CRAN Package Check Results for Package semtree

Last updated on 2018-11-17 10:51:24 CET.

Flavor Version Tinstall Tcheck Ttotal Status Flags
r-devel-linux-x86_64-debian-clang 0.9.13 9.06 622.92 631.98 OK
r-devel-linux-x86_64-debian-gcc 0.9.13 8.42 337.13 345.55 OK
r-devel-linux-x86_64-fedora-clang 0.9.13 520.99 OK
r-devel-linux-x86_64-fedora-gcc 0.9.13 405.74 OK
r-devel-windows-ix86+x86_64 0.9.13 23.00 457.00 480.00 OK
r-patched-linux-x86_64 0.9.13 7.88 452.24 460.12 OK
r-patched-solaris-x86 0.9.13 469.00 ERROR
r-release-linux-x86_64 0.9.13 8.23 437.99 446.22 OK
r-release-windows-ix86+x86_64 0.9.13 23.00 461.00 484.00 OK
r-release-osx-x86_64 0.9.13 OK
r-oldrel-windows-ix86+x86_64 0.9.13 13.00 462.00 475.00 OK
r-oldrel-osx-x86_64 0.9.13 OK

Check Details

Version: 0.9.13
Check: tests
Result: ERROR
     Running ‘invariance.R’ [19s/28s]
     Running ‘lavaan.R’ [43s/53s]
     Running ‘tree.R’ [11s/14s]
     Running ‘vim.R’ [227s/277s]
    Running the tests in ‘tests/tree.R’ failed.
    Complete output:
     > set.seed(789)
     > require("semtree")
     Loading required package: semtree
     Loading required package: OpenMx
     To take full advantage of multiple cores, use:
     mxOption(NULL, 'Number of Threads', parallel::detectCores()) #now
     Sys.setenv(OMP_NUM_THREADS=parallel::detectCores()) #before library(OpenMx)
     > data(lgcm)
     >
     > lgcm$agegroup <- as.ordered(lgcm$agegroup)
     > lgcm$training <- as.factor(lgcm$training)
     > lgcm$noise <- as.numeric(lgcm$noise)
     >
     > # LOAD IN OPENMX MODEL.
     > # A SIMPLE LINEAR GROWTH MODEL WITH 5 TIME POINTS FROM SIMULATED DATA
     >
     > manifests <- names(lgcm)[1:5]
     > lgcModel <- mxModel("Linear Growth Curve Model Path Specification",
     + type="RAM",
     + manifestVars=manifests,
     + latentVars=c("intercept","slope"),
     + # residual variances
     + mxPath(
     + from=manifests,
     + arrows=2,
     + free=TRUE,
     + values = c(1, 1, 1, 1, 1),
     + labels=c("residual1","residual2","residual3","residual4","residual5")
     + ),
     + # latent variances and covariance
     + mxPath(
     + from=c("intercept","slope"),
     + connect="unique.pairs",
     + arrows=2,
     + free=TRUE,
     + values=c(1, 1, 1),
     + labels=c("vari", "cov", "vars")
     + ),
     + # intercept loadings
     + mxPath(
     + from="intercept",
     + to=manifests,
     + arrows=1,
     + free=FALSE,
     + values=c(1, 1, 1, 1, 1)
     + ),
     + # slope loadings
     + mxPath(
     + from="slope",
     + to=manifests,
     + arrows=1,
     + free=FALSE,
     + values=c(0, 1, 2, 3, 4)
     + ),
     + # manifest means
     + mxPath(
     + from="one",
     + to=manifests,
     + arrows=1,
     + free=FALSE,
     + values=c(0, 0, 0, 0, 0)
     + ),
     + # latent means
     + mxPath(
     + from="one",
     + to=c("intercept", "slope"),
     + arrows=1,
     + free=TRUE,
     + values=c(1, 1),
     + labels=c("meani", "means")
     + ),
     + mxData(lgcm,type="raw")
     + )
     >
     >
     > # TREE CONTROL OPTIONS.
     > # TO OBTAIN BASIC/DEFAULT SMETREE OPTIONS, SIMPLY TPYE THE FOLLOWING:
     >
     > controlOptions <- semtree.control(method = "naive")
     > controlOptions$alpha <- 0.05
     >
     > # RUN TREE.
     >
     > tree <- semtree(model=lgcModel, data=lgcm, control = controlOptions)
     [x] Tree construction finished!
     >
     > # RERUN TREE WITH MODEL CONSTRAINTS.
     > # MODEL CONSTRAINTS CAN BE ADDED BY IDENTIFYING THE PARAMETERS TO BE
     > # CONSTRAINED IN EVERY NODE. ONLY UNCONSTRAINED PARAMETERS ARE THEN
     > # TESTED AT EACH NODE FOR GROUP DIFFERENCES. IN THIS EXAMPLE THE MODEL
     > # RESIDUALS ARE CONSTRAINED OVER THE NODES.
     >
     > constraints <- semtree.constraints(global.invariance = names(omxGetParameters(lgcModel))[1:5])
     >
     > treeConstrained <- semtree(model=lgcModel, data=lgcm, control = controlOptions,
     + constraints=constraints)
     Global Constraints:
     residual1 residual2 residual3 residual4 residual5
     Freely Estimated Parameters:
     vari cov vars meani means
     Error : The job for model 'INITIALIZED MODEL' exited abnormally with the error message: fit is not finite (0: The continuous part of the model implied covariance (loc2) is not positive definite in data 'INITIALIZED MODEL.data' row 245. Detail:
     covariance = matrix(c( # 5x5
     3.6378e+19, 7.27559e+19, 1.09134e+20, 1.45512e+20, 1.8189e+20
     , 7.27559e+19, 1.45512e+20, 2.18268e+20, 2.91024e+20, 3.6378e+20
     , 1.09134e+20, 2.18268e+20, 3.27402e+20, 4.36536e+20, 5.4567e+20
     , 1.45512e+20, 2.91024e+20, 4.36536e+20, 5.82048e+20, 7.27559e+20
     , 1.8189e+20, 3.6378e+20, 5.4567e+20, 7.27559e+20, 9.09449e+20), byrow=TRUE, nrow=5, ncol=5)
    
     )
     Model had a run error.
     [x] Tree construction finished!
     >
     > # SEE PLOT.
     > # THE PLOT FUNCTION WILL SHOW ALL FREE PARAMETERS AT EACH TERMINAL NODE.
     > # THIS CAN CREATE UNREADABLE FIGURES FOR MODELS WITH MANY FREE PARAMETERS.
     >
     > plot(tree)
     >
     > summary(tree)
     SEMtree Summary
     Template model:
     Total Sample Size: 400
     Number of nodes: 7
     Number of leaf nodes: 4
     Free Parameters: 10 ( residual1 residual2 residual3 residual4 residual5 vari cov vars meani means )
     >
     > summary(treeConstrained)
     Error in t.default(tree$params) : argument is not a matrix
     Calls: summary ... getTerminalNodes -> getTerminalNodes.rec -> cbind -> t -> t.default
     Execution halted
Flavor: r-patched-solaris-x86