New answers tagged

0

I don't have enough karma to comment on the above solution by @user20160 , so I'm posting this here. This provides the source code to implement the definition given by @user20160 for the gradient using GPR in sklearn. Here is a basic working example using an RBF kernel: gp = GaussianProcessRegressor(kernel=kernel, n_restarts_optimizer=9) gp.fit(X, y) # gets ...


0

Sklearn's Nystroem does not compute the Gram matrix itself, it returns the Feature map $\Phi$. The exact kernel matrix is approximated by $\tilde{G} = \Phi \Phi^\top$. Your code should look like this: kernel_approx = Nystroem(kernel, n_components=m) feature_matrix = kernel_approx.fit_transform(x) gram_matrix_approx = feature_matrix @ feature_matrix.T And ...


Top 50 recent answers are included