In lucas-castillo/SampleR: Tools To Compare Human Performance To Sampling Algorithms

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

library(SampleR)

Tuning epsilon

Make pd_func

target <- distr::Norm(0,3)
pd_func <- make_distr_pdf(target)

Epsilon too small

epsilon <- .01
set.seed(1)
hmc_results <- sampler_hmc(pd_func, epsilon = epsilon, L = 10, start = 0)
hist(hmc_results[[1]], probability = TRUE)
curve(dnorm(x, mean=0, sd=3), add=TRUE)

Right size

epsilon <- 1
set.seed(1)
hmc_results <- sampler_hmc(pd_func, epsilon = epsilon, L = 10, start = 0)
hist(hmc_results[[1]], probability = TRUE)
curve(dnorm(x, mean=0, sd=3), add=TRUE)

lucas-castillo/SampleR documentation built on Jan. 1, 2021, 8:25 a.m.