alibi.api.interfaces

Constants

logger

logger: logging.Logger = <Logger alibi.api.interfaces (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.

AlibiPrettyPrinter

Inherits from: PrettyPrinter

Overrides the built in dictionary pretty representation to look more similar to the external prettyprinter libary.

Constructor

AlibiPrettyPrinter(self, *args, **kwargs)

Base

Base class for all alibi algorithms. Implements a structured approach to handle metadata.

Constructor

Base(self, meta: dict = NOTHING) -> None

Name	Type	Default	Description
meta	dict	NOTHING

Explainer

Inherits from: ABC, Base

Base class for explainer algorithms from :py:mod:alibi.explainers.

Methods

explain

explain(X: typing.Any) -> alibi.api.interfaces.Explanation

Name	Type	Default	Description
X	typing.Any

Returns

• Type: alibi.api.interfaces.Explanation

load

load(path: Union[str, os.PathLike], predictor: typing.Any) -> alibi.api.interfaces.Explainer

Name	Type	Default	Description
path	Union[str, os.PathLike]		Path to a directory containing the saved explainer.
predictor	typing.Any		Model or prediction function used to originally initialize the explainer.

Returns

• Type: alibi.api.interfaces.Explainer

reset_predictor

reset_predictor(predictor: typing.Any) -> None

Name	Type	Default	Description
predictor	typing.Any		New predictor.

Returns

• Type: None

save

save(path: Union[str, os.PathLike]) -> None

Name	Type	Default	Description
path	Union[str, os.PathLike]		Path to a directory. A new directory will be created if one does not exist.

Returns

• Type: None

Explanation

Explanation class returned by explainers.

Constructor

Explanation(self, meta: dict, data: dict) -> None

Name	Type	Default	Description
meta	dict
data	dict

Methods

from_json

from_json(jsonrepr) -> alibi.api.interfaces.Explanation

Name	Type	Default	Description
jsonrepr			json representation of an explanation.

Returns

• Type: alibi.api.interfaces.Explanation

to_json

to_json() -> str

Serialize the explanation data and metadata into a json format.

Returns

• Type: str

FitMixin

Inherits from: ABC

Methods

fit

fit(X: typing.Any) -> alibi.api.interfaces.Explainer

Name	Type	Default	Description
X	typing.Any

Returns

• Type: alibi.api.interfaces.Explainer

Summariser

Inherits from: ABC, Base

Base class for prototype algorithms from :py:mod:alibi.prototypes.

Methods

load

load(path: Union[str, os.PathLike]) -> alibi.api.interfaces.Summariser

Name	Type	Default	Description
path	Union[str, os.PathLike]

Returns

• Type: alibi.api.interfaces.Summariser

save

save(path: Union[str, os.PathLike]) -> None

Name	Type	Default	Description
path	Union[str, os.PathLike]

Returns

• Type: None

summarise

summarise(num_prototypes: int) -> alibi.api.interfaces.Explanation

Name	Type	Default	Description
num_prototypes	int

Returns

• Type: alibi.api.interfaces.Explanation

Functions

default_meta

default_meta() -> dict

Returns

• Type: dict