CRAN Package Check Results for Package propagate

Last updated on 2017-12-20 00:47:09 CET.

Check Details

Version: 1.0-4
Check: R code for possible problems
Result: NOTE
    WelchSatter: no visible global function definition for ‘qt’
    bigcor: no visible binding for global variable ‘cor’
    bigcor: no visible binding for global variable ‘cov’
    contribution: no visible global function definition for ‘barplot’
    counter: no visible global function definition for ‘flush.console’
    dsn: no visible global function definition for ‘dnorm’
    dsn: no visible global function definition for ‘pnorm’
    dst: no visible global function definition for ‘dt’
    fitDistr: no visible global function definition for ‘var’
    fitDistr: no visible global function definition for ‘sd’
    fitDistr: no visible global function definition for ‘density’
    fitDistr: no visible global function definition for ‘hist’
    fitDistr : fitAIC: no visible global function definition for ‘AIC’
    fitDistr: no visible binding for global variable ‘dnorm’
    fitDistr: no visible binding for global variable ‘dlnorm’
    fitDistr: no visible binding for global variable ‘dlogis’
    fitDistr: no visible binding for global variable ‘dunif’
    fitDistr: no visible binding for global variable ‘dgamma’
    fitDistr: no visible binding for global variable ‘dcauchy’
    fitDistr: no visible global function definition for ‘lines’
    interval: no visible global function definition for ‘abline’
    makeGrad : FUN: no visible global function definition for ‘D’
    makeHess : FUN: no visible global function definition for ‘D’
    mixCov: no visible global function definition for ‘cov’
    plot.propagate: no visible global function definition for ‘par’
    plot.propagate: no visible global function definition for ‘quantile’
    plot.propagate: no visible global function definition for ‘hist’
    plot.propagate: no visible global function definition for ‘density’
    plot.propagate: no visible global function definition for ‘lines’
    plot.propagate: no visible global function definition for ‘title’
    plot.propagate: no visible global function definition for ‘abline’
    plot.propagate: no visible global function definition for ‘boxplot’
    predictNLS: no visible global function definition for ‘coef’
    predictNLS: no visible global function definition for ‘vcov’
    predictNLS: no visible global function definition for ‘tail’
    predictNLS: no visible global function definition for ‘qt’
    predictNLS: no visible global function definition for ‘df.residual’
    predictNLS: no visible global function definition for ‘residuals’
    propagate : <anonymous>: no visible global function definition for ‘sd’
    propagate: no visible global function definition for ‘cov’
    propagate: no visible global function definition for ‘quantile’
    propagate: no visible global function definition for ‘rnorm’
    propagate: no visible global function definition for ‘sd’
    propagate: no visible global function definition for ‘median’
    propagate: no visible global function definition for ‘mad’
    rJSB : erf.inv: no visible global function definition for ‘qnorm’
    rJSB: no visible global function definition for ‘runif’
    rJSU : erf.inv: no visible global function definition for ‘qnorm’
    rJSU: no visible global function definition for ‘runif’
    rarcsin: no visible global function definition for ‘runif’
    rbeta2: no visible global function definition for ‘qbeta’
    rbeta2: no visible global function definition for ‘runif’
    rctrap: no visible global function definition for ‘runif’
    rgnorm : invcerf: no visible global function definition for ‘qnorm’
    rgnorm: no visible global function definition for ‘runif’
    rgtrap: no visible global function definition for ‘runif’
    rgumbel: no visible global function definition for ‘runif’
    rlaplace: no visible global function definition for ‘runif’
    rmises: no visible global function definition for ‘runif’
    rsn: no visible global function definition for ‘rnorm’
    rst: no visible global function definition for ‘rt’
    rtrap: no visible global function definition for ‘runif’
    rtriang: no visible global function definition for ‘runif’
    rweibull2: no visible global function definition for ‘runif’
    summary.propagate: no visible global function definition for
     ‘shapiro.test’
    summary.propagate: no visible global function definition for ‘rnorm’
    summary.propagate: no visible global function definition for ‘sd’
    summary.propagate: no visible global function definition for ‘ks.test’
    Undefined global functions or variables:
     AIC D abline barplot boxplot coef cor cov dcauchy density df.residual
     dgamma dlnorm dlogis dnorm dt dunif flush.console hist ks.test lines
     mad median par pnorm qbeta qnorm qt quantile residuals rnorm rt runif
     sd shapiro.test tail title var vcov
    Consider adding
     importFrom("graphics", "abline", "barplot", "boxplot", "hist", "lines",
     "par", "title")
     importFrom("stats", "AIC", "D", "coef", "cor", "cov", "dcauchy",
     "density", "df.residual", "dgamma", "dlnorm", "dlogis",
     "dnorm", "dt", "dunif", "ks.test", "mad", "median", "pnorm",
     "qbeta", "qnorm", "qt", "quantile", "residuals", "rnorm",
     "rt", "runif", "sd", "shapiro.test", "var", "vcov")
     importFrom("utils", "flush.console", "tail")
    to your NAMESPACE file.
Flavors: r-devel-linux-x86_64-debian-clang, r-devel-linux-x86_64-debian-gcc, r-devel-linux-x86_64-fedora-clang, r-devel-linux-x86_64-fedora-gcc, r-devel-windows-ix86+x86_64, r-patched-linux-x86_64, r-patched-solaris-x86, r-release-linux-x86_64, r-release-windows-ix86+x86_64, r-release-osx-x86_64, r-oldrel-windows-ix86+x86_64, r-oldrel-osx-x86_64

