aisquared.config.postprocessing package

Submodules

aisquared.config.postprocessing.BinaryClassification module

class aisquared.config.postprocessing.BinaryClassification.BinaryClassification(label_map: list, threshold: float = 0.5)[source]

Bases: BaseObject

Postprocesssing configuration object for binary classification

Example usage

>>> import aisquared
>>> my_obj = aisquared.config.postprocessing.BinaryClassification(
    ['class1', 'class2']
)
>>> my_obj.to_dict()
{'className': 'BinaryClassification',
'params': {'labelMap': ['class1', 'class2'], 'threshold': 0.5}}

property label_map

property threshold

to_dict() → dict[source]: Get the configuration object as a dictionary

aisquared.config.postprocessing.MulticlassClassification module

class aisquared.config.postprocessing.MulticlassClassification.MulticlassClassification(label_map: list)[source]

Bases: BaseObject

Postprocessing configuration object for multiclass classification

Example usage:

>>> import aisquared
>>> my_obj = aisquared.config.postprocessing.MulticlassClassification(
    ['class1', 'class2', 'class3']
)
>>> my_obj.to_dict()
{'className': 'MulticlassClassification',
'params': {'labelMap': ['class1', 'class2', 'class3']}}

property label_map

to_dict() → dict[source]: Get the configuration object as a dictionary

aisquared.config.postprocessing.ObjectDetection module

class aisquared.config.postprocessing.ObjectDetection.ObjectDetection(label_map: list, threshold: float = 0.5)[source]

Bases: BaseObject

Postprocessing configuration object for object detection

Example usage:

>>> import aisquared
>>> my_obj = aisquared.config.postprocessing.ObjectDetection(
    ['class1', 'class2', 'class3']
)
>>> my_obj.to_dict()
{'className': 'ObjectDetection',
'params': {'labelMap': ['class1', 'class2', 'class3'], 'threshold': 0.5}}

property label_map

property threshold

to_dict() → dict[source]: Get the configuration object as a dictionary

aisquared.config.postprocessing.Regression module

class aisquared.config.postprocessing.Regression.Regression(min: int | float | None = None, max: int | float | None = None, round: bool = False)[source]

Bases: BaseObject

Postprocessing configuration object for Regression

Example usage:

>>> import aisquared
>>> my_obj = aisquared.config.postprocessing.Regression(
    10,
    100
)
>>> my_obj.to_dict()
{'className': 'Regression', 'params': {'min': 10, 'max': 100, 'round': False}}

property max

property min

property round

to_dict() → dict[source]: Get the configuration object as a dictionary

Module contents

The aisquared.config.postprocessing subpackage contains objects for configuring how predictions are postprocessed.