setup.py

#!/usr/bin/env python3

import setuptools
import os

with open("README.md", "r") as fh:
    long_description = fh.read()

__author__ = 'Kiran Karra, Chace Ashcraft, Nat Kavaler, Michael Majurski, Taylor Kulp-McDowall'
__email__ = 'kiran.karra@jhuapl.edu,chace.ashcraft@jhuapl.edu,nathaniel.kavaler@jhaupl.edu,' \
            'michael.majurski@nist.gov,taylor.kulp-mcdowall@iarpa.gov'
__version__ = '0.2.22'
# Additional credit for software design attributed to:
#  Cash Costello: cash.costello@jhuapl.edu
#  Nathan Drenkow: nathan.drenkow@jhuapl.edu
#  Neil Fendley: neil.fendley@jhuapl.edu

on_rtd = os.environ.get('READTHEDOCS') == 'True'
if on_rtd:
    install_requires = []
else:
    install_requires = ['numpy>=1.19.2',
                        'pandas>=1.0.3',
                        'scikit-image>=0.17.2',
                        'joblib>=0.14.1',
                        'scipy>=1.5.3',
                        'pillow>=7.1.2',
                        'scikit-learn>=0.23.0',
                        'tqdm>=4.46.0',
                        'opencv-python>=4.2.0.34',
                        'torch==1.7.0',
                        'torchvision==0.8.0',
                        'torchtext==0.8.0',
                        'blend_modes',
                        'spacy>=2.2.0',
                        'cloudpickle>=1.4.1',
                        'Wand>=0.5.9',
                        'nltk>=3.5',
                        'pyllist>=0.3',
                        'albumentations',
                        'blend_modes',
                        'advertorch',
                        'nltk',
                        'pyllist',
                        'transformers==4.5.1',  # TODO: this requires a rust compiler being installed on OSX ...
                        'kaggle',
                        'tb-nightly',
                        'future'
                        ]

setuptools.setup(
    name='trojai',
    version=__version__,

    description='TrojAI model and dataset generation library',
    long_description=long_description,
    long_description_content_type="text/markdown",

    url = 'https://github.com/trojai/trojai',

    author=__author__,
    author_email=__email__,

    license='Apache License 2.0',

    python_requires='>=3',
    packages=['trojai', 'trojai.datagen', 'trojai.modelgen', 'trojai.modelgen.architectures'],

    # See https://pypi.python.org/pypi?%3Aaction=list_classifiers
    classifiers=[
        'Development Status :: 3 - Alpha',

        'Intended Audience :: Developers',
        'Intended Audience :: Science/Research',
        'Topic :: Scientific/Engineering',
        'Topic :: Scientific/Engineering :: Artificial Intelligence',

        'License :: OSI Approved :: Apache Software License',

        'Programming Language :: Python :: 3 :: Only',
    ],

    keywords='deep-learning trojan adversarial',

    # Needed for albumentations install
    dependency_links=['git+https://bitbucket.xrcs.jhuapl.edu/scm/troj/'
                      'albumentations.git@reproducible_randomness_for_augmentations#egg=albumentations'],

    # List run-time dependencies here.  These will be installed by pip when
    # your project is installed. For an analysis of "install_requires" vs pip's
    # requirements files see:
    # https://packaging.python.org/en/latest/requirements.html

    install_requires=install_requires,

    # List additional groups of dependencies here (e.g. development
    # dependencies). You can install these using the following syntax,
    # for example:
    # $ pip install -e .[dev,test]
    extras_require={
          'test': ['nose==1.3.7', 'coverage==5.0.3', 'mock']
    },

    scripts=['scripts/datagen/mnist.py',
             'scripts/datagen/mnist_utils.py',
             'scripts/datagen/mnist_badnets.py',
             'scripts/datagen/mnist_badnets2.py',
             'scripts/datagen/mnist_badnets_one_class_trigger.py',
             'scripts/modelgen/gen_and_train_mnist.py',
             'scripts/modelgen/gen_and_train_mnist_sequential.py'
             ],

    zip_safe=False
)