Version: 1.0-4
Check: compiled code
Result: NOTE
    File ‘propagate/libs/propagate.so’:
     Found no calls to: ‘R_registerRoutines’, ‘R_useDynamicSymbols’
    
    It is good practice to register native routines and to disable symbol
    search.
    
    See ‘Writing portable packages’ in the ‘Writing R Extensions’ manual.
Flavors: r-devel-linux-x86_64-debian-clang, r-devel-linux-x86_64-debian-gcc, r-devel-linux-x86_64-fedora-clang, r-devel-linux-x86_64-fedora-gcc, r-patched-linux-x86_64, r-release-linux-x86_64

Version: 1.0-4
Check: examples
Result: ERROR
    Running examples in ‘propagate-Ex.R’ failed
    The error most likely occurred in:
    
    > base::assign(".ptime", proc.time(), pos = "CheckExEnv")
    > ### Name: predictNLS
    > ### Title: Confidence intervals for nonlinear models based on uncertainty
    > ### propagation
    > ### Aliases: predictNLS
    > ### Keywords: array algebra multivariate
    >
    > ### ** Examples
    >
    > ## Example from ?nls.
    > DNase1 <- subset(DNase, Run == 1)
    > fm3DNase1 <- nls(density ~ Asym/(1 + exp((xmid - log(conc))/scal)),
    + data = DNase1, start = list(Asym = 3, xmid = 0, scal = 1))
    >
    > ## Using a single predictor value without error.
    > PROP1 <- predictNLS(fm3DNase1, newdata = data.frame(conc = 2))
    Propagating predictor value #1 ...
    > PRED1 <- predict(fm3DNase1, newdata = data.frame(conc = 2))
    > PROP1$summary
     Prop.Mean.1 Prop.Mean.2 Prop.sd.1 Prop.sd.2 Prop.2.5% Prop.97.5%
    1 0.7480472 0.7481514 0.008042515 0.008052591 0.7307548 0.7655479
     Sim.Mean Sim.sd Sim.Median Sim.MAD Sim.2.5% Sim.97.5%
    1 0.7481505 0.008041165 0.7481565 0.008040288 0.7323719 0.7639674
    > PRED1
    [1] 0.7480472
    > ## => Prop.Mean.1 equal to PRED1
    >
    > ## Not run:
    > ##D ## Using a sequence of predictor values without error.
    > ##D CONC <- seq(1, 12, by = 1)
    > ##D PROP2 <- predictNLS(fm3DNase1, newdata = data.frame(conc = CONC))
    > ##D PRED2 <- predict(fm3DNase1, newdata = data.frame(conc = CONC))
    > ##D PROP2$summary
    > ##D PRED2
    > ##D ## => Prop.Mean.1 equal to PRED2
    > ##D
    > ##D ## Using a sequence of predictor values with error.
