search

twke18 / CAST

MIT License
29 stars 0 forks source link

Environment setting for dgl #6

Closed ByeongJuWoo closed 2 months ago

ByeongJuWoo commented 2 months ago

Hi, thank you for your great work!

I want to reproduce your code, but i have trouble in environment setting. After I initiate conda env cast, given pip install for dgl doesn't work. 

# install dgl (https://www.dgl.ai/pages/start.html)
> pip install dgl==1.1.3+cu116 -f https://data.dgl.ai/wheels/cu116/dgl-1.1.3%2Bcu116-cp38-cp38-manylinux1_x86_64.whl
ERROR: Could not find a version that satisfies the requirement dgl==1.1.3+cu116 (from versions: 0.1.0, 0.1.2, 0.1.3, 0.6.0, 0.6.0.post1, 0.6.1, 0.9.0, 0.9.1, 1.0.0, 1.0.1, 1.0.4, 1.1.0, 1.1.1, 1.1.2, 1.1.3, 2.1.0)
ERROR: No matching distribution found for dgl==1.1.3+cu116

In official dgl installation homepage they only support for torch 2.x. Also, I cannot find proper wheel file in repo, which is dgl==1.1.3+cu116.

How can I install proper dgl version for cuda 11.6?

twke18 commented 2 months ago

Hi,

Thanks for your interest. Can you confirm that you are using pytorch 1.13, cuda 11.6 and python 3.8?

If you use the same version of pytorch and cuda, but with a different version of python, you can search the matched one from the wheel repo.

Also, this is the wheel repo for cuda 11.8.

Let us know if you still have problems installing dgl.

ByeongJuWoo commented 2 months ago

The wheel repo you found works well. Thanks a lot!