A problem of install

[xiaosa269]

Hello,

Thank you very much for your pytorch_scatter library. I'm using CUDA: 11.1 and torch: 1.8.1. After installing torch, I conducted some tests, and it seems that the GPU is accessible.

However, after installing pytorch-scatter using conda install pytorch-scatter -c pyg, when I entered Python for testing, the following issue occurred: (frnet) xhy@sun:~/code/FRNet-master$ python Python 3.8.0 (default, Nov 6 2019, 21:49:08) [GCC 7.3.0] :: Anaconda, Inc. on linux Type "help", "copyright", "credits" or "license" for more information.

import torch torch.cuda.is_available() False

After uninstalling pytorch-scatter using conda uninstall pytorch-scatter, I tested torch again: (frnet) xhy@sun:~/code/FRNet-master$ python Python 3.8.0 (default, Nov 6 2019, 21:49:08) [GCC 7.3.0] :: Anaconda, Inc. on linux Type "help", "copyright", "credits" or "license" for more information.

import torch torch.cuda.is_available() Traceback (most recent call last): File "", line 1, in AttributeError: module 'torch' has no attribute 'cuda'

Could you please advise on why this is happening? Your assistance would be greatly appreciated. Thank you very much.

[rusty1s]

Try torch-scatter==2.0.8 for PyTorch 1.8, see here.

[github-actions[bot]]

This issue had no activity for 6 months. It will be closed in 2 weeks unless there is some new activity. Is this issue already resolved?