Skip to content

deeplabcut.pose_estimation_pytorch.data.helper

Classes:

Name Description
CombinedPropertyMeta

Combined metaclass that integrates the functionalities of both ABCMeta and
PropertyMeta

Metaclass for creating class properties in a more organized and systematic

Functions:

Name Description
class_property

Decorator to create a class property.

CombinedPropertyMeta

Bases: ABCMeta, PropertyMeta

Combined metaclass that integrates the functionalities of both ABCMeta and BasePropertyMeta.

This metaclass is useful in scenarios where a class needs to use both abstract methods (from ABCMeta) and the property definition utilities provided by BasePropertyMeta.

By using this metaclass, a class can be both an abstract class (with abstract methods and/or properties) and can also define properties in the structured manner facilitated by PropertyMeta.

Inherits: - ABCMeta: Metaclass for base classes that include abstract methods. - PropertyMeta: Metaclass that facilitates structured property definitions.

Note: When defining a class using CombinedPropertyMeta, ensure that the class also inherits from ABC to make it compatible with the ABCMeta behavior.

Source code in deeplabcut/pose_estimation_pytorch/data/helper.py 
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
class CombinedPropertyMeta(ABCMeta, PropertyMeta):
    """Combined metaclass that integrates the functionalities of both `ABCMeta` and
    `BasePropertyMeta`.

    This metaclass is useful in scenarios where a class needs to use both abstract methods (from `ABCMeta`)
    and the property definition utilities provided by `BasePropertyMeta`.

    By using this metaclass, a class can be both an abstract class (with abstract methods and/or properties)
    and can also define properties in the structured manner facilitated by `PropertyMeta`.

    Inherits:
    - ABCMeta: Metaclass for base classes that include abstract methods.
    - PropertyMeta: Metaclass that facilitates structured property definitions.

    Note:
    When defining a class using `CombinedPropertyMeta`, ensure that the class also inherits
    from `ABC` to make it compatible with the `ABCMeta` behavior.
    """

PropertyMeta

Bases: type

Metaclass for creating class properties in a more organized and systematic manner.

This metaclass allows a class to define its properties using a simple dictionary structure (properties). The dictionary keys represent the property names, while the values are tuples containing two callables: 1. The function that represents the logic of the property. 2. The function that provides the arguments for the logic function.

Usage: class MyClass(metaclass=PropertyMeta): properties = { 'property_name': (logic_function, arguments_function), # ... more properties ... }

For each property specified in the properties dictionary, the metaclass will generate a real property that uses the logic from logic_function and arguments from arguments_function.

  • properties (dict): Dictionary containing property names as keys and tuples of (logic_function, arguments_function) as values.
Source code in deeplabcut/pose_estimation_pytorch/data/helper.py 
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
71
72
73
74
75
class PropertyMeta(type):
    """Metaclass for creating class properties in a more organized and systematic
    manner.

    This metaclass allows a class to define its properties using a simple dictionary
    structure (`properties`). The dictionary keys represent the property names,
    while the values are tuples containing two callables:
    1. The function that represents the logic of the property.
    2. The function that provides the arguments for the logic function.

    Usage:
    class MyClass(metaclass=PropertyMeta):
        properties = {
            'property_name': (logic_function, arguments_function),
            # ... more properties ...
        }

    For each property specified in the `properties` dictionary, the metaclass will
    generate a real property that uses the logic from `logic_function` and
    arguments from `arguments_function`.

    Attributes:
    - properties (dict): Dictionary containing property names as keys and tuples
      of (logic_function, arguments_function) as values.
    """

    def __new__(cls, name, bases, attrs):
        if "properties" not in attrs:
            raise AttributeError(f"{name} must define a 'properties' dictionary.")
        properties = attrs.get("properties", {})
        for prop_name, (func, arg_func) in properties.items():
            attrs[prop_name] = class_property(func, arg_func)(lambda self: None)
        return super().__new__(cls, name, bases, attrs)

class_property 

class_property(func, arg_func)

Decorator to create a class property.

Parameters: - func: Callable that represents the logic of the property. - arg_func: Callable that provides the arguments for func.

Returns: - A property with the logic encapsulated in func and arguments derived from arg_func.

Source code in deeplabcut/pose_estimation_pytorch/data/helper.py 
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
def class_property(func, arg_func):
    """Decorator to create a class property.

    Parameters:
    - func: Callable that represents the logic of the property.
    - arg_func: Callable that provides the arguments for `func`.

    Returns:
    - A property with the logic encapsulated in `func` and arguments derived from `arg_func`.
    """

    def decorator_wrapper(method):
        def wrapper(self):
            return func(arg_func(self))

        return property(wrapper)

    return decorator_wrapper