CUDA is installed, but device gpu is not available (error: cuda unavailable)

Question

I've spent a few days trying to use my GPU with theano in Windows, but I've encountered several problems and after trying everything I've found in Google it still doesn't work.

The error I get is the following one:

===============================
nvcc warning : The 'compute_20', 'sm_20', and 'sm_21' architectures are deprecated, and may be removed in a future release (Use -Wno-deprecated-gpu-targets to suppress warning).
mod.cu
nvcc error   : 'cl.exe' died with status 0xC0000135

['nvcc', '-shared', '-O3', '--compiler-bindir', 'D:\\Programas\\Visual Studio\\2012\\VC\\bin', '-Xlinker', '/DEBUG', '-D HAVE_ROUND', '-m64', '-Xcompiler', '-DCUDA_NDARRAY_CUH=c72d035fdf91890f3b36710688069b2e,-DNPY_NO_DEPRECATED_API=NPY_1_7_API_VERSION,/Zi,/MD', '-ID:\\Anaconda2\\lib\\site-packages\\theano\\sandbox\\cuda', '-ID:\\Anaconda2\\lib\\site-packages\\numpy\\core\\include', '-ID:\\Anaconda2\\include', '-ID:\\Anaconda2\\lib\\site-packages\\theano\\gof', '-o', 'C:\\Users\\axelp\\AppData\\Local\\Theano\\compiledir_Windows-10-10.0.14393-Intel64_Family_6_Model_94_Stepping_3_GenuineIntel-2.7.13-64\\cuda_ndarray\\cuda_ndarray.pyd', 'mod.cu', '-LD:\\Anaconda2\\libs', '-LD:\\Anaconda2', '-lcublas', '-lpython27', '-lcudart']
ERROR (theano.sandbox.cuda): Failed to compile cuda_ndarray.cu: ('nvcc return status', 53, 'for cmd', 'nvcc -shared -O3 --compiler-bindir D:\\Programas\\Visual Studio\\2012\\VC\\bin -Xlinker /DEBUG -D HAVE_ROUND -m64 -Xcompiler -DCUDA_NDARRAY_CUH=c72d035fdf91890f3b36710688069b2e,-DNPY_NO_DEPRECATED_API=NPY_1_7_API_VERSION,/Zi,/MD -ID:\\Anaconda2\\lib\\site-packages\\theano\\sandbox\\cuda -ID:\\Anaconda2\\lib\\site-packages\\numpy\\core\\include -ID:\\Anaconda2\\include -ID:\\Anaconda2\\lib\\site-packages\\theano\\gof -o C:\\Users\\axelp\\AppData\\Local\\Theano\\compiledir_Windows-10-10.0.14393-Intel64_Family_6_Model_94_Stepping_3_GenuineIntel-2.7.13-64\\cuda_ndarray\\cuda_ndarray.pyd mod.cu -LD:\\Anaconda2\\libs -LD:\\Anaconda2 -lcublas -lpython27 -lcudart')
 WARNING (theano.sandbox.cuda): CUDA is installed, but device gpu is not available  (error: cuda unavailable)

Information about my installation:

1. I've installed CUDA v8.0 from the official NVidia site

2. My GPU is a GEFORCE GTX 960M

3. I've installed Visual Studio 2012. I previously tried to used the 2015 and 2017 versions, but I got a lot of errors (if I remember correctly, it stated that the architecture was unsupported) and saw that this one has better support. In case it matters, I didn't select the default directory (as it's on the SSD and I always try to save space on it), but on the regular HDD.

4. I use Windows 10 Home 64 bits

5. My Python version is the 2.7.13 from Anaconda

6. I've created a .theanorc.txt file in my User root directory with the following content:

   #!sh
   [global]
   device = gpu
   floatX = float32
   
   [nvcc]
   compiler_bindir=D:\Programas\Visual Studio\2012\VC\bin
   

7. My theano version is 0.8.2

Does anyone know how to solve it? Of course, don't hesitate to ask if I need to provide more information. Thank you so much in advance for your help!