Skip to content

abs

Author: Heli Qi Affiliation: NAIST Date: 2022.07

Criterion

Bases: ABC

Criterion is a Callable object which is the base class for all criterion objects in this toolkit. It serves the role of evaluating the model forward calculation results. Its output can be either a loss function used for training or an evaluation metric used for validation.

This base class has two abstract interface functions: criterion_init() for criterion initialization and call() for criterion forward calculation. 1. call() must be overridden if you want to make your own Criterion implementation. 2. criterion_init() is not mandatory to be overridden because some criteria can directly be applied to the input data without any initialization such as speechain.criterion.accuracy.Accuracy.

Source code in speechain/criterion/abs.py 
10
11
12
13
14
15
16
17
18
19
20
21
22
23
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
class Criterion(ABC):
    """Criterion is a Callable object which is the base class for all criterion objects
    in this toolkit. It serves the role of evaluating the model forward calculation
    results. Its output can be either a loss function used for training or an evaluation
    metric used for validation.

    This base class has two abstract interface functions: criterion_init() for criterion initialization and __call__()
    for criterion forward calculation.
    1. __call__() must be overridden if you want to make your own Criterion implementation.
    2. criterion_init() is not mandatory to be overridden because some criteria can directly be applied to the input data
        without any initialization such as speechain.criterion.accuracy.Accuracy.
    """

    def __init__(self, **criterion_conf):
        """This initialization function is shared by all Criterion subclasses.
        Currently, the shared logic only contains calling the initialization function of
        the parent class.

        Args:
            **criterion_conf:
                The arguments used by criterion_init() for your customized Criterion initialization.
        """
        super(Criterion, self).__init__()
        self.criterion_init(**criterion_conf)

    def criterion_init(self, **criterion_conf):
        """Abstract interface function for customized initialization of each Criterion
        subclass. This interface function is not mandatory to be overridden by your
        implementation.

        Args:
            **criterion_conf:
                The arguments used for customized Criterion initialization.
                For more details, please refer to the docstring of your target Criterion subclass.
        """
        pass

    @abstractmethod
    def __call__(self, **kwargs):
        """This abstract interface function receives the model forward calculation
        results and ground-truth labels. The output is a scalar which could be either
        trainable for parameter optimization or non- trainable for information
        recording.

        This interface function is mandatory to be overridden by your implementation.

        Args:
            **kwargs:
                model forward calculation results and ground-truth labels.
                For more details, please refer to the docstring of __call__() of your target Criterion subclass.

        Returns:
            A trainable or non-trainable scalar.
            For more details, please refer to the docstring of __call__() of your target Criterion subclass.
        """
        raise NotImplementedError

__call__(**kwargs) abstractmethod

This abstract interface function receives the model forward calculation results and ground-truth labels. The output is a scalar which could be either trainable for parameter optimization or non- trainable for information recording.

This interface function is mandatory to be overridden by your implementation.

Parameters:

Name Type Description Default
**kwargs

model forward calculation results and ground-truth labels. For more details, please refer to the docstring of call() of your target Criterion subclass.

 {}

Returns:

Type Description

A trainable or non-trainable scalar.

For more details, please refer to the docstring of call() of your target Criterion subclass.
Source code in speechain/criterion/abs.py 
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
@abstractmethod
def __call__(self, **kwargs):
    """This abstract interface function receives the model forward calculation
    results and ground-truth labels. The output is a scalar which could be either
    trainable for parameter optimization or non- trainable for information
    recording.

    This interface function is mandatory to be overridden by your implementation.

    Args:
        **kwargs:
            model forward calculation results and ground-truth labels.
            For more details, please refer to the docstring of __call__() of your target Criterion subclass.

    Returns:
        A trainable or non-trainable scalar.
        For more details, please refer to the docstring of __call__() of your target Criterion subclass.
    """
    raise NotImplementedError

__init__(**criterion_conf)

This initialization function is shared by all Criterion subclasses. Currently, the shared logic only contains calling the initialization function of the parent class.

Parameters:

Name Type Description Default
**criterion_conf

The arguments used by criterion_init() for your customized Criterion initialization.

 {}
Source code in speechain/criterion/abs.py 
23
24
25
26
27
28
29
30
31
32
33
def __init__(self, **criterion_conf):
    """This initialization function is shared by all Criterion subclasses.
    Currently, the shared logic only contains calling the initialization function of
    the parent class.

    Args:
        **criterion_conf:
            The arguments used by criterion_init() for your customized Criterion initialization.
    """
    super(Criterion, self).__init__()
    self.criterion_init(**criterion_conf)

criterion_init(**criterion_conf)

Abstract interface function for customized initialization of each Criterion subclass. This interface function is not mandatory to be overridden by your implementation.

Parameters:

Name Type Description Default
**criterion_conf

The arguments used for customized Criterion initialization. For more details, please refer to the docstring of your target Criterion subclass.

 {}
Source code in speechain/criterion/abs.py 
35
36
37
38
39
40
41
42
43
44
45
def criterion_init(self, **criterion_conf):
    """Abstract interface function for customized initialization of each Criterion
    subclass. This interface function is not mandatory to be overridden by your
    implementation.

    Args:
        **criterion_conf:
            The arguments used for customized Criterion initialization.
            For more details, please refer to the docstring of your target Criterion subclass.
    """
    pass