    > ##D DAT <- data.frame(conc = CONC, error = rnorm(12, 0, 0.1))
    > ##D PROP3 <- predictNLS(fm3DNase1, newdata = DAT)
    > ##D PRED3 <- predict(fm3DNase1, newdata = DAT)
    > ##D PROP3$summary
    > ##D PRED3
    > ##D ## => Prop.Mean.1 equal to PRED3
    > ##D
    > ##D ## Plot predicted and confidence values from
    > ##D ## first-/second-order Taylor expansion
    > ##D ## and Monte Carlo simulation.
    > ##D plot(DNase1$conc, DNase1$density)
    > ##D lines(DNase1$conc, fitted(fm3DNase1), lwd = 2, col = 1)
    > ##D points(CONC, PROP2$summary[, 1], col = 2, pch = 16)
    > ##D lines(CONC, PROP2$summary[, 5], col = 2)
    > ##D lines(CONC, PROP2$summary[, 6], col = 2)
    > ## End(Not run)
    >
    > ## Using multiple predictor values
    > ## 1: Setup of response values
    > ## with gaussian error of 10%.
    > x <- seq(1, 10, by = 0.01)
    > y <- seq(10, 1, by = -0.01)
    > a <- 2
    > b <- 5
    > c <- 10
    > z <- a * exp(b * x)^sin(y/c)
    > z <- z + sapply(z, function(x) rnorm(1, x, 0.10 * x))
    > ## 2: Fit 'nls' model.
    > MOD <- nls(z ~ a * exp(b * x)^sin(y/c),
    + start = list(a = 2, b = 5, c = 10))
    > ## 3: newdata without errors and prediction.
    > DAT1 <- data.frame(x = 4, y = 3)
    > PROP4 <- predictNLS(MOD, newdata = DAT1)
    Propagating predictor value #1 ...
    > PROP4$summary
     Prop.Mean.1 Prop.Mean.2 Prop.sd.1 Prop.sd.2 Prop.2.5% Prop.97.5% Sim.Mean
    1 1444.777 1442.775 54.96645 55.04017 1334.753 1550.798 1442.775
     Sim.sd Sim.Median Sim.MAD Sim.2.5% Sim.97.5%
    1 55.09039 1444.558 54.91878 1329.912 1545.241
    > ## 4: newdata with errors and prediction.
    > DAT2 <- data.frame(x = 4, y = 3, error.x = 0.2, error.y = 0.1)
    > PROP5 <- predictNLS(MOD, newdata = DAT2)
    Propagating predictor value #1 ...
    Error in `rownames<-`(`*tmp*`, value = make.unique(nameVEC)) :
     attempt to set 'rownames' on an object with no dimensions
    Calls: predictNLS -> mixCov -> makeCov -> rownames<-
    Execution halted
Flavors: r-devel-linux-x86_64-debian-clang, r-devel-linux-x86_64-debian-gcc

Version: 1.0-4
Check: examples
Result: ERROR
    Running examples in ‘propagate-Ex.R’ failed
    The error most likely occurred in:
    
