Error while 'make' cudamat

[Rietaros]

Hello, I want to start using deepnet (https://github.com/nitishsrivastava/deepnet). For some reason it was built on top of cudamat, so I need to install and i read the tutorial. But I have trouble while i call 'make' file like this:

~/Documents/deepnet-master/cudamat$ sudo make
nvcc -O3
-v
-gencode=arch=compute_10,code=sm_10
-gencode=arch=compute_20,code=sm_20
-gencode=arch=compute_30,code=sm_30
--compiler-options '-fPIC' -o libcudamat.so
--shared cudamat.cu cudamat_kernels.cu -lcublas -L
nvcc fatal : argument expected after '-L'
make: *** [libcudamat.so] Error 255

why it happen?
thank you

[Rietaros]

Solved.., I forgot to set the environment vari

Now the error become : able

unsupported gpu architecture 'compute_10'

I'm using cuda 7.5.

I need to edit the makefile:

-gencode=arch=compute_10,code=sm_10
-gencode=arch=compute_20,code=sm_20
-gencode=arch=compute_30,code=sm_30 \

what should I type to replace tha compute_10?

[Rietaros]

done by editing compute_10 to compute_30 depends on cuda version

And I come up with another problem while running the RBM examples like:

Traceback (most recent call last):
  File "../../trainer.py", line 1, in <module>
    from neuralnet import *
  File "/Documents/deepnet-master/deepnet/neuralnet.py", line 8, in <module>
    from datahandler import *
  File "/Documents/deepnet-master/deepnet/datahandler.py", line 4, in <module>e
    from choose_matrix_library import *
  File "/Documents/deepnet-master/deepnet/choose_matrix_library.py", line 13, in <module>
    from cudamat import cudamat_conv as cc
ImportError: cannot import name cudamat_conv

I have add the cudamat into my PYTHONPATH
export PYTHONPATH=$PYTHONPATH:$HOME/Documents/deepnet-master/cudamat
but still facing the problem.

But when I run:
import cudamat_conv
in my python console, nothing wrong

Any idea?

[Rietaros]

Now I just figure it out

in my choose.matrix.library.py line 13, instead of

from cudamat import cudamat_conv as cc
from cudamat import gpu_lock

I think it'll work with

import cudamat_conv as cc
import gpu_lock

because I've tested it in my python console. But after I edit choose.matrix.library.py, it looks my deepnet doesn't "refresh" the code and still have the same problem output.

Why the code doesn't "resfresh" ?