rjbgoudie / structmcmc Goto Github PK

I'm trying to get this script working. After executing line 14, the following error is thrown.

Error in if (nrow(arcs) == 0) return(TRUE) : argument is of length zero

minimal example

install.packages("devtools")
library(devtools)

install_github("rjbgoudie/parental")
install_github("rjbgoudie/structmcmc")
library(structmcmc)

# Setup data frame
x1 <- factor(c("a", "a", "g", "c", "c", "a", "g", "a", "a"))
x2 <- factor(c(2, 2, 4, 3, 1, 4, 4, 4, 1))
x3 <- factor(c(FALSE, FALSE, TRUE, FALSE, TRUE, TRUE, FALSE, FALSE, TRUE))
x <- data.frame(x1 = x1, x2 = x2, x3 = x3)

# Draw samples from the posterior using MC^3.
set.seed(1234)
initial <- bn(c(), c(), c())
priorgraph <- bn(c(), c(1), c(2))
prior <- priorGraph(priorgraph, 0.5)
mcmc <- posterior(data = x, method = "mc3", prior = prior,
                  nSamples = 10000, nBurnin = 1000, initial = initial)

environment

platform x86_64-apple-darwin15.6.0
arch x86_64
os darwin15.6.0
system x86_64, darwin15.6.0
status
major 3
minor 4.0
year 2017
month 04
day 21
svn rev 72570
language R
version.string R version 3.4.0 (2017-04-21)

I get the popup window "R session aborted. R encountered a fatal error. The session was terminated." when I try to run the examples from the README file with the argument method = "gibbs". Do I need to change the other arguments when methods = "gibbs"?

I am running R version 3.5.0

Hi, when executing the demo script for discrete data from the Readme, I get the following error

> mcmc <- posterior(data = x, method = "mc3", prior = prior,
+                   nSamples = 10000, nBurnin = 1000, initial = initial)
Error in logScoreOfflineFUN(x = currentNetwork[[1]], logScoreParameters = logScoreParameters,  : 
  class(logScoreParameters$data) == "matrix" are not all TRUE
Called from: logScoreOfflineFUN(x = currentNetwork[[1]], logScoreParameters = logScoreParameters,  cache = cache)

I get this error after running:

x1 <- factor(c("a", "a", "g", "c", "c", "a", "g", "a", "a"))
x2 <- factor(c(2, 2, 4, 3, 1, 4, 4, 4, 1))
x3 <- factor(c(FALSE, FALSE, TRUE, FALSE, TRUE, TRUE, FALSE, FALSE, TRUE))
x <- data.frame(x1 = x1, x2 = x2, x3 = x3)
set.seed(1234)
initial <- bn(c(), c(), c())
priorgraph <- bn(c(), c(1), c(2))
prior <- priorGraph(priorgraph, 0.5)
mcmc <- posterior(data = x, method = "mc3", prior = prior,
                  nSamples = 10000, nBurnin = 1000, initial = initial)

Any ideas how to fix this? I am just exploring the package and would like to debug a walkthrough of a few examples to see how it works.

Thanks!

The browser() function is called inside the priorGraph function, meaning that Rstudio breaks and won't continue until enter is pressed. As this function is called many times I've not been able to run the example code given here

Section of code is below:

  difference <- setdiff(x, graph[[i]])
  difference <- length(difference)
  browser()
  exp(-lambda * difference)

RStudio Version 1.1.419

Just FYI, I also had to install the following packages, in addition to parental:

install.packages(c("bit", "fastmatch", "testthat"))