feature: release

.gitignore

@@ -1,3 +1,7 @@
 __pycache__
 .vscode
 .idea
+
+/build
+/dist
+*.egg-info

easytorch/__init__.py

@@ -1,3 +1,4 @@
 from .config import import_config
 from .core import *
 from .utils import *
+from .version import __version__

easytorch/core/launcher.py

@@ -1,5 +1,4 @@
 import os
-import random
 from typing import Callable
 
 import torch
@@ -39,7 +38,8 @@ def train(cfg: dict, use_gpu: bool, tf32_mode: bool):
     runner.train(cfg)
 
 
-def train_ddp(local_rank: int, world_size: int, backend: str or Backend, init_method: str, cfg: dict, tf32_mode: bool, node_rank: int = 0):
+def train_ddp(local_rank: int, world_size: int, backend: str or Backend, init_method: str, cfg: dict, tf32_mode: bool,
+              node_rank: int = 0):
     """Start training with DistributedDataParallel
 
     Args:

easytorch/train.py

@@ -0,0 +1,30 @@
+import os
+import sys
+from argparse import ArgumentParser
+
+from easytorch import launch_training
+
+
+def parse_args():
+    parser = ArgumentParser(description='Welcome to EasyTorch!')
+    parser.add_argument('-c', '--cfg', help='training config', required=True)
+    parser.add_argument('--node-rank', default=0, type=int, help='node rank for distributed training')
+    parser.add_argument('--gpus', help='visible gpus', type=str)
+    parser.add_argument('--tf32', help='enable tf32 on Ampere device', action='store_true')
+    return parser.parse_args()
+
+
+def main():
+    # work dir
+    path = os.getcwd()
+    sys.path.append(path)
+
+    # parse arguments
+    args = parse_args()
+
+    # train
+    launch_training(args.cfg, args.gpus, args.tf32, args.node_rank)
+
+
+if __name__ == '__main__':
+    main()

easytorch/version.py

@@ -0,0 +1,2 @@
+__version__ = '1.0.0'
+__all__ = ['__version__']

requirements.txt

@@ -1,3 +1,4 @@
+torch>=1.7
 easydict
 tensorboard
 tqdm

setup.py

@@ -0,0 +1,54 @@
+import os
+from setuptools import find_packages, setup
+
+
+def readme():
+    with open('README.md', encoding='utf-8') as f:
+        content = f.read()
+    return content
+
+
+def get_version():
+    version_file = 'easytorch/version.py'
+    with open(version_file, 'r', encoding='utf-8') as f:
+        exec(compile(f.read(), version_file, 'exec'))
+    return locals()['__version__']
+
+
+def get_requirements(filename='requirements.txt'):
+    here = os.path.dirname(os.path.realpath(__file__))
+    with open(os.path.join(here, filename), 'r') as f:
+        requires = [line.replace('\n', '') for line in f.readlines()]
+    return requires
+
+
+if __name__ == '__main__':
+    setup(
+        name='easytorch',
+        version=get_version(),
+        description='Simple and powerful pytorch framework.',
+        long_description=readme(),
+        long_description_content_type='text/markdown',
+        author='Yuhao Wang',
+        author_email='yuhaow97@gmail.com',
+        keywords='pytorch, deep learning',
+        url='https://github.com/cnstark/easytorch',
+        include_package_data=True,
+        packages=find_packages(exclude=('tests',)),
+        classifiers=[
+            'Development Status :: 4 - Beta',
+            'License :: OSI Approved :: Apache Software License',
+            'Operating System :: OS Independent',
+            'Programming Language :: Python :: 3',
+            'Programming Language :: Python :: 3.7',
+            'Programming Language :: Python :: 3.8',
+            'Programming Language :: Python :: 3.9',
+            'Topic :: Utilities'
+        ],
+        entry_points={
+            'console_scripts': ['easytrain=easytorch.train:main'],
+        },
+        license='Apache License 2.0',
+        install_requires=get_requirements(),
+        zip_safe=False
+    )