Missing nvcc compiler - theano - environment-variables

Question

I use ubuntu 14.04 and cuda 7.5. I get cuda version information using $ nvcc --version :
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2015 NVIDIA Corporation
Built on Tue_Aug_11_14:27:32_CDT_2015
Cuda compilation tools, release 7.5, V7.5.17
$PATH and $LD_LIBRARY_PATH are below :
$ echo $PATH
/usr/local/cuda-7.5/bin:/usr/local/cuda-7.5/bin/:/opt/ros/indigo/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games
$ echo $LD_LIBRARY_PATH
/usr/local/cuda-7.5/lib64
I install theano. I use it with cpu but not gpu. This guide says that
Testing Theano with GPU¶
To see if your GPU is being used, cut and paste the following program into a file and run it.
from theano import function, config, shared, sandbox import
> theano.tensor as T import numpy import time
>
> vlen = 10 * 30 * 768 # 10 x #cores x # threads per core iters = 1000
>
> rng = numpy.random.RandomState(22) x =
> shared(numpy.asarray(rng.rand(vlen), config.floatX)) f = function([],
> T.exp(x)) print(f.maker.fgraph.toposort()) t0 = time.time() for i in
> range(iters):
> r = f() t1 = time.time() print("Looping %d times took %f seconds" % (iters, t1 - t0)) print("Result is %s" % (r,)) if
> numpy.any([isinstance(x.op, T.Elemwise) for x in
> f.maker.fgraph.toposort()]):
> print('Used the cpu') else:
> print('Used the gpu') The program just computes the exp() of a bunch of random numbers. Note that we use the shared function to make
> sure that the input x is stored on the graphics device.
If I run this program (in check1.py) with device=cpu, my computer
takes a little over 3 seconds, whereas on the GPU it takes just over
0.64 seconds. The GPU will not always produce the exact same floating-point numbers as the CPU. As a benchmark, a loop that calls
numpy.exp(x.get_value()) takes about 46 seconds.
$ THEANO_FLAGS=mode=FAST_RUN,device=cpu,floatX=float32 python
check1.py [Elemwise{exp,no_inplace}()]
Looping 1000 times took 3.06635117531 seconds Result is [ 1.23178029
1.61879337 1.52278066 ..., 2.20771813 2.29967761
1.62323284] Used the cpu
$ THEANO_FLAGS=mode=FAST_RUN,device=gpu,floatX=float32 python
check1.py Using gpu device 0: GeForce GTX 580
[GpuElemwise{exp,no_inplace}(),
HostFromGpu(GpuElemwise{exp,no_inplace}.0)] Looping 1000 times took
0.638810873032 seconds Result is [ 1.23178029 1.61879349 1.52278066 ..., 2.20771813 2.29967761
1.62323296] Used the gpu Note that GPU operations in Theano require for now floatX to be float32 (see also below).
I run gpu version command without sudo, it throws permission denied error :
/theano/gof/cmodule.py", line 741, in refresh
files = os.listdir(root)
OSError: [Errno 13] Permission denied: '/home/user/.theano/compiledir_Linux-3.16--generic-x86_64-with-Ubuntu-14.04-trusty-x86_64-2.7.6-64/tmp077r7U'
If I use it with sudo, the compiler cannot find nvcc path.
ERROR (theano.sandbox.cuda): nvcc compiler not found on $PATH. Check your nvcc installation and try again.
How can I fix this error?

Answer 1

Try running
chown -R user /home/user/.theano
chmod -R 775 /home/user/.theano
this will change the permissions of the folder that your python script can't access. The first one will make the folder belong to your user and the second one will change the permissions to be readable, writable and executable by the user.

Answer 2

Regarding this error only:
You can check where your NVCC is installed , default path is '/usr/local/cuda/bin', if you could see it there then do as below:
$ export PATH="/usr/local/cuda/bin:$PATH"
$ source .bashrc
This worked for me and now I can use NVCC and it is no longer missing.