sliced.datasets.
make_quadratic
(n_samples=500, n_features=10, n_informative=2, random_state=None)[source]¶Generates a dataset with a quadratic response curve.
Inputs X are independent normally distributed features. The output y is created according to the formula:
beta = np.hstack((
np.ones(n_informative), np.zeros(n_features - n_informative)))
h = np.dot(X, beta)
y(h) = 0.125 * h ** 2 + 0.5 * N(0, 1).
Out of the n_features features, only n_informative are actually used to compute y. The remaining features are independent of y. As such the central subspace is one dimensional and consists of the h axis.
Parameters: |
|
---|---|
Returns: |
|