CRAN Package Check Results for Package xgboost

Last updated on 2019-10-16 22:56:14 CEST.

Flavor Version Tinstall Tcheck Ttotal Status Flags
r-devel-linux-x86_64-debian-clang 0.90.0.2 111.39 200.43 311.82 NOTE
r-devel-linux-x86_64-debian-gcc 0.90.0.2 111.42 166.14 277.56 NOTE
r-devel-linux-x86_64-fedora-clang 0.90.0.2 395.88 NOTE
r-devel-linux-x86_64-fedora-gcc 0.90.0.2 417.87 NOTE
r-devel-windows-ix86+x86_64 0.90.0.2 276.00 348.00 624.00 NOTE
r-patched-linux-x86_64 0.90.0.2 107.73 179.65 287.38 NOTE
r-patched-solaris-x86 0.90.0.2 613.70 ERROR
r-release-linux-x86_64 0.90.0.2 104.84 182.70 287.54 NOTE
r-release-windows-ix86+x86_64 0.90.0.2 199.00 355.00 554.00 NOTE
r-release-osx-x86_64 0.90.0.2 NOTE
r-oldrel-windows-ix86+x86_64 0.90.0.2 195.00 236.00 431.00 NOTE
r-oldrel-osx-x86_64 0.90.0.2 NOTE

Check Details

Version: 0.90.0.2
Check: for GNU extensions in Makefiles
Result: NOTE
    GNU make is a SystemRequirements.
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: 0.90.0.2
Check: installed package size
Result: NOTE
     installed size is 21.8Mb
     sub-directories of 1Mb or more:
     libs 21.0Mb
Flavors: r-devel-linux-x86_64-fedora-clang, r-devel-windows-ix86+x86_64, r-patched-solaris-x86, r-release-windows-ix86+x86_64, r-release-osx-x86_64, r-oldrel-windows-ix86+x86_64, r-oldrel-osx-x86_64

Version: 0.90.0.2
Check: examples
Result: ERROR
    Running examples in ‘xgboost-Ex.R’ failed
    The error most likely occurred in:
    
    > ### Name: cb.gblinear.history
    > ### Title: Callback closure for collecting the model coefficients history
    > ### of a gblinear booster during its training.
    > ### Aliases: cb.gblinear.history
    >
    > ### ** Examples
    >
    > #### Binary classification:
    > #
    > # In the iris dataset, it is hard to linearly separate Versicolor class from the rest
    > # without considering the 2nd order interactions:
    > require(magrittr)
    Loading required package: magrittr
    > x <- model.matrix(Species ~ .^2, iris)[,-1]
    > colnames(x)
     [1] "Sepal.Length" "Sepal.Width"
     [3] "Petal.Length" "Petal.Width"
     [5] "Sepal.Length:Sepal.Width" "Sepal.Length:Petal.Length"
     [7] "Sepal.Length:Petal.Width" "Sepal.Width:Petal.Length"
     [9] "Sepal.Width:Petal.Width" "Petal.Length:Petal.Width"
    > dtrain <- xgb.DMatrix(scale(x), label = 1*(iris$Species == "versicolor"))
    > param <- list(booster = "gblinear", objective = "reg:logistic", eval_metric = "auc",
    + lambda = 0.0003, alpha = 0.0003, nthread = 2)
    > # For 'shotgun', which is a default linear updater, using high eta values may result in
    > # unstable behaviour in some datasets. With this simple dataset, however, the high learning
    > # rate does not break the convergence, but allows us to illustrate the typical pattern of
    > # "stochastic explosion" behaviour of this lock-free algorithm at early boosting iterations.
    > bst <- xgb.train(param, dtrain, list(tr=dtrain), nrounds = 200, eta = 1.,
    + callbacks = list(cb.gblinear.history()))
    Error: segfault from C stack overflow
    Execution halted
Flavor: r-patched-solaris-x86

Version: 0.90.0.2
Check: tests
Result: ERROR
     Running ‘testthat.R’
    Running the tests in ‘tests/testthat.R’ failed.
    Complete output:
     > library(testthat)
     > library(xgboost)
     >
     > test_check("xgboost")
     [1] train-error:0.047136
     Error: segfault from C stack overflow
     Execution halted
Flavor: r-patched-solaris-x86