Support for incremental training?
bdeng3 opened this issue · comments
Currently if we wanted to train a model, we call the "fit" method who takes in a list of networkX graphs.
What if we have a really large number of graphs that can not be fit into RAM, is incremental training supported?
Do you want to open a PR?
Hmm, sorry I don't have plan on doing this yet. But it would be a great feature to add to Karateclub
Dear @bdeng3,
The newest 1.3.0. release on PyPI can handle inference! Can you star the repo and hit star?
Benedek