CRAN Package Check Results for Package tidyLPA

Last updated on 2019-03-23 00:47:04 CET.

Check Details

Version: 1.0.0
Check: tests
Result: ERROR
     Running 'testthat.R' [27s/29s]
    Running the tests in 'tests/testthat.R' failed.
    Complete output:
     > library(testthat)
     > library(tidyLPA)
     tidyLPA has received a major update, with a much easier workflow and improved functionality. However, you might have to update old syntax to account for the new workflow. See vignette('introduction-to-major-changes') for details!
    
     Mplus is not installed. Use only package = 'mclust' when calling estimate_profiles().
     Warning message:
     In system2("type", args = "mplus", stdout = FALSE, stderr = FALSE) :
     error in running command
     >
     > test_check("tidyLPA")
     -- 1. Error: (unknown) (@test-compare-solutions.R#10) -------------------------
     object must be of class 'Mclust' or 'densityMclust'
     1: pisaUSA15[1:100, ] %>% single_imputation() %>% estimate_profiles(1:3, variances = c("equal",
     "varying"), covariances = c("zero", "varying")) %>% compare_solutions(statistics = c("AIC",
     "BIC")) at testthat/test-compare-solutions.R:10
     2: withVisible(eval(quote(`_fseq`(`_lhs`)), env, env))
     3: eval(quote(`_fseq`(`_lhs`)), env, env)
     4: eval(quote(`_fseq`(`_lhs`)), env, env)
     5: `_fseq`(`_lhs`)
     6: freduce(value, `_function_list`)
     7: function_list[[i]](value)
     8: estimate_profiles(., 1:3, variances = c("equal", "varying"), covariances = c("zero",
     "varying"))
     9: estimate_profiles.data.frame(., 1:3, variances = c("equal", "varying"), covariances = c("zero",
     "varying"))
     10: NextMethod("estimate_profiles", df)
     11: estimate_profiles.default(., 1:3, variances = c("equal", "varying"), covariances = c("zero",
     "varying"))
     12: estimate_profiles_mclust(df, n_profiles, model_numbers, ...)
     13: mapply(FUN = function(this_class, this_model) {
     out <- list(model = Mclust(full_data, G = this_class, modelNames = ifelse(ncol(full_data) ==
     1, substr(get_modelname(this_model), 1, 1), get_modelname(this_model)), warn = FALSE,
     verbose = FALSE, ...))
     out$model$mclustBootstrap <- MclustBootstrap(out$model, nboot = 100, type = "bs",
     verbose = FALSE)
     out$model$LRTS <- ifelse(this_class == 1, NA, boot_blrt[[which(model_numbers ==
     this_model)]]$obs[this_class - 1])
     out$model$LRTS_p <- ifelse(this_class == 1, NA, boot_blrt[[which(model_numbers ==
     this_model)]]$p.value[this_class - 1])
     out$fit <- c(Model = this_model, Classes = this_class, calc_fitindices(out$model))
     estimates <- estimates(out$model)
     estimates$Model <- this_model
     estimates$Classes <- this_class
     if (this_class == 1) {
     estimates$se[estimates$Category == "Means"] <- estimates$Estimate[estimates$Category ==
     "Variances"]/out$model$n
     estimates$se[estimates$Category == "Variances"] <- sapply(sqrt(estimates$Estimate[estimates$Category ==
     "Variances"]), se_s, n = out$model$n)^2
     estimates$p <- stats::pnorm(abs(estimates$Estimate), sd = estimates$se, lower.tail = FALSE)
     }
     out$estimates <- estimates
     outdat <- cbind(out$model$z, out$model$classification)
     dff <- matrix(NA, dim(df)[1], dim(outdat)[2])
     dff[no_na_rows, ] <- outdat
     colnames(dff) <- c(paste0("CPROB", 1:ncol(out$model$z)), "Class")
     out$dff <- as_tibble(cbind(df, dff))
     out$dff$model_number <- this_model
     out$dff$classes_number <- this_class
     out$dff <- out$dff[, c((ncol(out$dff) - 1), ncol(out$dff), 1:(ncol(out$dff) -
     2))]
     warnings <- NULL
     if (out$fit[["prob_min"]] < 0.001)
     warnings <- c(warnings, "Some classes were not assigned any cases with more than .1% probability. Consequently, these solutions are effectively identical to a solution with one class less.")
     if (out$fit[["n_min"]] < 0.01)
     warnings <- c(warnings, "Less than 1% of cases were assigned to one of the profiles. Interpret this solution with caution and consider other models.")
     out$warnings <- warnings
     class(out) <- c("tidyProfile.mclust", "tidyProfile", "list")
     out
     }, this_class = run_models$prof, this_model = run_models$mod, SIMPLIFY = FALSE)
     14: (function (this_class, this_model)
     {
     out <- list(model = Mclust(full_data, G = this_class, modelNames = ifelse(ncol(full_data) ==
     1, substr(get_modelname(this_model), 1, 1), get_modelname(this_model)), warn = FALSE,
     verbose = FALSE, ...))
     out$model$mclustBootstrap <- MclustBootstrap(out$model, nboot = 100, type = "bs",
     verbose = FALSE)
     out$model$LRTS <- ifelse(this_class == 1, NA, boot_blrt[[which(model_numbers ==
     this_model)]]$obs[this_class - 1])
     out$model$LRTS_p <- ifelse(this_class == 1, NA, boot_blrt[[which(model_numbers ==
     this_model)]]$p.value[this_class - 1])
     out$fit <- c(Model = this_model, Classes = this_class, calc_fitindices(out$model))
     estimates <- estimates(out$model)
     estimates$Model <- this_model
     estimates$Classes <- this_class
     if (this_class == 1) {
     estimates$se[estimates$Category == "Means"] <- estimates$Estimate[estimates$Category ==
     "Variances"]/out$model$n
     estimates$se[estimates$Category == "Variances"] <- sapply(sqrt(estimates$Estimate[estimates$Category ==
     "Variances"]), se_s, n = out$model$n)^2
     estimates$p <- stats::pnorm(abs(estimates$Estimate), sd = estimates$se, lower.tail = FALSE)
     }
     out$estimates <- estimates
     outdat <- cbind(out$model$z, out$model$classification)
     dff <- matrix(NA, dim(df)[1], dim(outdat)[2])
     dff[no_na_rows, ] <- outdat
     colnames(dff) <- c(paste0("CPROB", 1:ncol(out$model$z)), "Class")
     out$dff <- as_tibble(cbind(df, dff))
     out$dff$model_number <- this_model
     out$dff$classes_number <- this_class
     out$dff <- out$dff[, c((ncol(out$dff) - 1), ncol(out$dff), 1:(ncol(out$dff) -
     2))]
     warnings <- NULL
     if (out$fit[["prob_min"]] < 0.001)
     warnings <- c(warnings, "Some classes were not assigned any cases with more than .1% probability. Consequently, these solutions are effectively identical to a solution with one class less.")
     if (out$fit[["n_min"]] < 0.01)
     warnings <- c(warnings, "Less than 1% of cases were assigned to one of the profiles. Interpret this solution with caution and consider other models.")
     out$warnings <- warnings
     class(out) <- c("tidyProfile.mclust", "tidyProfile", "list")
     out
     })(this_class = dots[[1L]][[6L]], this_model = dots[[2L]][[6L]])
     15: MclustBootstrap(out$model, nboot = 100, type = "bs", verbose = FALSE)
     16: stop("object must be of class 'Mclust' or 'densityMclust'")
    
