Skip to content

instance_sampling

UniformInstanceAttributeSampling

Bases: OutliersGenerator

Randomly generates outliers by sampling from existing instances attributes uniformly at random

Source code in badgers/generators/tabular_data/outliers/instance_sampling.py 
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
class UniformInstanceAttributeSampling(OutliersGenerator):
    """
    Randomly generates outliers by sampling from existing instances attributes uniformly at random
    """

    def __init__(self, random_generator=default_rng(seed=0)):
        """
        Initialize the UniformInstanceAttributeSampling with a random number generator.

        :param random_generator: An instance of numpy's random number generator (default is a new generator with seed 0).
        """
        super().__init__(random_generator)

    @preprocess_inputs
    def generate(self, X, y, n_outliers: int = 10):
        """


        :param X: the input features (pandas DataFrame or numpy array).
        :param y: the class labels, target values, or None (if not provided).
        :param n_outliers: The number of outliers to generate.
        :return: A tuple containing the augmented feature matrix with added outliers and the corresponding target values.
                 If `y` is None, the returned target values will also be None.
        """
        assert n_outliers > 0

        # Vectorized: pick random row indices for each (outlier, column) pair,
        # then index into the underlying numpy array directly (avoids expensive
        # .iloc per-column slicing in a Python loop).
        row_indices = self.random_generator.integers(0, len(X), size=(n_outliers, X.shape[1]))
        outlier_values = X[row_indices, np.arange(X.shape[1])]

        outliers = outlier_values

        # Append outliers to original data
        Xt = np.vstack([X, outliers])

        # Build yt labels
        n_samples = len(X)
        if y is None:
            yt = np.array(["original"] * n_samples + ["outliers"] * n_outliers)
        else:
            yt = np.append(np.asarray(y, dtype=str), ["outliers"] * n_outliers)

        return Xt, yt

__init__(random_generator=default_rng(seed=0))

Initialize the UniformInstanceAttributeSampling with a random number generator.

Parameters:

Name Type Description Default
random_generator

An instance of numpy's random number generator (default is a new generator with seed 0).

 default_rng(seed=0)
Source code in badgers/generators/tabular_data/outliers/instance_sampling.py 
13
14
15
16
17
18
19
def __init__(self, random_generator=default_rng(seed=0)):
    """
    Initialize the UniformInstanceAttributeSampling with a random number generator.

    :param random_generator: An instance of numpy's random number generator (default is a new generator with seed 0).
    """
    super().__init__(random_generator)

generate(X, y, n_outliers=10)

Parameters:

Name Type Description Default
X

the input features (pandas DataFrame or numpy array).

 required
y

the class labels, target values, or None (if not provided).

 required
n_outliers int

The number of outliers to generate.

 10

Returns:

Type Description

A tuple containing the augmented feature matrix with added outliers and the corresponding target values. If y is None, the returned target values will also be None.
Source code in badgers/generators/tabular_data/outliers/instance_sampling.py 
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
@preprocess_inputs
def generate(self, X, y, n_outliers: int = 10):
    """


    :param X: the input features (pandas DataFrame or numpy array).
    :param y: the class labels, target values, or None (if not provided).
    :param n_outliers: The number of outliers to generate.
    :return: A tuple containing the augmented feature matrix with added outliers and the corresponding target values.
             If `y` is None, the returned target values will also be None.
    """
    assert n_outliers > 0

    # Vectorized: pick random row indices for each (outlier, column) pair,
    # then index into the underlying numpy array directly (avoids expensive
    # .iloc per-column slicing in a Python loop).
    row_indices = self.random_generator.integers(0, len(X), size=(n_outliers, X.shape[1]))
    outlier_values = X[row_indices, np.arange(X.shape[1])]

    outliers = outlier_values

    # Append outliers to original data
    Xt = np.vstack([X, outliers])

    # Build yt labels
    n_samples = len(X)
    if y is None:
        yt = np.array(["original"] * n_samples + ["outliers"] * n_outliers)
    else:
        yt = np.append(np.asarray(y, dtype=str), ["outliers"] * n_outliers)

    return Xt, yt