Closed LMXB closed 5 years ago
Hi Yi, thank you for reporting this. I’ll take a look once I get to the office today.
You are using Intel compilers here which I have never tested before. This would be a great opportunity for me to test it with a different environment. Let's try to solve this problem with the Intel compiler you have. If it turns out to be too complicated, I might suggest you change your compiler to GCC for simplicity.
The error seems to reflect a conflict in some installed packages. Could you please show me the output of the following commands?
installed.packages()[c('BH', 'Rcpp'), 'Version']
rownames(installed.packages())
Hi Weiming, I found it works well with GCC compiler. Next, you may want to test it with Intel compiler, we can try to fix this.
Here is the output of these commands.
> installed.packages()[c('BH', 'Rcpp'), 'Version']
BH Rcpp
"1.69.0-1" "1.0.1"
> rownames(installed.packages())
[1] "abind" "ADGofTest" "akima"
[4] "askpass" "assertthat" "backports"
[7] "base64enc" "BH" "bitops"
[10] "BMA" "callr" "caTools"
[13] "CDFt" "cellranger" "chron"
[16] "CircStats" "cli" "climdex.pcic"
[19] "climdex.pcic.ncdf" "clipr" "clisymbols"
[22] "colorspace" "combinat" "copula"
[25] "corrplot" "crayon" "curl"
[28] "DataLoader" "DBI" "deepnet"
[31] "DEoptimR" "desc" "devtools"
[34] "dichromat" "digest" "doMC"
[37] "doParallel" "dotCall64" "dtw"
[40] "ellipsis" "ensembleBMA" "EnvStats"
[43] "esd" "eva" "evaluate"
[46] "evd" "fansi" "fields"
[49] "fitdistrplus" "forcats" "foreach"
[52] "formatR" "fs" "functional"
[55] "gdata" "generics" "ggplot2"
[58] "gh" "git2r" "glmnet"
[61] "glue" "goftest" "gplots"
[64] "gridExtra" "gridGraphics" "gsl"
[67] "gtable" "gtools" "highr"
[70] "htmltools" "htmlwidgets" "httpuv"
[73] "httr" "ini" "inline"
[76] "ismev" "iterators" "JBTools"
[79] "jsonlite" "knitr" "labeling"
[82] "latticeExtra" "lazyeval" "leaps"
[85] "lmomco" "Lmoments" "lmtest"
[88] "loadeR" "loadeR.java" "locfit"
[91] "lsei" "lubridate" "magrittr"
[94] "mapplots" "maps" "markdown"
[97] "matlabr" "memoise" "mime"
[100] "munsell" "mvtnorm" "ncdf"
[103] "ncdf.tools" "ncdf4" "ncdf4.helpers"
[106] "network" "nortest" "npsurv"
[109] "numDeriv" "openssl" "pcaPP"
[112] "PCICt" "pillar" "pkgbuild"
[115] "pkgconfig" "pkgload" "plogr"
[118] "plotrix" "plyr" "praise"
[121] "prettyunits" "processx" "proj4"
[124] "proxy" "ps" "pspline"
[127] "purrr" "qrng" "quadprog"
[130] "R.matlab" "R.methodsS3" "R.oo"
[133] "R.utils" "R6" "RAnEn"
[136] "raster" "rChoiceDialogs" "rcmdcheck"
[139] "RCMIP5" "RColorBrewer" "Rcpp"
[142] "RcppArmadillo" "RcppEigen" "RCurl"
[145] "rematch" "remotes" "reshape2"
[148] "rgl" "rJava" "rlang"
[151] "RNetCDF" "robustbase" "rprojroot"
[154] "rrcov" "rstudioapi" "rvest"
[157] "sandwich" "scales" "sdsmR"
[160] "seas" "selectr" "sessioninfo"
[163] "shiny" "snow" "sourcetools"
[166] "sp" "spam" "SPEI"
[169] "stabledist" "stringi" "stringr"
[172] "sys" "termstrc" "testthat"
[175] "tibble" "tinytex" "transformeR"
[178] "udunits2" "urca" "usethis"
[181] "utf8" "vctrs" "verification"
[184] "viridisLite" "whisker" "withr"
[187] "xfun" "xlsx" "xlsxjars"
[190] "xml2" "xopen" "xtable"
[193] "yaml" "zeallot" "zoo"
[196] "base" "boot" "class"
[199] "cluster" "codetools" "compiler"
[202] "datasets" "foreign" "graphics"
[205] "grDevices" "grid" "KernSmooth"
[208] "lattice" "MASS" "Matrix"
[211] "methods" "mgcv" "nlme"
[214] "nnet" "parallel" "rpart"
[217] "spatial" "splines" "stats"
[220] "stats4" "survival" "tcltk"
[223] "tools" "utils"
By searching around the Internet, people have encountered similar issues like this (errors triggered by the same file in BH and by the same line of code). Looks like this can be caused by several reasons:
But I have managed to install RAnEn
using icpc
/icc
on a cluster machine using either -std=gnu++11
or -std=c++11
. My Rcpp
and BH
versions are the same as yours.
So I guess this is going down the direction that a mysterious package is creating the conflict. People have reported StanHeaders
and bigstatsr
. But apparently you don't have them.
Since you already have a working version with GCC, I would suggest you use that one. Please let me know if you particular reason for Intel compilers.
I have the following errors while installing the RAnEn package on Linux.