Hi, I am implementing non-stationary GP regression using Gibbs kernel. The lengthscale and signal covariance are also GP. with pm.Model() as model: # Mean mean = pm.gp.mean.Zero() # Noise noise_obs = pm.Normal('noise_obs', mu=0, sd=1) # Prior for hyperparameters l_l = pm.Normal('l_l', mu=…

Non Stationary GP Regression

rohanchauhan July 4, 2018, 3:58pm 5

Thanks for your reply! I will check it and get back with results!

Topic		Replies	Views
Behaviour of Gibbs kernel seems off Questions	6	849	May 8, 2022
Multiple (uncertain) function observations of the same Gaussian process Questions	15	6568	October 16, 2017
GP and Bayesian linear regression in PyMC Questions	8	1434	August 18, 2018
GP.Marginal_likelihood without noise Questions	2	561	June 18, 2018
GP with multiplicative factor Questions	4	612	August 31, 2018