[antonsoo]

ClassificationKriging.fit documents a one-dimensional target, but passes it directly to OneHotEncoder, which requires a column. Reshape the target for the encoder, retaining support for column targets.

import numpy as np
from sklearn.naive_bayes import GaussianNB
from pykrige.ck import ClassificationKriging

x = np.array([[0., 0.], [0., 1.], [1., 0.], [1., 1.]])
y = np.array([0, 0, 1, 1])
model = ClassificationKriging(
    classification_model=GaussianNB(),
    n_closest_points=2, variogram_parameters=[1., 0.],
)
model.fit(x, x, y)  # Previously: ValueError: Expected 2D array
print(model.predict(x, x))  # [0 0 1 1]

The regression covers vector and column targets with ordinary and universal kriging. On the recorded Meuse dataset (155 observations), six zero-based vector-target configurations now complete fitting, prediction, and scoring against independent ILR/ordinary-kriging calculations. The corresponding six column-target controls stay exact. The configurations use three stratified splits and seeded logistic/random-forest classifiers.

Validation: the full suite passes (60 tests) with Python 3.12 / NumPy 2.5 / scikit-learn 1.9 and Python 3.10 / NumPy 1.21.6 / scikit-learn 1.0.2. Black, isort, cython-lint, and git diff --check pass.