search

snap-stanford / ogb

Benchmark datasets, data loaders, and evaluators for graph machine learning
https://ogb.stanford.edu
MIT License
1.89k stars 397 forks source link

is there any plan to add dynamic or temporal graph dataset for benchmark evaluation? #429

Closed hanzhichao2000 closed 10 months ago

hanzhichao2000 commented 1 year ago

in this article, the author said "We are now working on creating new dynamic graph datasets and tasks as part of the Open Graph Benchmark."

is there any progress?

weihua916 commented 1 year ago

Yes, that's ongoing. We will let you know once we reach there (hopefully in the next few months).

LucaCappelletti94 commented 1 year ago

Hi @weihua916, I may have just the thing ready for these tasks. The graphs I have mentioned in this other issue are released in permalinks every month or so, so you have this rather large list of evolving graphs with many checkpoints. Several of these graphs have both node and edge labels and have features associated with both nodes, node types and edges.

Possibly they just might do the trick, and since these graphs are actively used in research, getting good predictions would be readily useful to the team at Berkeley that develops them.

weihua916 commented 10 months ago

Our TGB paper is online: https://arxiv.org/abs/2307.01026