unable to load mxnet library in R

[wittyalias]

Looks to be closely related to #13936, and possibly related to the issues in #13952, but I'm not completely sure.

I've followed the instructions to install CPU version of MXNet for R on MacOS (10.14.2).
I successfully install openblas (0.3.5) and opencv (4.0.1) via brew using

brew install opencv
brew install openblas

Also, added the symbolic links as suggested (this is a different openblas version that I have installed):

ln -sf /usr/local/opt/openblas/lib/libopenblas.dylib /usr/local/opt/openblas/lib/libopenblasp-r0.3.1.dylib

And then installed the package within R with:

cran <- getOption("repos")
cran["dmlc"] <- "https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/R/CRAN/"
options(repos = cran)
install.packages("mxnet")

Description

When I try to load the mxnet library I get an error message below

>library(mxnet)
Error: package or namespace load failed for ‘mxnet’:
 .onLoad failed in loadNamespace() for 'mxnet', details:
  call: dyn.load(file, DLLpath = DLLpath, ...)
  error: unable to load shared object '/Library/Frameworks/R.framework/Versions/3.5/Resources/library/mxnet/libs/libmxnet.so':
  dlopen(/Library/Frameworks/R.framework/Versions/3.5/Resources/library/mxnet/libs/libmxnet.so, 10): Library not loaded: /usr/local/opt/openblas/lib/libopenblasp-r0.3.1.dylib
  Referenced from: /Library/Frameworks/R.framework/Versions/3.5/Resources/library/mxnet/libs/libmxnet.so
  Reason: image not found

Environment info (Required)

I'm using R in RStudio 1.1.463

R version 3.5.2 (2018-12-20)
Platform: x86_64-apple-darwin15.6.0 (64-bit)
Running under: macOS Mojave 10.14.2

Matrix products: default
BLAS: /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libBLAS.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/3.5/Resources/lib/libRlapack.dylib

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods  
[7] base     

What have you tried to solve it?

1. My guess is that its looking for different versions of openblas and opencv than I have installed, so I tried to install a previous versions of them via brew, but I can't figure out how to do it for openblas and 3.4 opencv installs to a folder 'opencv@3' - so the library won't be able to find it anyway.

I did not change the symbolic link because before I had the openblas 0.3.1 error, I had a very similar looking error that said it couldn't find opencv 3.4.

I was thinking of cloning the repository and seeing if I could change what openblas and opencv versions mxnet looks for, but that seems like a lot of work for something that you may already be looking into.

[mxnet-label-bot]

Hey, this is the MXNet Label Bot.
Thank you for submitting the issue! I will try and suggest some labels so that the appropriate MXNet community members can help resolve it.
Here are my recommended labels: Installation

[frankfliu]

@mxnet-label-bot add [build, R]

[saberioon]

I have the same issue, Is there any solution available?

my session info is :

R version 3.5.1 (2018-07-02)
Platform: x86_64-apple-darwin15.6.0 (64-bit)
Running under: macOS  10.14.3

Matrix products: default
BLAS: /Library/Frameworks/R.framework/Versions/3.5/Resources/lib/libRblas.0.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/3.5/Resources/lib/libRlapack.dylib

locale:
[1] C/UTF-8/C/C/C/C

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base

loaded via a namespace (and not attached):
 [1] Rcpp_1.0.0         lubridate_1.7.4    lattice_0.20-35    tidyr_0.8.2
 [5] class_7.3-14       visNetwork_2.0.5   assertthat_0.2.0   digest_0.6.18
 [9] ipred_0.9-8        foreach_1.4.4      R6_2.3.0           plyr_1.8.4
[13] stats4_3.5.1       ggplot2_3.1.0      pillar_1.3.0       rlang_0.3.0.1
[17] lazyeval_0.2.1     caret_6.0-81       rstudioapi_0.8     data.table_1.11.8
[21] kernlab_0.9-27     rpart_4.1-13       Matrix_1.2-14      DiagrammeR_1.0.0
[25] splines_3.5.1      gower_0.1.2        downloader_0.4     readr_1.3.1
[29] stringr_1.3.1      htmlwidgets_1.3    igraph_1.2.2       munsell_0.5.0
[33] compiler_3.5.1     influenceR_0.1.0   rgexf_0.15.3       pkgconfig_2.0.2
[37] htmltools_0.3.6    nnet_7.3-12        tidyselect_0.2.5   tibble_1.4.2
[41] gridExtra_2.3      prodlim_2018.04.18 codetools_0.2-15   XML_3.98-1.16
[45] viridisLite_0.3.0  crayon_1.3.4       dplyr_0.7.8        withr_2.1.2
[49] MASS_7.3-50        recipes_0.1.4      ModelMetrics_1.2.2 grid_3.5.1
[53] jsonlite_1.6       nlme_3.1-137       gtable_0.2.0       magrittr_1.5
[57] pROC_1.13.0        scales_1.0.0       stringi_1.2.4      reshape2_1.4.3
[61] viridis_0.5.1      bindrcpp_0.2.2     timeDate_3043.102  brew_1.0-6
[65] generics_0.0.2     lava_1.6.4         RColorBrewer_1.1-2 iterators_1.0.10
[69] tools_3.5.1        glue_1.3.0         purrr_0.2.5        hms_0.4.2
[73] Rook_1.1-1         survival_2.42-3    colorspace_1.3-2   bindr_0.1.1
>

