I am trying to make predictions using “gp.predict”, however, every time I use “gp.predict” I get a different result. The code I am using: import matplotlib.pyplot as plt import pymc3 as pm import numpy as np import theano.tensor as tt import arviz as az %config InlineBackend.figure_format = 'retin…

What do I need to do to get the same prediction every time I run the code? Your issue is that you need to pass a single set of model parameters into the kwarg for point. I’m not confident if this is actually what’s going on but I think by passing in the entire trace (which contains many samples …

Question about "gp.predict": Gaussian Processes

Questions

MCruzReyes September 2, 2021, 11:54pm 3

Thank you!

1 Like

Topic		Replies	Views
GP Predict Point Questions	9	1033	March 6, 2020
The predict function of GP Questions	10	1267	June 15, 2021
Using Marginal.predict method with GP trace InferenceData v5 gaussian_process	2	480	May 21, 2022
Understanding the Predictive Covariance of Gaussian Processes Questions	1	459	June 23, 2018
`predictt` in PyMC5? v5 gaussian_process	2	549	February 6, 2023

Question about "gp.predict": Gaussian Processes

Related topics