alibi.confidence.trustscore

Constants

logger

logger: logging.Logger = <Logger alibi.confidence.trustscore (WARNING)>

Instances of the Logger class represent a single logging channel. A "logging channel" indicates an area of an application. Exactly how an "area" is defined is up to the application developer. Since an application can have any number of areas, logging channels are identified by a unique string. Application areas can be nested (e.g. an area of "input processing" might include sub-areas "read CSV files", "read XLS files" and "read Gnumeric files"). To cater for this natural nesting, channel names are organized into a namespace hierarchy where levels are separated by periods, much like the Java or Python package namespace. So in the instance given above, channel names might be "input" for the upper level, and "input.csv", "input.xls" and "input.gnu" for the sub-levels. There is no arbitrary limit to the depth of nesting.

TrustScore

Constructor

TrustScore(self, k_filter: int = 10, alpha: float = 0.0, filter_type: Optional[str] = None, leaf_size: int = 40, metric: str = 'euclidean', dist_filter_type: str = 'point') -> None

Name	Type	Default	Description
k_filter	int	10	Number of neighbors used during either kNN distance or probability filtering.
alpha	float	0.0	Fraction of instances to filter out to reduce impact of outliers.
filter_type	Optional[str]	None	Filter method: 'distance_knn'
leaf_size	int	40	Number of points at which to switch to brute-force. Affects speed and memory required to build trees. Memory to store the tree scales with n_samples / leaf_size.
metric	str	'euclidean'	Distance metric used for the tree. See sklearn DistanceMetric class for a list of available metrics.
dist_filter_type	str	'point'	Use either the distance to the k-nearest point (dist_filter_type = 'point') or the average distance from the first to the k-nearest point in the data (dist_filter_type = 'mean').

Methods

filter_by_distance_knn

filter_by_distance_knn(X: numpy.ndarray) -> numpy.ndarray

Name	Type	Default	Description
X	numpy.ndarray		Data.

Returns

• Type: numpy.ndarray

filter_by_probability_knn

filter_by_probability_knn(X: numpy.ndarray, Y: numpy.ndarray) -> Tuple[numpy.ndarray, numpy.ndarray]

Name	Type	Default	Description
X	numpy.ndarray		Data.
Y	numpy.ndarray		Predicted class labels.

Returns

• Type: Tuple[numpy.ndarray, numpy.ndarray]

fit

fit(X: numpy.ndarray, Y: numpy.ndarray, classes: Optional[int] = None) -> None

Name	Type	Default	Description
X	numpy.ndarray		Data.
Y	numpy.ndarray		Target labels, either one-hot encoded or the actual class label.
classes	Optional[int]	None	Number of prediction classes, needs to be provided if Y equals the predicted class.

Returns

• Type: None

score

score(X: numpy.ndarray, Y: numpy.ndarray, k: int = 2, dist_type: str = 'point') -> Tuple[numpy.ndarray, numpy.ndarray]

Name	Type	Default	Description
X	numpy.ndarray		Instances to calculate trust score for.
Y	numpy.ndarray		Either prediction probabilities for each class or the predicted class.
k	int	2	Number of nearest neighbors used for distance calculation.
dist_type	str	'point'	Use either the distance to the k-nearest point (dist_type = 'point') or the average distance from the first to the k-nearest point in the data (dist_type = 'mean').

Returns

• Type: Tuple[numpy.ndarray, numpy.ndarray]