Unable to run test [generalized_rspmm related error]

#1
by valiavee - opened

Hi,

Thank you for the great package. Could you please help me resolve an error associated with importing 'generalized_rspmm'?

Issue: When I run test, it results in the following error which I assume is related to 'generalized_rspmm'

test(model, mode="test", dataset=dataset, gpus=None)

Error: 'subprocess.CalledProcessError: Command '['which', 'c++']' returned non-zero exit status 1. ' .

Platform - WSL (Linux)

Cause of error after investigation

I was able to reproduce and get the same error by running the following command in a Python terminal:

from ultra.rspmm.rspmm import generalized_rspmm

Specifically in file rspmm.py, line 199 "cpp_extension.load(name, sources, extra_cflags, extra_cuda_cflags, **kwargs)" is not working for me.
I also made sure that the 'path = os.path.join(os.path.dirname(file), "source")' in line 203 points to the right file.

May I know how to fix this error?

Owner

Hi,
you probably don't have C++ build tools installed, please follow the torchdrug docs to install one on Windows: https://github.com/DeepGraphLearning/torchdrug?tab=readme-ov-file#windows-powershell

Thank you @mgalkin . I will give a try.

Sign up or log in to comment