    > ### Name: predictNLS
    > ### Title: Confidence intervals for nonlinear models based on uncertainty
    > ### propagation
    > ### Aliases: predictNLS
    > ### Keywords: array algebra multivariate
    >
    > ### ** Examples
    >
    > ## Example from ?nls.
    > DNase1 <- subset(DNase, Run == 1)
    > fm3DNase1 <- nls(density ~ Asym/(1 + exp((xmid - log(conc))/scal)),
    + data = DNase1, start = list(Asym = 3, xmid = 0, scal = 1))
    >
    > ## Using a single predictor value without error.
    > PROP1 <- predictNLS(fm3DNase1, newdata = data.frame(conc = 2))
    Propagating predictor value #1 ...
    > PRED1 <- predict(fm3DNase1, newdata = data.frame(conc = 2))
    > PROP1$summary
     Prop.Mean.1 Prop.Mean.2 Prop.sd.1 Prop.sd.2 Prop.2.5% Prop.97.5%
    1 0.7480472 0.7481514 0.008042515 0.008052591 0.7307548 0.7655479
     Sim.Mean Sim.sd Sim.Median Sim.MAD Sim.2.5% Sim.97.5%
    1 0.7481505 0.008041165 0.7481565 0.008040288 0.7323719 0.7639674
    > PRED1
    [1] 0.7480472
    > ## => Prop.Mean.1 equal to PRED1
    >
    > ## Not run:
    > ##D ## Using a sequence of predictor values without error.
    > ##D CONC <- seq(1, 12, by = 1)
    > ##D PROP2 <- predictNLS(fm3DNase1, newdata = data.frame(conc = CONC))
    > ##D PRED2 <- predict(fm3DNase1, newdata = data.frame(conc = CONC))
    > ##D PROP2$summary
    > ##D PRED2
    > ##D ## => Prop.Mean.1 equal to PRED2
    > ##D
    > ##D ## Using a sequence of predictor values with error.
    > ##D DAT <- data.frame(conc = CONC, error = rnorm(12, 0, 0.1))
    > ##D PROP3 <- predictNLS(fm3DNase1, newdata = DAT)
    > ##D PRED3 <- predict(fm3DNase1, newdata = DAT)
    > ##D PROP3$summary
    > ##D PRED3
    > ##D ## => Prop.Mean.1 equal to PRED3
    > ##D
    > ##D ## Plot predicted and confidence values from
    > ##D ## first-/second-order Taylor expansion
    > ##D ## and Monte Carlo simulation.
    > ##D plot(DNase1$conc, DNase1$density)
    > ##D lines(DNase1$conc, fitted(fm3DNase1), lwd = 2, col = 1)
    > ##D points(CONC, PROP2$summary[, 1], col = 2, pch = 16)
    > ##D lines(CONC, PROP2$summary[, 5], col = 2)
    > ##D lines(CONC, PROP2$summary[, 6], col = 2)
    > ## End(Not run)
    >
    > ## Using multiple predictor values
    > ## 1: Setup of response values
    > ## with gaussian error of 10%.
    > x <- seq(1, 10, by = 0.01)
    > y <- seq(10, 1, by = -0.01)
    > a <- 2
    > b <- 5
    > c <- 10
    > z <- a * exp(b * x)^sin(y/c)
    > z <- z + sapply(z, function(x) rnorm(1, x, 0.10 * x))
    > ## 2: Fit 'nls' model.
    > MOD <- nls(z ~ a * exp(b * x)^sin(y/c),
    + start = list(a = 2, b = 5, c = 10))
    > ## 3: newdata without errors and prediction.
    > DAT1 <- data.frame(x = 4, y = 3)
    > PROP4 <- predictNLS(MOD, newdata = DAT1)
    Propagating predictor value #1 ...
    > PROP4$summary
     Prop.Mean.1 Prop.Mean.2 Prop.sd.1 Prop.sd.2 Prop.2.5% Prop.97.5% Sim.Mean
    1 1444.777 1442.775 54.96645 55.04017 1334.753 1550.798 1442.775
     Sim.sd Sim.Median Sim.MAD Sim.2.5% Sim.97.5%
    1 55.09039 1444.558 54.91878 1329.912 1545.241
    > ## 4: newdata with errors and prediction.
    > DAT2 <- data.frame(x = 4, y = 3, error.x = 0.2, error.y = 0.1)
    > PROP5 <- predictNLS(MOD, newdata = DAT2)
    Propagating predictor value #1 ...
    Error in `rownames<-`(`*tmp*`, value = make.unique(nameVEC)) :
     attempt to set 'rownames' on an object with no dimensions
    Calls: predictNLS -> mixCov -> makeCov -> rownames<-
    Execution halted
Flavors: r-devel-linux-x86_64-fedora-clang, r-devel-linux-x86_64-fedora-gcc

Version: 1.0-4
Check: running examples for arch ‘i386’
Result: ERROR
    Running examples in 'propagate-Ex.R' failed
    The error most likely occurred in:
    