     == testthat results ===========================================================
     OK: 20 SKIPPED: 13 FAILED: 1
     1. Error: (unknown) (@test-compare-solutions.R#10)
    
     Error: testthat unit tests failed
     Execution halted
Flavor: r-devel-linux-x86_64-debian-clang

Version: 1.0.0
Check: tests
Result: ERROR
     Running ‘testthat.R’ [18s/27s]
    Running the tests in ‘tests/testthat.R’ failed.
    Complete output:
     > library(testthat)
     > library(tidyLPA)
     tidyLPA has received a major update, with a much easier workflow and improved functionality. However, you might have to update old syntax to account for the new workflow. See vignette('introduction-to-major-changes') for details!
    
     Mplus is not installed. Use only package = 'mclust' when calling estimate_profiles().
     Warning message:
     In system2("type", args = "mplus", stdout = FALSE, stderr = FALSE) :
     error in running command
     >
     > test_check("tidyLPA")
     ── 1. Error: (unknown) (@test-compare-solutions.R#10) ─────────────────────────
     replacement has length zero
     1: pisaUSA15[1:100, ] %>% single_imputation() %>% estimate_profiles(1:3, variances = c("equal",
     "varying"), covariances = c("zero", "varying")) %>% compare_solutions(statistics = c("AIC",
     "BIC")) at testthat/test-compare-solutions.R:10
     2: withVisible(eval(quote(`_fseq`(`_lhs`)), env, env))
     3: eval(quote(`_fseq`(`_lhs`)), env, env)
     4: eval(quote(`_fseq`(`_lhs`)), env, env)
     5: `_fseq`(`_lhs`)
     6: freduce(value, `_function_list`)
     7: function_list[[i]](value)
     8: estimate_profiles(., 1:3, variances = c("equal", "varying"), covariances = c("zero",
     "varying"))
     9: estimate_profiles.data.frame(., 1:3, variances = c("equal", "varying"), covariances = c("zero",
     "varying"))
     10: NextMethod("estimate_profiles", df)
     11: estimate_profiles.default(., 1:3, variances = c("equal", "varying"), covariances = c("zero",
     "varying"))
     12: estimate_profiles_mclust(df, n_profiles, model_numbers, ...)
     13: lapply(boot_model_names, function(mod_name) {
     mclustBootstrapLRT(full_data, modelName = mod_name, nboot = ifelse(methods::hasArg("nboot"),
     arg_list$nboot, 100), maxG = max(n_profiles), verbose = FALSE)
     })
     14: FUN(X[[i]], ...)
     15: mclustBootstrapLRT(full_data, modelName = mod_name, nboot = ifelse(methods::hasArg("nboot"),
     arg_list$nboot, 100), maxG = max(n_profiles), verbose = FALSE)
    
     ══ testthat results ═══════════════════════════════════════════════════════════
     OK: 21 SKIPPED: 13 FAILED: 1
     1. Error: (unknown) (@test-compare-solutions.R#10)
    
     Error: testthat unit tests failed
     Execution halted
Flavor: r-devel-linux-x86_64-debian-gcc

Version: 1.0.0
Check: dependencies in R code
Result: NOTE
    Namespace in Imports field not imported from: ‘mix’
     All declared Imports should be used.
Flavors: r-devel-linux-x86_64-fedora-clang, r-devel-linux-x86_64-fedora-gcc, r-patched-solaris-x86

Version: 1.0.0
Check: whether package can be installed
Result: WARN
    Found the following significant warnings:
     Warning: running command '"where" mplus.exe' had status 1
Flavor: r-oldrel-windows-ix86+x86_64