Trainer for graph with billion-scale nodes

Questions
#1

Hi,
I have come across one interesting paper, “Distributed Hybrid CPU and GPU training for Graph Neural Networks on Billion-Scale Heterogeneous Graphs”, and have found the implementation is merged,I would like to ask where is the script and documentation?

Thanks

#2

DistDGLv2 has not been fully merged into DGL yet though distributed train on hetero-graph has been available since DGL 0.7.

Please refer to latest DGL doc page to see if it works for you: Chapter 7: Distributed Training — DGL 0.9.1post1 documentation

#3

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.