Reducer#

class deeptrack.statistics.Reducer(function: Callable, feature: Feature | None = None, distributed: bool = True, **kwargs: Any)#

Base class that reduce input dimensionality with a statistical function.

Parameters#

function: Callable: The function used to reduce the input.
feature: Feature, optional: If not None, the output of this feature is used as the input.
distributed: bool: Whether to apply the reducer to each image in the input list individually.
axis: int or tuple of int: The axis / axes to reduce over.
keepdims: bool: Whether to keep the singleton dimensions after reducing or squeezing them.
**kwargs: Additional keyword arguments passed to the parent class and the function.

The distributed keyword is passed to the parent class to determine
how to handle the input list of images. If distributed is True, the reducer will be applied to each image in the list individually. If False, the reducer will be applied to the entire list as a single array.

Methods Summary

get(image, axis[, keepdims])

Apply the reduction function to the input image.

Methods Documentation

Apply the reduction function to the input image.

image: array-like or list/tuple of array-like: The input image or list/tuple of images to reduce.
axis: int or None: The axis or axes along which the reduction is performed. If None, the reduction is performed over all axes.
keepdims: bool or None: Whether to keep the singleton dimensions after reducing or squeezing them. If None, the default behavior of the reduction function is used.
**kwargs: Additional keyword arguments passed to the parent class and the reduction function.