cristal.helper_classes.base module#

Base classes for outlier detection methods, plotters, inverters, incrementers, and polynomial basis functions.

Also contains the NotFittedError exception.

class cristal.helper_classes.base.BaseDetector#

Bases: ABC

Base class for outlier detection methods

assert_shape_fitted(x)#

Asserts that the input array x has the correct shape to use the model after it has been fitted.

Parameters:

x (np.ndarray) – The input data to check. Should be a 2D array with shape (N, d) where N is the number of samples and d is the number of features (the same as during fitting).

Raises:

NotFittedError – If the model has not been fitted yet.
ValueError – If the input array does not have the expected shape.

static assert_shape_unfitted(x)#

Asserts that the input array x has the correct shape to fit the model.

Parameters:: x (np.ndarray) – The input data to check. Should be a 2D array with shape (N, d) where N is the number of samples and d is the number of features.
Raises:: ValueError – If the input array does not have the expected shape.

copy()#

Returns a copy of the model.

Return type:: Self
Returns:: A copy of the current model.
Return type:: Self

abstractmethod decision_function(x)#

Computes the decision function for the samples in x.

Return type:: ndarray
Parameters:: x (np.ndarray (L, d)) – The input data to compute the decision function.
Returns:: The decision function values for each sample.
Return type:: np.ndarray (L,)

abstractmethod eval_update(x)#

Evaluates the model on the samples in x and updates it with inliers. This method iterates over each sample in x.

Return type:: ndarray
Parameters:: x (np.ndarray (L, d)) – The input data to evaluate and update the model.
Returns:: The decision function values for each sample.
Return type:: np.ndarray (L,)

abstractmethod fit(x)#

Generates a model that fit dataset x.

Return type:: Self
Parameters:: x (np.ndarray (N, d)) – The input data to fit the model.
Returns:: The fitted model.
Return type:: Self

fit_predict(x)#

Fits the model with x and predicts the outlier labels for the samples in x.

Return type:: ndarray
Parameters:: x (np.ndarray (N, d)) – The input data to fit and predict.
Returns:: The predicted labels for each sample.
Return type:: np.ndarray (N,)

abstractmethod is_fitted()#

Checks if the model is fitted.

Return type:: bool
Returns:: True if the model is fitted, False otherwise.
Return type:: bool

abstractmethod load_model(model_dict)#

Loads the model from a dictionary.

Return type:: Self
Parameters:: model_dict (dict) – The dictionary containing the model parameters returned by save_model().
Returns:: The loaded model.
Return type:: Self

abstractmethod method_name()#

Returns the name of the method.

Return type:: str
Returns:: The name of the method.
Return type:: str

abstractmethod predict(x)#

Predicts the outlier labels for the samples in x.

Return type:: ndarray
Parameters:: x (np.ndarray (L, d)) – The input data to predict the labels.
Returns:: The predicted labels for each sample.
Return type:: np.ndarray (L,)

abstractmethod predict_update(x)#

Predict the model on the samples in x and updates it with inliers.

Return type:: ndarray
Parameters:: x (np.ndarray (L, d)) – The input data to evaluate and update the model.
Returns:: The predicted labels for each sample.
Return type:: np.ndarray (L,)

abstractmethod save_model()#

Saves the model as a dictionary.

Return type:: dict
Returns:: The model represented as a dictionary.
Return type:: dict

abstractmethod score_samples(x)#

Computes the outlier scores for the samples in x.

Return type:: ndarray
Parameters:: x (np.ndarray (L, d)) – The input data to compute the scores.
Returns:: The outlier scores for each sample.
Return type:: np.ndarray (L,)

abstractmethod update(x)#

Updates the current model with instances in x.

Return type:: Self
Parameters:: x (np.ndarray (N', d)) – The input data to update the model.
Returns:: The updated model.
Return type:: Self

class cristal.helper_classes.base.BaseIncrementer#

Bases: ABC

Base class for moments matrix incrementers.

abstractmethod static increment(mm, x, n, inv_class, sym=True)#

Increment the inverse moments matrix (and possibly the moments matrix).

