MMD

MMD#

class frouros.detectors.data_drift.batch.distance_based.MMD(kernel: ~typing.Callable = <function rbf_kernel>, chunk_size: int | None = None, callbacks: ~frouros.callbacks.batch.base.BaseCallbackBatch | list[~frouros.callbacks.batch.base.BaseCallbackBatch] | None = None)#

MMD (Maximum Mean Discrepancy) [gretton2012kernel] detector.

Parameters:

  • kernel (Callable) – kernel function, defaults to rbf_kernel()

  • chunk_size (Optional[int]) – chunk size value, defaults to None

  • callbacks (Optional[Union[BaseCallbackBatch, list[BaseCallbackBatch]]]) – callbacks, defaults to None

References:

[gretton2012kernel]

Gretton, Arthur, et al. “A kernel two-sample test.” The Journal of Machine Learning Research 13.1 (2012): 723-773.

Example:

>>> from functools import partial
>>> from frouros.detectors.data_drift import MMD
>>> from frouros.utils.kernels import rbf_kernel
>>> import numpy as np
>>> np.random.seed(seed=31)
>>> X = np.random.multivariate_normal(mean=[1, 1], cov=[[2, 0], [0, 2]], size=100)
>>> Y = np.random.multivariate_normal(mean=[0, 0], cov=[[2, 1], [1, 2]], size=100)
>>> detector = MMD(kernel=partial(rbf_kernel, sigma=0.5))
>>> _ = detector.fit(X=X)
>>> detector.compare(X=Y)[0]
DistanceResult(distance=0.02146955300299802)
property chunk_size: int | None#

Chunk size property.

Returns:

chunk size to use

Return type:

int

property kernel: Callable#

Kernel property.

Returns:

kernel function to use

Return type:

Callable

property X_ref: ndarray | None#

Reference data property.

Returns:

reference data

Return type:

Optional[numpy.ndarray]

property callbacks: list[BaseCallback] | None#

Callbacks property.

Returns:

callbacks

Return type:

Optional[list[BaseCallback]]

compare(X: ndarray, **kwargs: Any) → Tuple[ndarray, dict[str, Any]]#

Compare values.

Parameters:

X (numpy.ndarray) – test data

Returns:

compare result and callbacks logs

Return type:

Tuple[numpy.ndarray, dict[str, Any]]

property data_type: BaseDataType#

Data type property.

Returns:

data type

Return type:

BaseDataType

fit(X: ndarray, **kwargs: Any) → dict[str, Any]#

Fit detector.

Parameters:

  • X (numpy.ndarray) – feature data

  • kwargs (Any) – additional fit parameters

Returns:

callbacks logs

Return type:

dict[str, Any]

reset() → None#

Reset method.

property statistical_kwargs: dict[str, Any]#

Statistical kwargs property.

Returns:

statistical kwargs

Return type:

dict[str, Any]

property statistical_method: Callable#

Statistical method property.

Returns:

statistical method

Return type:

Callable

property statistical_type: BaseStatisticalType#

Statistical type property.

Returns:

statistical type

Return type:

BaseStatisticalType

Contents