Hello, I am trying to perform inference on the following model to get posterior distributions on some parameters of a black-box function and the length-scale of a gaussian process. The function eta(theta,x) can be any arbitrary black box function, where theta are the parameters to be inferred and x…

Sorry I’m late, yup, totally agree with @DanhPhan ’s suggestions. You can definitely use the GP covariance functions without the any of the other stuff in the GP library, like this: cov = pm.gp.cov.ExpQuad(1, ls=lengthscale) Sigma = cov(X) likelihood = pm.MvNormal('likelihood', mu=mu, cov=Sigma, o…

Adding Black box Model to GP

Questions v3

Rahul_Rameshbabu May 18, 2022, 1:55am 6

Yes, this is what I meant. Thanks!

1 Like

Topic		Replies	Views
Gaussian Process Regression level 1 inference: Re-producing Mauna Loa CO2 Example with PyMC3 Questions	10	2244	August 16, 2017
Hierarchical Time Series w/Gaussian Process: Any examples? Questions	3	2833	November 4, 2017
Using a GP as Covariance Matrix for an MvNormal Questions	4	767	June 15, 2020
Multiple (uncertain) function observations of the same Gaussian process Questions	15	6568	October 16, 2017
Gaussian Process predict over trace Questions	2	834	September 4, 2018

Adding Black box Model to GP

Related topics