Skip to content

glassbox.models.neighbors._knn

KNeighborsClassifier and KNeighborsRegressor.

BaseKNN 

BaseKNN(k=5, metric=EUCLIDEAN, algorithm=BRUTE_FORCE)

Bases: BaseModel

Initialize the BaseKNN estimator.

Parameters:

Name Type Description Default
k int

Number of neighbors to use.

 5
metric DistanceMetric

Distance metric to compute distances.

 DistanceMetric.EUCLIDEAN
algorithm SearchAlgorithm

Algorithm used to compute the nearest neighbors.

 SearchAlgorithm.BRUTE_FORCE
Source code in glassbox/models/neighbors/_knn.py 
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
def __init__(
    self,
    k: int = 5,
    metric: DistanceMetric = DistanceMetric.EUCLIDEAN,
    algorithm: SearchAlgorithm = SearchAlgorithm.BRUTE_FORCE,
) -> None:
    """
    Initialize the BaseKNN estimator.

    Parameters
    ----------
    k : int, default=5
        Number of neighbors to use.
    metric : DistanceMetric, default=DistanceMetric.EUCLIDEAN
        Distance metric to compute distances.
    algorithm : SearchAlgorithm, default=SearchAlgorithm.BRUTE_FORCE
        Algorithm used to compute the nearest neighbors.
    """
    self.k: int = k
    self.metric: DistanceMetric = metric
    self.algorithm: SearchAlgorithm = algorithm
    self.index: BaseIndex | None = None
    self.y_train: np.ndarray | None = None

fit 

fit(X, y)

Fits the model to the training data.

Parameters:

Name Type Description Default
X ndarray

Training data of shape (n_samples, n_features).

 required
y ndarray

Target values of shape (n_samples,).

 required

Returns:

Type Description
Self

The fitted model.
Source code in glassbox/models/neighbors/_knn.py 
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
def fit(self, X: np.ndarray, y: np.ndarray) -> Self:
    """
    Fits the model to the training data.

    Parameters
    ----------
    X : np.ndarray
        Training data of shape (n_samples, n_features).
    y : np.ndarray
        Target values of shape (n_samples,).

    Returns
    -------
    Self
        The fitted model.
    """
    X_arr = np.asarray(X)
    self.y_train = np.asarray(y)

    if self.algorithm == SearchAlgorithm.BRUTE_FORCE:
        self.index = BruteForceIndex(metric=self.metric)
    elif self.algorithm == SearchAlgorithm.KD_TREE:
        self.index = KDTreeIndex(metric=self.metric)
    else:
        raise ValueError(f"Unsupported algorithm: {self.algorithm}")

    self.index.build(X_arr)
    return self

predict 

predict(X, **kwargs)

Predicts target values for the given data.

Parameters:

Name Type Description Default
X ndarray

Data to predict on, of shape (n_samples, n_features) or (n_features,).

 required
**kwargs Any

Additional keyword arguments.

 {}

Returns:

Type Description
ndarray

Predicted target values.
Source code in glassbox/models/neighbors/_knn.py 
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
def predict(self, X: np.ndarray, **kwargs: Any) -> np.ndarray:
    """
    Predicts target values for the given data.

    Parameters
    ----------
    X : np.ndarray
        Data to predict on, of shape (n_samples, n_features) or (n_features,).
    **kwargs : Any
        Additional keyword arguments.

    Returns
    -------
    np.ndarray
        Predicted target values.
    """
    if self.index is None or self.y_train is None:
        raise ValueError("Model must be fitted before calling predict.")

    X_arr = np.asarray(X)
    single_query = X_arr.ndim == 1

    nearest_indices = self.index.query(X_arr, self.k)

    if single_query:
        nearest_y = self.y_train[nearest_indices].reshape(1, -1)
    else:
        nearest_y = self.y_train[nearest_indices]

    preds = self._aggregate(nearest_y)

    if single_query:
        return preds[0]
    return preds

KNeighborsClassifier 

KNeighborsClassifier(
    k=5, metric=EUCLIDEAN, algorithm=BRUTE_FORCE
)

Bases: BaseKNN

Source code in glassbox/models/neighbors/_knn.py 
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
def __init__(
    self,
    k: int = 5,
    metric: DistanceMetric = DistanceMetric.EUCLIDEAN,
    algorithm: SearchAlgorithm = SearchAlgorithm.BRUTE_FORCE,
) -> None:
    """
    Initialize the BaseKNN estimator.

    Parameters
    ----------
    k : int, default=5
        Number of neighbors to use.
    metric : DistanceMetric, default=DistanceMetric.EUCLIDEAN
        Distance metric to compute distances.
    algorithm : SearchAlgorithm, default=SearchAlgorithm.BRUTE_FORCE
        Algorithm used to compute the nearest neighbors.
    """
    self.k: int = k
    self.metric: DistanceMetric = metric
    self.algorithm: SearchAlgorithm = algorithm
    self.index: BaseIndex | None = None
    self.y_train: np.ndarray | None = None

KNeighborsRegressor 

KNeighborsRegressor(
    k=5, metric=EUCLIDEAN, algorithm=BRUTE_FORCE
)

Bases: BaseKNN

Source code in glassbox/models/neighbors/_knn.py 
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
def __init__(
    self,
    k: int = 5,
    metric: DistanceMetric = DistanceMetric.EUCLIDEAN,
    algorithm: SearchAlgorithm = SearchAlgorithm.BRUTE_FORCE,
) -> None:
    """
    Initialize the BaseKNN estimator.

    Parameters
    ----------
    k : int, default=5
        Number of neighbors to use.
    metric : DistanceMetric, default=DistanceMetric.EUCLIDEAN
        Distance metric to compute distances.
    algorithm : SearchAlgorithm, default=SearchAlgorithm.BRUTE_FORCE
        Algorithm used to compute the nearest neighbors.
    """
    self.k: int = k
    self.metric: DistanceMetric = metric
    self.algorithm: SearchAlgorithm = algorithm
    self.index: BaseIndex | None = None
    self.y_train: np.ndarray | None = None