Probability#

class deeptrack.features.Probability(feature: Feature, probability: float | Callable[[...], float], **kwargs: Any)#

Bases: StructuralFeature

Resolve a feature with a certain probability.

This feature conditionally applies a given feature to an input based on a sampled uniform random number. If the sampled number is less than the specified probability, the feature is resolved; otherwise, the input is returned unchanged.

To resample the decision, call .update() before evaluating the feature.

Parameters#

feature: Feature: The feature to resolve conditionally.
probability: PropertyLike[float]: The probability (from 0 to 1) of resolving the feature.
**kwargs: Any: Additional keyword arguments passed to the parent StructuralFeature class.

Methods#

get(inputs, probability, random_number, **kwargs) -> Any: Resolves the feature if the sampled random number is less than the specified probability.

Examples#

>>> import deeptrack as dt

In this example, the Add feature is applied to the input image with a 70% chance.

Define a feature and wrap it with Probability:

>>> add_feature = dt.Add(value=2)
>>> probabilistic_feature = dt.Probability(add_feature, probability=0.7)

Define inputs:

>>> import numpy as np
>>>
>>> inputs = np.zeros((2, 3))

Apply the feature:

>>> probabilistic_feature.update()  # Update the random number
>>> outputs = probabilistic_feature(inputs)

With 70% probability, the output is:

>>> outputs
array([[2., 2., 2.],
    [2., 2., 2.]])

With 30% probability, it remains:

>>> outputs
array([[0., 0., 0.],
    [0., 0., 0.]])

Methods Summary

get(inputs, probability, random_number, **kwargs)

Resolve the feature if random number is less than probability.

Methods Documentation

get(inputs: Any, probability: float, random_number: float, **kwargs: Any) → Any#

Resolve the feature if random number is less than probability.

Parameters#

inputs: Any or list[Any]: The inputs to process.
probability: float: The probability (between 0 and 1) of resolving the feature.
random_number: float: A random number sampled to determine whether to resolve the feature. It is initialized when this feature is initialized. It can be updated using the .update() method.
**kwargs: Any: Additional arguments passed to the feature’s resolve() method.

Returns#

Any: The processed outputs. If the feature is resolved, this is the output of the feature; otherwise, it is the unchanged inputs.