Error : The RAM model 'blahblah' does not contain any paths
Posted on
CharlesD
Joined: 04/30/2013
Forums
This occurs after I change the data associated with an otherwise working matrix specification model... I don't think it should :)
gosh
The error is triggered by the following test,
if ((length(model$A) == 0) ||
(length(model$S) == 0) ||
(length(model$F) == 0)) {
Are you using non-default names for these matrices?
I am puzzled why changing the data would cause this error to trigger. Can you provide a script to reproduce the problem?
Log in or register to post comments
OpenMx thinks your matrix model is type="RAM"?
Certainly it ought not happen, but I am not quite following the use case
The error implies the model type ="RAM". But model type for matrix models is default, and the expectations etc are different.
Can you please check the following on your pre and post change models?
class(tmpMatMode)[1] == "MxRAMModel"
class(tmpMatMode$objective[[1]])[1] == "MxExpectationRAM"
e.g.
tmpMatMod <- mxModel("tmpMatMod",
mxExpectationNormal(covariance="expCov", means="expMeans"),
mxFitFunctionML()
)
class(tmpMatMod)[1] == "MxRAMModel" # FALSE
class(tmpMatMod$objective[[1]])[1] == "MxExpectationRAM" # FALSE
data(myFADataRaw)
manifests = names(myFADataRaw); latents = c("G")
myFADataRaw = myFADataRaw[, manifests]
tmpRAM <- mxModel("tmpRAM", type="RAM",
manifestVars = manifests, latentVars = latents,
mxPath(from = latents, to = manifests),
mxPath(from = manifests, arrows = 2), # manifest residuals
mxPath(from = latents, arrows = 2, free = F, values = 1), # latents fixed@1
mxData(cov(myFADataRaw, use="complete"), type = "cov", numObs = nrow(myFADataRaw))
)
class(tmpRAM)[1] == "MxRAMModel" # TRUE
class(tmpRAM$objective[[1]])[1] == "MxExpectationRAM" # TRUE
Log in or register to post comments
In reply to OpenMx thinks your matrix model is type="RAM"? by tbates
Ok, good pickup. I don't
Ok, good pickup. I don't remember *why*, but I do remember that for earlier versions of OpenMx I had to specify type='RAM' or else various things seemed to break. Could also be a bug in my own code that somehow improved along the way I guess... or in worst case I'll find over the coming week that certain things are not working :) Anyway, when I leave out the type argument, I'm able to change my data later with no problem. So, cheers!
Log in or register to post comments