[DNM] test which version of DGL will work

choderalab / espaloma

Extensible Surrogate Potential of Ab initio Learned and Optimized by Message-passing Algorithm 🍹https://arxiv.org/abs/2010.01196

https://docs.espaloma.org/en/latest/

MIT License

212 stars 23 forks source link

[DNM] test which version of DGL will work #177

Closed mikemhenry closed 1 year ago

codecov-commenter commented 1 year ago

Codecov Report

:exclamation: No coverage uploaded for pull request base (main@7165038). Click here to learn what that means. The diff coverage is n/a.

mikemhenry commented 1 year ago

It looks like we need to use 0.9.0 only because it looks like dglteam builds their dgl.sparse package (which is new) against pytorch from defaults so when we install pytorch from conda-forge we get an error. The older dgl packages don't do this, which is why they work. Our dgl package that we are getting onto conda-forge builds/links against the correct packages, so while it looks like newer dgl versions don't work, they should!

mikemhenry commented 1 year ago

Looks like dgl<1 is what we want