unable to load mxnet in Rstudio using GPU

[mpearmain]

changed config.mk

whether use CUDA during compile

USE_CUDA = 1

add the path to CUDA libary to link and compile flag

if you have already add them to enviroment variable, leave it as NONE

USE_CUDA_PATH = /usr/local/cuda
#USE_CUDA_PATH = NONE

everything builds and R-package installs

Compiles and can load via CLI

library(mxnet)

Within Rstudio fails

library(mxnet)
Error : .onLoad failed in loadNamespace() for 'mxnet', details:
call: dyn.load(file, DLLpath = DLLpath, ...)
error: unable to load shared object '/home/mpearmain/R/x86_64-pc-linux-gnu-library/3.2/mxnet/libs/libmxnet.so':
libcudart.so.7.0: cannot open shared object file: No such file or directory
Error: package or namespace load failed for ‘mxnet’

[tqchen]

You want to add the path to /usr/local/cuda to LD_LIBRARY_PATH on your environment variable

[dselivanov]

On my ubuntu machine I added following strings to .profile:

export CUDA_HOME=/usr/local/cuda 
export LD_LIBRARY_PATH=${CUDA_HOME}/lib64:${LD_LIBRARY_PATH}

Works for me.

[Sat9443321]

I did the same in windows platform. but it returns the same error "enable the cuda". Please suggest how to proceed the gpu computing with windows platform.

[thirdwing]

@Sat9443321 Can you give more info on error "enable the cuda"?