alibi_detect.base module

class alibi_detect.base.BaseDetector[source]

Bases: ABC

Base class for outlier, adversarial and drift detection algorithms.

property meta: Dict

Return type:: Dict

abstract predict(X)[source]

abstract score(X)[source]

class alibi_detect.base.ConfigurableDetector(*args, **kwargs)[source]

Bases: Detector, Protocol

Type Protocol for detectors that have support for saving via config.

Used for typing save and load functionality in alibi_detect.saving.saving.

classmethod from_config(config)[source]

get_config()[source]

Return type:: dict

class alibi_detect.base.Detector(*args, **kwargs)[source]

Bases: Protocol

Type Protocol for all detectors.

Used for typing legacy save and load functionality in alibi_detect.saving._tensorflow.saving.py.

Note:: This exists to distinguish between detectors with and without support for config saving and loading. Once all detector support this then this protocol will be removed.

meta: Dict

predict()[source]

Return type:: Any

class alibi_detect.base.DriftConfigMixin[source]

Bases: object

A mixin class containing methods related to a drift detector’s configuration dictionary.

config: Optional[dict] = None

classmethod from_config(config)[source]

Instantiate a drift detector from a fully resolved (and validated) config dictionary.

Parameters:: config (dict) – A config dictionary matching the schema’s in schemas.

get_config()[source]

Get the detector’s configuration dictionary.

Return type:: dict
Returns:: The detector’s configuration dictionary.

class alibi_detect.base.FitMixin[source]

Bases: ABC

abstract fit(*args, **kwargs)[source]

Return type:: None

class alibi_detect.base.NumpyEncoder(*, skipkeys=False, ensure_ascii=True, check_circular=True, allow_nan=True, sort_keys=False, indent=None, separators=None, default=None)[source]

Bases: JSONEncoder

default(obj)[source]

Implement this method in a subclass such that it returns a serializable object for o, or calls the base implementation (to raise a TypeError).

For example, to support arbitrary iterators, you could implement default like this:

def default(self, o):
    try:
        iterable = iter(o)
    except TypeError:
        pass
    else:
        return list(iterable)
    # Let the base class default method raise the TypeError
    return JSONEncoder.default(self, o)

class alibi_detect.base.StatefulDetectorOnline(*args, **kwargs)[source]

Bases: ConfigurableDetector, Protocol

Type Protocol for detectors that have support for save/loading of online state.

Used for typing save and load functionality in alibi_detect.saving.saving.

load_state(filepath)[source]

save_state(filepath)[source]

t: int = 0

class alibi_detect.base.ThresholdMixin[source]

Bases: ABC

abstract infer_threshold(*args, **kwargs)[source]

Return type:: None

alibi_detect.base.adversarial_correction_dict()[source]

alibi_detect.base.adversarial_prediction_dict()[source]

alibi_detect.base.concept_drift_dict()[source]

alibi_detect.base.outlier_prediction_dict()[source]