gpgpu - How to use theano within screen session? -
i use theano on remote server first ssh (i don't have root on system). works fine, however, if start screen
, error when trying import theano.
behavior when not using screen
:
>>> import theano using gpu device 0: geforce gtx titan x (cnmem disabled, cudnn 4007)
behavior when using screen
:
>>> import theano error (theano.sandbox.cuda): failed compile cuda_ndarray.cu: libcublas.so.7.5: cannot open shared object file: no such file or directory traceback (most recent call last): file "<stdin>", line 1, in <module> file "/home/2012/enewel3/.local/lib/python2.7/site-packages/theano/__init__.py", line 103, in <module> import theano.sandbox.cuda file "/home/2012/enewel3/.local/lib/python2.7/site-packages/theano/sandbox/cuda/__init__.py", line 697, in <module> use(device=config.device, force=config.force_device, test_driver=false) file "/home/2012/enewel3/.local/lib/python2.7/site-packages/theano/sandbox/cuda/__init__.py", line 496, in use device, cuda_initialization_error_message)) environmenterror: forced use of gpu device gpu, cuda initialization failed error: cuda unavailable
how should use theano within screen session?
the problem due fact that, although screen inherits environment variables, adds, removes, , alters of them. in case, changing value of ld_library_path
.
upon entering screen, manually setting environment variable it's proper value fixed problem. me, looks this:
export ld_library_path=/usr/local/cuda/lib64:/usr/local/pkgs/gurobi502/linux64/lib
adding line .bashrc
makes sure screens right value. note, adding .bash_profile won't work, because .bash_profile run @ login, while .bashrc run every new shell.
Comments
Post a Comment