deepnet icon indicating copy to clipboard operation
deepnet copied to clipboard

OSError: libeigenmat.so: cannot open shared object file: No such file or directory

Open tengshaofeng opened this issue 10 years ago • 4 comments

hello, when i run runall_dbn.sh,everything is ok. But when i install IDE of wingIde,and set the environment,when i run a test.py and do "from deepnet import datahandler as dh",it reports the error like this:

OSError: libeigenmat.so: cannot open shared object file: No such file or directory File "/home/tbq/Downloads/deepnet-master/test.py", line 5, in from deepnet import dbm File "/home/tbq/Downloads/deepnet-master/deepnet/dbm.py", line 2, in from neuralnet import * File "/home/tbq/Downloads/deepnet-master/deepnet/neuralnet.py", line 8, in from datahandler import * File "/home/tbq/Downloads/deepnet-master/deepnet/datahandler.py", line 4, in from choose_matrix_library import * File "/home/tbq/Downloads/deepnet-master/deepnet/choose_matrix_library.py", line 16, in import eigenmat as cm File "/home/tbq/Downloads/deepnet-master/eigenmat/init.py", line 1, in from eigenmat import * File "/home/tbq/Downloads/deepnet-master/eigenmat/eigenmat.py", line 10, in _eigenmat = ct.cdll.LoadLibrary('libeigenmat.so') File "/usr/lib/python2.7/ctypes/init.py", line 443, in LoadLibrary return self._dlltype(name) File "/usr/lib/python2.7/ctypes/init.py", line 365, in init self._handle = _dlopen(self._name, mode)

then i change LoadLibrary('libcudamat.so') to LoadLibrary('/home/tbq/Downloads/deepnet-master/cudamat/libcudamat.so'),it does not works,too. i know error comes from "import cudamat".But when i import in terminal after python ,it is ok. it does not work in wingide. i need your help,thanks in advance

tengshaofeng avatar Nov 11 '14 11:11 tengshaofeng

same problem!!!

lireagan avatar Dec 10 '14 06:12 lireagan

Got the same problem when I try to run the example for rbm. Here's the error:-

titan4@titan4-Precision-Tower-5810:~/deepnet/deepnet/examples/rbm$ ./runall.sh Failed to import cudamat. Using eigenmat. No GPU will be used. Traceback (most recent call last): File "../../trainer.py", line 1, in from neuralnet import * File "/home/titan4/deepnet/deepnet/neuralnet.py", line 8, in from datahandler import * File "/home/titan4/deepnet/deepnet/datahandler.py", line 4, in from choose_matrix_library import * File "/home/titan4/deepnet/deepnet/choose_matrix_library.py", line 16, in import eigenmat as cm File "/home/titan4/deepnet/eigenmat/init.py", line 1, in from eigenmat import * File "/home/titan4/deepnet/eigenmat/eigenmat.py", line 10, in _eigenmat = ct.cdll.LoadLibrary('libeigenmat.so') File "/usr/lib/python2.7/ctypes/init.py", line 443, in LoadLibrary return self._dlltype(name) File "/usr/lib/python2.7/ctypes/init.py", line 365, in init self._handle = _dlopen(self._name, mode) OSError: libeigenmat.so: cannot open shared object file: No such file or directory

Any help appreciated :)

ghost avatar Nov 19 '15 12:11 ghost

I also get the same error!

rivergold avatar Sep 10 '16 10:09 rivergold

I also have the similar problem. I rerun the make file in the cudamat directory and reduce the gpu memory (gpu_mem) in runall_dbn.sh for the multi modal data. Now, it works.

a-taherkhani avatar Mar 29 '17 10:03 a-taherkhani