Baole Ai
Baole Ai
Hi, thanks for your interest. Graph-Learn started out by encapsulating GNN models in a dense EgoGraph fashion for large-scale graphs, and built models primarily based on TensorFlow 1.12. With the...
Yes, we partition and store the original graph in a distributed manner, and then sample and train from the original graph. EgoGraph is just the organization form of the sampling...
Not yet, do you have any suggested metrics? We can follow up with a comparison and analysis.
No, we removed this part of the code in the open source version because the previous coupling was a bit high. We are designing new feature storage and caching solutions...
Please try v1.1.0.
Hi, it seems that you use v1.0.1 pip wheel but run example at master branch. We will release v1.1.0(check out from master) asap. For now, you can just checkout branch...
Could you provide more details (e.g., your graph-learn version, error log.)
https://github.com/alibaba/graph-learn/blob/master/graphlearn/src/core/partition/hash_partitioner.h
Please provide information about your compilation environment commands, etc.
What is your compilation environment and do you get any errors after `make`?