setup.py

# -*- coding: utf-8 -*-
from setuptools import setup

__author__ = "Martin Uhrin"
__contributors__ = "Sonia Collaud"
__license__ = "GPLv3"

about = {}
with open("milad/version.py") as f:
    exec(f.read(), about)  # nosec

setup(
    name="milad",
    version=about["__version__"],
    description="Moment Invariants Local Atomic Descriptor",
    long_description=open("README.rst").read(),
    url="https://github.com/muhrin/milad.git",
    author="Martin Uhrin",
    author_email="martin.uhrin.10@ucl.ac.uk",
    license=__license__,
    classifiers=[
        "License :: OSI Approved :: GNU General Public License v3 (GPLv3)",
        "Programming Language :: Python :: 3.7",
        "Programming Language :: Python :: 3.8",
        "Programming Language :: Python :: 3.9",
        "Programming Language :: Python :: 3.10",
    ],
    keywords="machine learning, atomic descriptor, moment invariants",
    python_requires=">=3.6",
    install_requires=[
        "dscribe",
        "matplotlib",
        "miniball",
        "numba",
        "numpy",
        "rmsd",
        "torch",
        "scipy",
        "sympy",
        "scikit-learn",
        "scikit-image",
        "typing-extensions",
    ],
    extras_require={
        "dev": [
            "ipython",
            "pip",
            "pytest~=7.2",
            "pytest-cov",
            "pre-commit~=2.2",
            "twine",
        ],
        #        'functorch':
        #        ['torch==1.9.0.dev20210604', 'functorch @ git+https://github.com/zou3519/functorch.git#egg=functorch']
    },
    dependency_links=[
        "https://download.pytorch.org/whl/nightly/cu102/torch_nightly.html"
    ],
    packages=[
        "milad",
        "milad.dat",
        "milad.descriptors",
        "milad.models",
        "milad.optimisers",
        "milad.play",
        "milad.invariants",
    ],
    include_package_data=True,
    test_suite="test",
    entry_points={
        "mincepy.plugins.types": ["milad_types = milad.provides:get_mince_types"],
    },
)