Chainer GPU about chainer-fast-neuralstyle HOT 3 OPEN

Hi, what is the message error ?
I'm on Win10 with nvidia & cudnn & GTX 980 and everything works fine

from chainer-fast-neuralstyle.

Microsoft Windows [Version 6.1.7601]
Copyright (c) 2009 Microsoft Corporation. All rights reserved.

E:_PYTOOLS\chainer-fast>python generate.py sample_images/snail.png -m models/se
urat.model -o results/snail_s.png -g 0
Traceback (most recent call last):
File "generate.py", line 30, in
y = model(x)
File "E:_PYTOOLS\chainer-fast\net.py", line 55, in call
h = self.b1(F.elu(self.c1(x)), test=test)
File "C:\Python27\lib\site-packages\chainer\links\connection\convolution_2d.py
", line 82, in call
x, self.W, self.b, self.stride, self.pad, self.use_cudnn)
File "C:\Python27\lib\site-packages\chainer\functions\connection\convolution_2
d.py", line 316, in convolution_2d
return func(x, W, b)
File "C:\Python27\lib\site-packages\chainer\function.py", line 130, in _call
_
outputs = self.forward(in_data)
File "C:\Python27\lib\site-packages\chainer\function.py", line 234, in forward

return self.forward_gpu(inputs)

File "C:\Python27\lib\site-packages\chainer\functions\connection\convolution_2
d.py", line 90, in forward_gpu
x = cuda.cupy.ascontiguousarray(x)
File "C:\Python27\lib\site-packages\cupy\creation\from_data.py", line 78, in a
scontiguousarray
return core.ascontiguousarray(a, dtype)
File "cupy/core/core.pyx", line 1427, in cupy.core.core.ascontiguousarray (cup
y\core\core.cpp:49842)
File "cupy/core/core.pyx", line 1437, in cupy.core.core.ascontiguousarray (cup
y\core\core.cpp:49724)
File "cupy/core/core.pyx", line 1293, in cupy.core.core.elementwise_copy (cupy
\core\core.cpp:48776)
File "cupy/core/elementwise.pxi", line 768, in cupy.core.core.ufunc.call (
cupy\core\core.cpp:39715)
File "cupy/util.pyx", line 36, in cupy.util.memoize.decorator.ret (cupy\util.c
pp:1264)
File "cupy/core/elementwise.pxi", line 576, in cupy.core.core._get_ufunc_kerne
l (cupy\core\core.cpp:36329)
File "cupy/core/elementwise.pxi", line 32, in cupy.core.core._get_simple_eleme
ntwise_kernel (cupy\core\core.cpp:26928)
File "cupy/core/carray.pxi", line 87, in cupy.core.core.compile_with_cache (cu
py\core\core.cpp:26615)
File "C:\Python27\lib\site-packages\cupy\cuda\compiler.py", line 119, in compi
le_with_cache
base = _empty_file_preprocess_cache[env] = preprocess('', options)
File "C:\Python27\lib\site-packages\cupy\cuda\compiler.py", line 82, in prepro
cess
pp_src = _run_nvcc(cmd, root_dir)
File "C:\Python27\lib\site-packages\cupy\cuda\compiler.py", line 44, in _run_n
vcc
raise RuntimeError(msg)
RuntimeError: nvcc command returns non-zero exit status.
command: ['nvcc', '--preprocess', '-Xcompiler', '/wd 4819', '-m64', 'c:\users
jimmy\appdata\local\temp\tmpo4ds4g\kern.cu']
return-code: 1
stdout/stderr:
nvcc fatal : Cannot find compiler 'cl.exe' in PATH

E:_PYTOOLS\chainer-fast>

from chainer-fast-neuralstyle.

Oh, I've been through that error. You have to install Microsoft Visual Studio and then CUDA. If it still doesn't work, you should tweak your environment PATH so that chainer can find cl.exe somethjng like C:\Program Files\Microsoft Visual Studio 10.0\VC\bin . However, maybe cl.exe can be found in another MS package, than Visual Studio, I don't know.

If it still doesn't work, try to use Miniconda2 (I use that).

from chainer-fast-neuralstyle.