[drag05]

I have installed mxnet for CPU and worked fine on Windows 10, R version 3.6.2. Then I wanted to try mxnet for GPU: I installed mxnet for GPU using the command:

install.packages("https://s3.ca-central-1.amazonaws.com/jeremiedb/share/mxnet/GPU/mxnet.zip", repos = NULL)

everything went file. However when trying to load in R version 3.6.2 or R Studio, I get the following error:

Loading required package: mxnet

Error: package or namespace load failed for ‘mxnet’: .onLoad failed in loadNamespace() for 'mxnet', details: call: inDL(x, as.logical(local), as.logical(now), ...) error: unable to load shared object 'C:/R/R-3.6.2/library/mxnet/libs/x64/libmxnet.dll': LoadLibrary failure: The specified module could not be found.

I have been using my GPU for R/keras/tensorflow having installed cuda100 on miniconda3. It works quite smoothly on NVIDIA Geforce RTX 2080

Please advise

In case this helps:

> sessionInfo()

R version 3.6.2 (2019-12-12)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 10 x64 (build 18363)

Matrix products: default

locale:

attached base packages:
[1] stats graphics grDevices utils datasets methods base

loaded via a namespace (and not attached):
[1] htmlwidgets_1.5.1 DiagrammeR_1.0.5 compiler_3.6.2 magrittr_1.5 visNetwork_2.0.9
[6] htmltools_0.4.0 tools_3.6.2 glue_1.3.1 RColorBrewer_1.1-2 Rcpp_1.0.3
[11] stringi_1.4.5 stringr_1.4.0 jsonlite_1.6.1 digest_0.6.23 rlang_0.4.4

Thank you!

[sbar365]

I am also trying to install mxnet for R on MacOS (10.15.4). I installed openblas 0.3.9 and opencv 4.2.0_3.

cran <- getOption("repos")
cran["dmlc"] <- "https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/R/CRAN/"
options(repos = cran)
install.packages("mxnet")
library(mxnet)

and I get the following error:

[1] "Loading local: inst/libs/libmxnet.so"
Error: package or namespace load failed for ‘mxnet’:
.onLoad failed in loadNamespace() for 'mxnet', details:
call: dyn.load("R-package/inst/libs/libmxnet.so", local = FALSE)
error: unable to load shared object '/Users/Stefanos/R-package/inst/libs/libmxnet.so':
dlopen(/Users/Stefanos/R-package/inst/libs/libmxnet.so, 10): image not found

sessionInfo()
R version 3.5.2 (2018-12-20)
Platform: x86_64-apple-darwin15.6.0 (64-bit)
Running under: macOS 10.15.4

Matrix products: default
BLAS: /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libBLAS.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/3.5/Resources/lib/libRlapack.dylib

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats graphics grDevices utils datasets methods base

other attached packages:
[1] devtools_2.2.2 usethis_1.5.1

loaded via a namespace (and not attached):
[1] Rcpp_1.0.4 tidyr_1.0.2 prettyunits_1.1.1 ps_1.3.2 visNetwork_2.0.9 assertthat_0.2.1
[7] rprojroot_1.3-2 digest_0.6.25 R6_2.4.1 backports_1.1.5 ggplot2_3.3.0 pillar_1.4.3
[13] rlang_0.4.5 rstudioapi_0.11 callr_3.4.3 drat_0.1.5 DiagrammeR_0.9.2 desc_1.2.0
[19] downloader_0.4 servr_0.16 readr_1.3.1 stringr_1.4.0 htmlwidgets_1.5.1 igraph_1.2.5
[25] munsell_0.5.0 compiler_3.5.2 influenceR_0.1.0 rgexf_0.16.0 httpuv_1.5.2 xfun_0.12
[31] pkgconfig_2.0.3 pkgbuild_1.0.6 htmltools_0.4.0 tidyselect_1.0.0 tibble_3.0.0 gridExtra_2.3
[37] XML_3.99-0.3 fansi_0.4.1 viridisLite_0.3.0 crayon_1.3.4 dplyr_0.8.5 withr_2.1.2
[43] later_1.0.0 grid_3.5.2 jsonlite_1.6.1 gtable_0.3.0 lifecycle_0.2.0 magrittr_1.5
[49] scales_1.1.0 neuralnet_1.44.2 cli_2.0.2 stringi_1.4.6 viridis_0.5.1 fs_1.3.2
[55] promises_1.1.0 remotes_2.1.1 testthat_2.3.2 ellipsis_0.3.0 vctrs_0.2.4 RColorBrewer_1.1-2
[61] tools_3.5.2 glue_1.3.2 purrr_0.3.3 hms_0.5.3 processx_3.4.2 pkgload_1.0.2
[67] yaml_2.2.1 colorspace_1.4-1 sessioninfo_1.1.1 memoise_1.1.0

Any ideas how to fix this issue?
Thank you.