Ipulatov
Ipulatov

Reputation: 195

When I try to run modules from pycuda, I get the error: "No module named 'pycuda._driver'"

I want to run code that I got off this site - https://documen.tician.de/pycuda/tutorial.html - it's what's below.

import pycuda.gpuarray as gpuarray
import pycuda.driver as cuda
import pycuda.autoinit
import numpy

a_gpu = gpuarray.to_gpu(numpy.random.randn(4,4).astype(numpy.float32))
a_doubled = (2*a_gpu).get()
print(a_doubled)
print(a_gpu)

Basically, for the line "import pycuda.driver as cuda", I get the error message:

File "C:\Users\David\Anaconda3\lib\site-packages\pycuda\driver.py", line 5, in from pycuda._driver import * # noqa

ModuleNotFoundError: No module named 'pycuda._driver'

This makes sense, because when I look as the driver text file, I see the following lines

try:
    from pycuda._driver import *  # noqa
except ImportError as e:
    if "_v2" in str(e):
        from warnings import warn
        warn("Failed to import the CUDA driver interface, with an error "
                "message indicating that the version of your CUDA header "
                "does not match the version of your CUDA driver.")
    raise

And there is in fact no text file called _driver in my pycuda folder. So how can I fix this? I thought I should have all the folders when I write "pip install pycuda" in my terminal.

Upvotes: 0

Views: 8978

Answers (1)

sansa
sansa

Reputation: 904

Depending on your OS make sure you have configured pycuda on correct cuda lib path.. Follow https://wiki.tiker.net/PyCuda/Installation/Windows carefully.

HOWEVER: It looks like you want to install this on Anaconda, therefore I would advise you to follow this guide instead. https://www.ibm.com/developerworks/community/blogs/jfp/entry/Installing_PyCUDA_On_Anaconda_For_Windows?lang=en

Upvotes: 0

Related Questions