Skip to content

configclass

chanfig.configclasses

configclass

Python
configclass(cls=None)

Construct a Config in dataclass style.

This decorator creates a Config instance with the provided class attributes.

See Also

dataclass

Parameters:

Name Type Description Default

cls

Type[Any]

The class to be enhanced, provided directly if no parentheses are used.

 None

Returns:

Type Description

A modified class with Config functionalities or a decorator with bound parameters.

Examples:

Python Console Session
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
>>> @configclass
... class DataloaderConfig:
...     batch_size: int = 64
...     num_workers: int = 4
...     pin_memory: bool = True
>>> config = DataloaderConfig()
>>> print(config)
DataloaderConfig(<class 'chanfig.config.Config'>,
  ('batch_size'): 64
  ('num_workers'): 4
  ('pin_memory'): True
)
Source code in chanfig/configclasses.py
Python
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
def configclass(cls=None):
    """
    Construct a Config in [`dataclass`][dataclasses.dataclass] style.

    This decorator creates a Config instance with the provided class attributes.

    See Also:
        [`dataclass`][dataclasses.dataclass]

    Args:
        cls (Type[Any]): The class to be enhanced, provided directly if no parentheses are used.

    Returns:
        A modified class with Config functionalities or a decorator with bound parameters.

    Examples:
        >>> @configclass
        ... class DataloaderConfig:
        ...     batch_size: int = 64
        ...     num_workers: int = 4
        ...     pin_memory: bool = True
        >>> config = DataloaderConfig()
        >>> print(config)
        DataloaderConfig(<class 'chanfig.config.Config'>,
          ('batch_size'): 64
          ('num_workers'): 4
          ('pin_memory'): True
        )
    """

    warn(
        "This decorator is deprecated and may be removed in the future release. "
        "All chanfig classes will copy variable identified in `__annotations__` by default."
        "This decorator is equivalent to inheriting from `Config`.",
        PendingDeprecationWarning,
    )

    def decorator(cls: Type[Any]):
        if not issubclass(cls, Config):
            config_cls = type(cls.__name__, (Config, cls), dict(cls.__dict__))
            cls = config_cls

        return cls

    if cls is None:
        return decorator
    return decorator(cls)