I am in the process of experimenting and learning about Gaussian Process regression. I tried running the simple 1D example using generalized_exponential autocorrelation. Modified version of the example is:
X = np.atleast_2d([1., 3., 5., 6., 7., 8.]).T y = f(X).ravel() x = np.atleast_2d(np.linspace(0, 10, 1000)).T gp = GaussianProcess(regr='constant', corr='generalized_exponential', theta0=(1e-2,2.0), thetaL=(1e-4,1.9), thetaU=(1e-1,2.1), random_start=100) gp.fit(X, y) I get the following error: capi_return is NULL Call-back cb_calcfc_in_ *cobyla*_user__routines failed. Optimization failed. Try increasing the nugget I also tried running it using the Welch optimizer but got similar error. Am I missing something here???
------------------------------------------------------------------------------
_______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
