pyproject.toml

[tool.poetry]
name = "machine_learning_engineering"
version = "0.1"
description = "MLE-STAR: A novel machine learning engineering (MLE) agent to automate the implementation of ML models."
authors = ["Jiefeng Chen <jiefengc@google.com>", "Raj Sinha <sinharaj@google.com", "Jaehyun Nam <jaehyunnam@google.com>", "Jinsung Yoon <jinsungyoon@google.com>"]
license = "Apache License 2.0"
readme = "README.md"

[tool.poetry.dependencies]
python = "^3.12"
google-adk = "^1.5.0"
google-genai = "^1.9.0"
pydantic = "^2.10.6"
python-dotenv = "^1.0.1"
google-cloud-aiplatform = { version = "^1.93", extras = [
    "adk",
    "agent-engines",
] }
numpy = "^2.2.3"
pandas = "^2.3.1"
scikit-learn = "^1.7.1"
# The CPU version of PyTorch is installed here. Change this is you need GPUs, or
# want to use PyTorch on a different hardware architecture or CPU.
# (https://download.pytorch.org/whl)
torch = [
  {url = "https://download.pytorch.org/whl/cpu-cxx11-abi/torch-2.7.1%2Bcpu.cxx11.abi-cp312-cp312-linux_x86_64.whl", markers = "sys_platform == 'linux'"},
]
[tool.poetry.group.dev]
optional = true

[tool.poetry.group.dev.dependencies]
google-adk = { extras = ["eval"], version = "^1.5.0" }
google-cloud-aiplatform = { version = "^1.93", extras = [
    "adk",
    "agent-engines",
    "evaluation",
] }
pytest = "^8.3.5"
black = "^25.1.0"
pytest-asyncio = "^0.26.0"
numpy = "^2.2.3"
pandas = "^2.3.1"
scikit-learn = "^1.7.1"
# The CPU version of PyTorch is installed here. Change this is you need GPUs, or
# want to use PyTorch on a different hardware architecture or CPU.
# (https://download.pytorch.org/whl)
torch = [
  {url = "https://download.pytorch.org/whl/cpu-cxx11-abi/torch-2.7.1%2Bcpu.cxx11.abi-cp312-cp312-linux_x86_64.whl", markers = "sys_platform == 'linux'"},
]

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

[tool.poetry.group.deployment.dependencies]
absl-py = "^2.2.1"

[tool.pytest.ini_options]
asyncio_mode = "auto"

[build-system]
requires = ["poetry-core>=2.0.0,<3.0.0"]
build-backend = "poetry.core.masonry.api"