alibi.confidence.model_linearity module

class alibi.confidence.model_linearity.LinearityMeasure(method='grid', epsilon=0.04, nb_samples=10, res=100, alphas=None, model_type='classifier', agg='pairwise', verbose=False)[source]

Bases: object

__init__(method='grid', epsilon=0.04, nb_samples=10, res=100, alphas=None, model_type='classifier', agg='pairwise', verbose=False)[source]
Parameters:

  • method (str) – Method for sampling. Supported methods: 'knn' | 'grid'.

  • epsilon (float) – Size of the sampling region around the central instance as a percentage of the features range.

  • nb_samples (int) – Number of samples to generate.

  • res (int) – Resolution of the grid. Number of intervals in which the feature range is discretized.

  • alphas (Optional[ndarray]) – Coefficients in the superposition.

  • agg (str) – Aggregation method. Supported values: 'global' | 'pairwise'.

  • model_type (str) – Type of task. Supported values: 'regressor' | 'classifier'.

fit(X_train)[source]
Parameters:

X_train (ndarray) – Training set.

Return type:

None

score(predict_fn, x)[source]
Parameters:

  • predict_fn (Callable) – Prediction function.

  • x (ndarray) – Instance of interest.

Return type:

ndarray

Returns:

Linearity measure.

alibi.confidence.model_linearity.infer_feature_range(X_train)[source]

Infers the feature range from the training set.

Parameters:

X_train (ndarray) – Training set.

Return type:

ndarray

Returns:

Feature range.

alibi.confidence.model_linearity.linearity_measure(predict_fn, x, feature_range=None, method='grid', X_train=None, epsilon=0.04, nb_samples=10, res=100, alphas=None, agg='global', model_type='classifier')[source]

Calculate the linearity measure of the model around an instance of interest x.

Parameters:

  • predict_fn (Callable) – Predict function.

  • x (ndarray) – Instance of interest.

  • feature_range (Union[List, ndarray, None]) – Array with min and max values for each feature.

  • method (str) – Method for sampling. Supported values: 'knn' | 'grid'.

  • X_train (Optional[ndarray]) – Training set.

  • epsilon (float) – Size of the sampling region as a percentage of the feature range.

  • nb_samples (int) – Number of samples to generate.

  • res (int) – Resolution of the grid. Number of intervals in which the features range is discretized.

  • alphas (Optional[ndarray]) – Coefficients in the superposition.

  • agg (str) – Aggregation method. Supported values: 'global' | 'pairwise'.

  • model_type (str) – Type of task. Supported values: 'regressor' | 'classifier'.

Return type:

ndarray

Returns:

Linearity measure.