Source code for ezyrb.approximation.radius_neighbors_regressor

"""Wrapper for RadiusNeighborsRegressor."""

from sklearn.neighbors import RadiusNeighborsRegressor as Regressor

from .neighbors_regressor import NeighborsRegressor


[docs] class RadiusNeighborsRegressor(NeighborsRegressor): """ Radius Neighbors Regressor for multidimensional approximation. :param kwargs: arguments passed to the internal instance of RadiusNeighborsRegressor. :Example: >>> import numpy as np >>> from ezyrb import RadiusNeighborsRegressor >>> x = np.random.uniform(-1, 1, size=(20, 2)) >>> y = np.sin(x[:, 0]) * np.cos(x[:, 1]) >>> rnn = RadiusNeighborsRegressor(radius=0.5) >>> rnn.fit(x, y) >>> new_x = np.array([[0.0, 0.0]]) >>> y_pred = rnn.predict(new_x) """ def __init__(self, **kwargs): """ Initialize a Radius Neighbors Regressor. :param kwargs: Arguments passed to sklearn's RadiusNeighborsRegressor. """ self.regressor = Regressor(**kwargs)