argparse with type annotations
Project description
TypedArgs
Strong type args.
This project is inspired by TeXitoi/structopt.
Install
From pypi
pip install typed-args
If you want to use it on python 3.5 and 3.6 please install dataclasses
:
pip install dataclasses
Usage
import argparse
from dataclasses import dataclass
from typed_args import TypedArgs, add_argument
@dataclass
class Args(TypedArgs):
data: str = add_argument(metavar='DIR', help='path to dataset')
arch: str = add_argument('-a', '--arch', metavar='ARCH', default='resnet18',
help='model architecture (default: resnet18)')
num_workers: int = add_argument('-j', '--workers', default=4, metavar='N',
help='number of data loading workers (default: 4)')
def parser_factory(self):
return argparse.ArgumentParser('PROG')
def test_args():
data = '/path/to/dataset'
arch = 'resnet50'
num_workers = 8
argv = f'{data} -a {arch} --workers {num_workers}'.split()
args = Args.from_args(argv)
assert args.arch == arch
assert args.data == data
assert args.num_workers == num_workers
if __name__ == "__main__":
test_args()
Limitation
Currently, we don't support add_group
and sub parser
.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
typed-args-0.4.0.tar.gz
(6.3 kB
view hashes)
Built Distribution
Close
Hashes for typed_args-0.4.0-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 512f9ada84a917609c1e6279d751bb8553e0b905cf273688f5c4b4b3636e6ac2 |
|
MD5 | 195f18abfc1301ce7c08666fb78280bf |
|
BLAKE2b-256 | 164a7f6df8d13de1cbe157c8d6c588699cc43d6f2a6491999c8edbf2920f4048 |