Hello,
I'm trying to use parallel computing in MuMIn package. It worked a couple
month ago, but now I'm getting the error. Here is a code:
#Data
x1 = rnorm(1000)
x2 = rnorm(1000)
x3 = rnorm(1000)
z = 1 + 0.5*x1 - x2
pr = 1/(1+exp(-z))
y = rbinom(1000, 1, pr)
dataD = cbind(y=y, x1=x1 ,x2=x2, x3=x3)
#Fixed variable
fix.var = c("x1")
#Model averaging
cl <- makeCluster(detectCores())
clusterExport(cl, c("dataD"), envir=environment())
clusterExport(cl, c("y"), envir=environment())
clusterExport(cl, c("fix.var"), envir=environment())
clusterEvalQ(cl, library(MuMIn))
d = data.frame(dataD)
options(na.action = "na.fail")
m.inc = glm(y~., family=binomial, data=d)
MA.inc = pdredge(global.model = m.inc, rank = "AIC" , fixed =
fix.var, cluster=cl)
mo.inc = model.avg(MA.inc, rank = "AIC")
atr.inc = attr(mo.inc, "modelList")
stopCluster(cl)
The error says that data frame 'd' is not found and the pdredge result is
empty:
Warning messages:
1: In is.data.frame(data) : object 'd' not found (model 0 skipped)
2: In is.data.frame(data) : object 'd' not found (model 1 skipped)
3: In is.data.frame(data) : object 'd' not found (model 2 skipped)
4: In is.data.frame(data) : object 'd' not found (model 3 skipped)
5: In is.data.frame(data) : object 'd' not found (model 4 skipped)
Error in pdredge(global.model = m.inc, rank = "AIC" , fixed = fix.var, :
the result is empty
--
*I like to pretend I'm alone*. *Completely alone*. *Maybe post-apocalypse
or plague*... *Whatever*. *No-one left to act normal for. No need to hide
who I really am. It would be... freeing*. *...*
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
