gpu = [
  "torch==2.5.1",
  "torchaudio==2.5.1",
  "fairseq2n==0.3.0rc1",
  "fairseq2[arrow]==0.3.0rc1",
]
# cpu = [
#   "torch==2.5.1+cpu",
#   "torchaudio==2.5.1+cpu",
#   "fairseq2n==0.3.0rc1",
#   "fairseq2[arrow]==0.3.0rc1",
# ]
eval = [
  "accelerate>=1.2.0",
  "bert-score>=0.3.13",
  "editdistance>=0.8.1",
  "jinja2>=3.1.3",
  "nltk>=3.9.1",
  "rouge-score>=0.1.2",
  "sacrebleu>=2.4.3",
  "scikit-learn>=1.5.2",
  "spacy>=3.7.5",
  "textdescriptives>=2.8.2",
  "tiktoken>=0.8.0",
  "transformers>=4.45.0",
  "fairscale>=0.4.13",
]
data = [
  "numpy>=1.21",
  "numba>=0.60.0",
  "spacy>=3.7.5",
  "en_core_web_sm@https://github.com/explosion/spacy-models/releases/download/en_core_web_sm-3.7.1/en_core_web_sm-3.7.1-py3-none-any.whl",
  "sacremoses>=0.1.1",
  "nltk>=3.8.1",
  "scipy>=1.14",
  "editdistance>=0.8.1",
  "sacrebleu>=2.4.1",
  "datasets>=2.18.0",
  "wtpsplit>=2.1.0",
  "transformers>=4.45.0",
]


[tool.ruff]
target-version = "py310"

[tool.mypy]
python_version = "3.10"
show_error_codes = true
check_untyped_defs = true
ignore_missing_imports = true
implicit_optional = true
implicit_reexport = true

files = [
  "lcm/", # TODO
]

[tool.uv]
prerelease = "explicit" # for fairseq2 0.3.0rc1

# TODO Change versions
[tool.uv.sources]
fairseq2 = [
  { index = "fairseq2-gpu", extra = 'gpu' },
  # { index = "fairseq2-cpu", extra = 'cpu' }
]
fairseq2n = [
  { index = "fairseq2-gpu", extra = 'gpu' },
  # { index = "fairseq2-cpu", extra = 'cpu' }
]
torch={index="pytorch-gpu"}
torchaudio={index="pytorch-gpu"}
# torch={index="pytorch-cpu"}
# torchaudio={index="pytorch-cpu"}
# sonar-space = { git = "https://github.com/facebookresearch/SONAR", branch = "update_fs2" } # TODO

[[tool.uv.index]]
name = "fairseq2-gpu"
url = "https://fair.pkg.atmeta.com/fairseq2/whl/rc/pt2.5.1/cu121"
explicit = true

# [[tool.uv.index]]
# name = "fairseq2-cpu"
# url = "https://fair.pkg.atmeta.com/fairseq2/whl/rc/pt2.5.1/cpu/"
# explicit = true

[[tool.uv.index]]
name = "pytorch-gpu"
url = "https://download.pytorch.org/whl/cu121"
explicit = true

# [[tool.uv.index]]
# name = "pytorch-cpu"
# url = "https://download.pytorch.org/whl/cpu"
# explicit = true

[dependency-groups]
dev = ["pytest-asyncio>=0.23.2", "pytest>=8.0.0"]


[project.entry-points."fairseq2"]
"fairseq2" = "lcm:setup_fairseq2" # TODO update lcm