Error in jags.model

[masaba79]

Hello! I used MixSIAR successfully since ~2018 (thanks a lot by the way for this modeling framework!!). I´ve used the gui and modified the example scripts to use my data without problems in the past. But now, after a year or so without touching the script I want to run it with another set of data, I get an error when I try to run the test (and any jags model). I get through everything ok (loading the mixture, source and discrimination files, making the isospace and priors plot, calculating the convex hull area all ok). I think the model is written ok too (attached) MixSIAR_modelDate1Loc1.txt.

This is my code (like I said I basically modified the example script): model_filename <- "MixSIAR_modelDate1Loc1.txt" resid_err <- TRUE process_err <- FALSE write_JAGS_model(model_filename, resid_err, process_err, mix, source) jags<- run_model(run="test", mix, source, discr, model_filename, alpha.prior=1, resid_err, process_err)

But I get this error:

Compiling model graph Resolving undeclared variables Allocating nodes Graph information: Observed stochastic nodes: 9 Unobserved stochastic nodes: 47 Total graph size: 652 Initializing model Deleting model

Error in jags.model(model.file, data = data, inits = init.values, n.chains = n.chains, : Error in node src_tau[3,1] Invalid parent values

Also, not sure if related, mixsiar_gui() doesnt work:

Error in mixsiar_gui() : could not find function "mixsiar_gui"

any idea what is wrong? and how can I fix this?

More info: I had to uninstall and install again R, Rstudio and MixSIAR because I had a previous problem with ggplot2 not being able to work (aka got stuck in ploting the isospace, but got resoved by fresh install of everything). To install MixSIAR now I followed instructions just like the first time and just noticed that now I have two versions of JAGS in my computer: JAGS-4.2.0 and JAGS-4.3.0...how do I know which one uses MixSIAR, or does it matter? Shall I delete one of these? which one? The freshly installed one is the oldest 4.2.0 (not sure how this happended).

Any help appreciated! thnks

Susana.

[masaba79]

update: I ran it with previous datasets (which have worked before) and now the Jags model ran smoothly! So it might be something about the dataset... cant figure what it is.

[DSchimmenti]

I think I might be having a similar issue and I am also looking for a solution. I am working on Mac OSX with R version 4.3.1 in RStudio (I have checked and I get this error regardless of whether I run my code in RStudio or the R console). I have run MixSIAR many times with my own data using 2 isotopes/tracers, 11 sources, and 35 mixture data points with no errors when I run the JAGS model. I am also using age as a fixed effect and running the model with residual error only no process error. Now that I am attempting 3 isotopes/tracers with the same amount of sources, mixture data points, fixed effects, and error settings, I get the following error message when I attempt to run the JAGS model:

Compiling model graph Resolving undeclared variables Allocating nodes Deleting model

Error in jags.model(model.file, data = data, inits = init.values, n.chains = n.chains, : RUNTIME ERROR: Unable to resolve the following parameters: Factor.1[67] (line 40) cross[67,1,1:10] (line 50) cross[67,2,1:10] (line 50) cross[67,3,1:10] (line 50) cross[67,4,1:10] (line 50) cross[67,5,1:10] (line 50) cross[67,6,1:10] (line 50) cross[67,7,1:10] (line 50) cross[67,8,1:10] (line 50) cross[67,9,1:10] (line 50) cross[67,10,1:10] (line 50) cross[67,11,1:10] (line 50) mix.mu[1:3,67] (line 92, 93) p.ind[67,1:11] (line 83) tmp.p[67,1:11] (line 53) Either supply values for these nodes with the data or define them on the left hand side of a relation.

I attempted to run the Isopod example to see if for some reason it was an issue related to using more than 2 isotopes/tracers, but I was able to run that vignette with no errors. I also double checked my mixture, source, and discrimination factor data files and everything looked okay and consistent.

My code for my data is as follows:

library(MixSIAR) setwd("/Users/Danielle/Documents/R Data/PhD_Ch3/Manuscript Data Files") mix206208eNd <- load_mix_data(filename = "Mixture_206208eNd_MixSIAR_withAges.csv", iso_names = c("Pb206204", "Pb208204", "eNd"), factors = c("Age..kyr.BP."), fac_random = FALSE, fac_nested = NULL, cont_effects = NULL) source206208eNd <- load_source_data(filename = "Sources_206208eNd_AllPSAs_MixSIAR.csv", source_factors = NULL, conc_dep = FALSE, data_type = "means", mix206208eNd) discr206208eNd <- load_discr_data(filename = "Discrimination_206208eNd_AllPSAs.csv", mix206208eNd) model_filename206208eNd <- "MixSIAR_model_206208eNd_AllPSAs.txt" resid_err <- TRUE process_err <- FALSE write_JAGS_model(model_filename206208eNd, resid_err, process_err, mix206208eNd, source206208eNd) jags.1_206208eNd_test <- run_model(run="test", mix206208eNd, source206208eNd, discr206208eNd, model_filename206208eNd, alpha.prior = 1)