causally.scm.causal_mechanism.GaussianProcessMechanism.predict

GaussianProcessMechanism.predict(X: array) → array

Generate the effect given the observations of the parent nodes.

The effect is generated as a nonlinear function sampled from a gaussian process.

Parameters:

X (np.array, shape (num_samples, num_parents)) – Input of the RBF kernel.

Returns:

y – Causal effect sampled from the gaussian process with covariance matrix given by the RBF kernel with X as input.

Return type:

np.array, shape (num_samples)