    > ### Name: predictNLS
    > ### Title: Confidence intervals for nonlinear models based on uncertainty
    > ### propagation
    > ### Aliases: predictNLS
    > ### Keywords: array algebra multivariate
    >
    > ### ** Examples
    >
    > ## Example from ?nls.
    > DNase1 <- subset(DNase, Run == 1)
    > fm3DNase1 <- nls(density ~ Asym/(1 + exp((xmid - log(conc))/scal)),
    + data = DNase1, start = list(Asym = 3, xmid = 0, scal = 1))
    >
    > ## Using a single predictor value without error.
    > PROP1 <- predictNLS(fm3DNase1, newdata = data.frame(conc = 2))
    Propagating predictor value #1 ...
    > PRED1 <- predict(fm3DNase1, newdata = data.frame(conc = 2))
    > PROP1$summary
     Prop.Mean.1 Prop.Mean.2 Prop.sd.1 Prop.sd.2 Prop.2.5% Prop.97.5%
    1 0.7480472 0.7481514 0.008042515 0.008052591 0.7307548 0.7655479
     Sim.Mean Sim.sd Sim.Median Sim.MAD Sim.2.5% Sim.97.5%
    1 0.7481507 0.008043127 0.7481808 0.008074998 0.7323472 0.7637585
    > PRED1
    [1] 0.7480472
    > ## => Prop.Mean.1 equal to PRED1
    >
    > ## Not run:
    > ##D ## Using a sequence of predictor values without error.
    > ##D CONC <- seq(1, 12, by = 1)
    > ##D PROP2 <- predictNLS(fm3DNase1, newdata = data.frame(conc = CONC))
    > ##D PRED2 <- predict(fm3DNase1, newdata = data.frame(conc = CONC))
    > ##D PROP2$summary
    > ##D PRED2
    > ##D ## => Prop.Mean.1 equal to PRED2
    > ##D
    > ##D ## Using a sequence of predictor values with error.
    > ##D DAT <- data.frame(conc = CONC, error = rnorm(12, 0, 0.1))
    > ##D PROP3 <- predictNLS(fm3DNase1, newdata = DAT)
    > ##D PRED3 <- predict(fm3DNase1, newdata = DAT)
    > ##D PROP3$summary
    > ##D PRED3
    > ##D ## => Prop.Mean.1 equal to PRED3
    > ##D
    > ##D ## Plot predicted and confidence values from
    > ##D ## first-/second-order Taylor expansion
    > ##D ## and Monte Carlo simulation.
    > ##D plot(DNase1$conc, DNase1$density)
    > ##D lines(DNase1$conc, fitted(fm3DNase1), lwd = 2, col = 1)
    > ##D points(CONC, PROP2$summary[, 1], col = 2, pch = 16)
    > ##D lines(CONC, PROP2$summary[, 5], col = 2)
    > ##D lines(CONC, PROP2$summary[, 6], col = 2)
    > ## End(Not run)
    >
    > ## Using multiple predictor values
    > ## 1: Setup of response values
    > ## with gaussian error of 10%.
    > x <- seq(1, 10, by = 0.01)
    > y <- seq(10, 1, by = -0.01)
    > a <- 2
    > b <- 5
    > c <- 10
    > z <- a * exp(b * x)^sin(y/c)
    > z <- z + sapply(z, function(x) rnorm(1, x, 0.10 * x))
    > ## 2: Fit 'nls' model.
    > MOD <- nls(z ~ a * exp(b * x)^sin(y/c),
    + start = list(a = 2, b = 5, c = 10))
    > ## 3: newdata without errors and prediction.
    > DAT1 <- data.frame(x = 4, y = 3)
    > PROP4 <- predictNLS(MOD, newdata = DAT1)
    Propagating predictor value #1 ...
    > PROP4$summary
     Prop.Mean.1 Prop.Mean.2 Prop.sd.1 Prop.sd.2 Prop.2.5% Prop.97.5% Sim.Mean
    1 1444.777 1442.775 54.96645 55.04017 1334.753 1550.798 1442.775
     Sim.sd Sim.Median Sim.MAD Sim.2.5% Sim.97.5%
    1 55.09039 1444.558 54.91878 1329.912 1545.241
    > ## 4: newdata with errors and prediction.
    > DAT2 <- data.frame(x = 4, y = 3, error.x = 0.2, error.y = 0.1)
    > PROP5 <- predictNLS(MOD, newdata = DAT2)
    Propagating predictor value #1 ...
    Error in `rownames<-`(`*tmp*`, value = make.unique(nameVEC)) :
     attempt to set 'rownames' on an object with no dimensions
    Calls: predictNLS -> mixCov -> makeCov -> rownames<-
    Execution halted
Flavor: r-devel-windows-ix86+x86_64

Version: 1.0-4
Check: running examples for arch ‘x64’
Result: ERROR
    Running examples in 'propagate-Ex.R' failed
    The error most likely occurred in:
    
