I am trying to estimate the MAP for a logistic regression model. For simplicity, I’ve reduced the model down to only the intercept term: import numpy as np import pymc3 as pm def sigmoid(x: np.array) -> np.array: """Return the sigmoid of X.""" sigm = 1.0 / (1.0 + np.exp(-x)) return si…

Bernoulli only accept binary data like [0, 1, 0, 0, …]

Best practices for MAP estimation

nolan September 6, 2019, 3:41pm 3

Ah, maybe that’s it! Thanks for the quick reply. What distribution would you suggest for modeling probabilities (real numbers in range 0-1)?

Topic		Replies	Views
Unconsistent logp, Evaluating the results of find_MAP on Custom Likelihood Questions	3	610	August 28, 2018
Evaluating log likelihood using estimates from find_MAP results Questions	3	1360	January 25, 2020
A little change to the sample code of MAP, however Questions	10	1136	April 1, 2018
Maximum likelihood estimation of a Bayesian model Questions	11	5575	April 12, 2018
Beta Regression in pyMC3 Questions	7	5589	September 18, 2017