Currently, I’m considering to use GP regression for a my project. I typically use following script is to train a GP regression model with PyMC3. X // features y // target with pm.Model() as model: ℓ = pm.Gamma("ℓ", alpha=2, beta=1) η = pm.HalfCauchy("η", beta=5) cov = η **…

Thanks @indian312 , Don’t need a loop, just by setting shape of length scale as X.shape[1] (number of features) we can achieve that. ℓ = pm.Gamma("ℓ", alpha=2, beta=1, shape = X.shape[1]])

Gaussian Process regression with Automatic Relevance Determination

Questions

Nadheesh February 11, 2019, 5:47am 13

Thanks @bwengals, I did not change the size of eta, but instead of a scalar I used a RV as shown in the scripts. I assumed that is better than using a scaler according to Bayesian.

Topic		Replies	Views
Gaussian Process Regression level 1 inference: Re-producing Mauna Loa CO2 Example with PyMC3 Questions	10	2244	August 16, 2017
Gaussian Process Regression-- reparameterizing to us pm.gp class Questions	2	505	October 3, 2019
GP regularization effect Questions	6	517	January 11, 2021
Multiple (uncertain) function observations of the same Gaussian process Questions	15	6568	October 16, 2017
Gaussian Processes: Sampling kernel hyperparameters? Questions	2	893	January 15, 2021

Gaussian Process regression with Automatic Relevance Determination

Related topics