    > ### Name: predictNLS
    > ### Title: Confidence intervals for nonlinear models based on uncertainty
    > ### propagation
    > ### Aliases: predictNLS
    > ### Keywords: array algebra multivariate
    >
    > ### ** Examples
    >
    > ## Example from ?nls.
    > DNase1 <- subset(DNase, Run == 1)
    > fm3DNase1 <- nls(density ~ Asym/(1 + exp((xmid - log(conc))/scal)),
    + data = DNase1, start = list(Asym = 3, xmid = 0, scal = 1))
    >
    > ## Using a single predictor value without error.
    > PROP1 <- predictNLS(fm3DNase1, newdata = data.frame(conc = 2))
    Propagating predictor value #1 ...
    > PRED1 <- predict(fm3DNase1, newdata = data.frame(conc = 2))
    > PROP1$summary
     Prop.Mean.1 Prop.Mean.2 Prop.sd.1 Prop.sd.2 Prop.2.5% Prop.97.5%
    1 0.7480472 0.7481514 0.008042515 0.008052591 0.7307548 0.7655479
     Sim.Mean Sim.sd Sim.Median Sim.MAD Sim.2.5% Sim.97.5%
    1 0.7481505 0.008041165 0.7481565 0.008040288 0.7323719 0.7639674
    > PRED1
    [1] 0.7480472
    > ## => Prop.Mean.1 equal to PRED1
    >
    > ## Not run:
    > ##D ## Using a sequence of predictor values without error.
    > ##D CONC <- seq(1, 12, by = 1)
    > ##D PROP2 <- predictNLS(fm3DNase1, newdata = data.frame(conc = CONC))
    > ##D PRED2 <- predict(fm3DNase1, newdata = data.frame(conc = CONC))
    > ##D PROP2$summary
    > ##D PRED2
    > ##D ## => Prop.Mean.1 equal to PRED2
    > ##D
    > ##D ## Using a sequence of predictor values with error.
    > ##D DAT <- data.frame(conc = CONC, error = rnorm(12, 0, 0.1))
    > ##D PROP3 <- predictNLS(fm3DNase1, newdata = DAT)
    > ##D PRED3 <- predict(fm3DNase1, newdata = DAT)
    > ##D PROP3$summary
    > ##D PRED3
    > ##D ## => Prop.Mean.1 equal to PRED3
    > ##D
    > ##D ## Plot predicted and confidence values from
    > ##D ## first-/second-order Taylor expansion
    > ##D ## and Monte Carlo simulation.
    > ##D plot(DNase1$conc, DNase1$density)
    > ##D lines(DNase1$conc, fitted(fm3DNase1), lwd = 2, col = 1)
    > ##D points(CONC, PROP2$summary[, 1], col = 2, pch = 16)
    > ##D lines(CONC, PROP2$summary[, 5], col = 2)
    > ##D lines(CONC, PROP2$summary[, 6], col = 2)
    > ## End(Not run)
    >
    > ## Using multiple predictor values
    > ## 1: Setup of response values
    > ## with gaussian error of 10%.
    > x <- seq(1, 10, by = 0.01)
    > y <- seq(10, 1, by = -0.01)
    > a <- 2
    > b <- 5
    > c <- 10
    > z <- a * exp(b * x)^sin(y/c)
    > z <- z + sapply(z, function(x) rnorm(1, x, 0.10 * x))
    > ## 2: Fit 'nls' model.
    > MOD <- nls(z ~ a * exp(b * x)^sin(y/c),
    + start = list(a = 2, b = 5, c = 10))
    > ## 3: newdata without errors and prediction.
    > DAT1 <- data.frame(x = 4, y = 3)
    > PROP4 <- predictNLS(MOD, newdata = DAT1)
    Propagating predictor value #1 ...
    > PROP4$summary
     Prop.Mean.1 Prop.Mean.2 Prop.sd.1 Prop.sd.2 Prop.2.5% Prop.97.5% Sim.Mean
    1 1444.777 1442.775 54.96645 55.04016 1334.753 1550.798 1442.775
     Sim.sd Sim.Median Sim.MAD Sim.2.5% Sim.97.5%
    1 55.09055 1444.517 54.85169 1329.825 1545.365
    > ## 4: newdata with errors and prediction.
    > DAT2 <- data.frame(x = 4, y = 3, error.x = 0.2, error.y = 0.1)
    > PROP5 <- predictNLS(MOD, newdata = DAT2)
    Propagating predictor value #1 ...
    Error in `rownames<-`(`*tmp*`, value = make.unique(nameVEC)) :
     attempt to set 'rownames' on an object with no dimensions
    Calls: predictNLS -> mixCov -> makeCov -> rownames<-
    Execution halted
Flavor: r-devel-windows-ix86+x86_64