Parameters:

mm (MomentsMatrix) – The moments matrix to increment.
x (np.ndarray (N', d)) – The input data.
n (int) – The number of points integrated in the moments matrix.
inv_class (type[BaseInverter]) – The inversion class to use.
sym (bool, optional) – Whether to consider the matrix as symmetric, by default True

update_moments_matrix: bool = False#: Whether the method updates the moments matrix during incrementing

class cristal.helper_classes.base.BaseInverter#

Bases: ABC

Base class for matrix inversion methods.

abstractmethod static invert(matrix)#

Invert the given matrix.

Return type:: ndarray
Parameters:: matrix (np.ndarray (n, n)) – The input matrix to invert.
Returns:: The inverse of the input matrix.
Return type:: np.ndarray (n, n)

class cristal.helper_classes.base.BasePlotter#

Bases: ABC

Base class for outlier detection results plotters. This requires that data is 2-dimensional.

abstractmethod boundary(x, *, n_x1, n_x2, show, save, save_title, close)#

Generates the boundary plot and optionally show, save and close.

Return type:

None | tuple[Figure, Axes]

Parameters:

x (np.ndarray) – The input data to plot, shape (n_samples, n_features=2).
n_x1 (int) – Number of points along the first dimension for the grid.
n_x2 (int) – Number of points along the second dimension for the grid.
show (bool) – Whether to show the plot.
save (bool) – Whether to save the plot to a file.
save_title (str) – Title for the saved plot file.
close (bool) – Whether to close the plot or return it after saving or showing.

Returns:

The figure and axes objects if not closing the plot, otherwise None.

Return type:

None | tuple[Figure, Axes]

abstractmethod levelset(x, *, n_x1, n_x2, levels, percentiles, show, save, save_title, close)#

Generates the levelset plot and optionally show, save and close.

Return type:

None | tuple[Figure, Axes]

Parameters:

x (np.ndarray) – The input data to plot, shape (n_samples, n_features=2).
n_x1 (int) – Number of points along the first dimension for the grid.
n_x2 (int) – Number of points along the second dimension for the grid.
levels (list[int | float] | None) – The levels to plot.
percentiles (list[int | float] | None) – The percentiles to compute and plot as levels.
show (bool) – Whether to show the plot.
save (bool) – Whether to save the plot to a file.
save_title (str) – Title for the saved plot file.
close (bool) – Whether to close the plot or return it after saving or showing.

Returns:

The figure and axes objects if not closing the plot, otherwise None.

Return type:

None | tuple[Figure, Axes]

class cristal.helper_classes.base.BasePolynomialBasis#

Bases: ABC

Base class for polynomial basis functions.

abstractmethod static func(x, monomials_matrix)#

Apply the polynomial basis function to the input data.

Return type:

ndarray

Parameters:

x (np.ndarray (1, d) or (d,)) – The input data.
monomials_matrix (np.ndarray (s(n), d)) – The monomials matrix. Should be of shape (s(n), d) where s(n) is the number of monomials and d is the dimension of the input data.

Returns:

The transformed input data.

Return type:

np.ndarray (s(n), d)

class cristal.helper_classes.base.BaseRegularizer#

Bases: ABC

Base class for regularization methods.

abstractmethod static regularizer(n, d, C)#

Compute the regularization value.

Return type:

float

Parameters:

n (int) – The polynomial basis maximum degree.
d (int) – The dimension of the input data.
C (float | int) – A constant used in the regularization computation.

Returns:

The computed regularization value.

Return type:

float

exception cristal.helper_classes.base.NotFittedError(message)#

Bases: Exception

This exception is raised when a BaseDetector is used before it has been fitted.

__init__(message)#

cristal.helper_classes.base module#

This Page