billshitg
billshitg
just tried this with TB 2.7 and it worked very well. amazing! thanks for the fix. why has it not been merged already?
good point. We don't have benchmarks. We should probably do it once we have the actual dataloaders.
Nice feature to have, indeed. We don't have bandwidth for this right now unfortunately, but we should be able to include it in 1.1. If folks would like to build...
hmmm, weird error. Could you shut down all services `docker compose down`, and then update the images `docker compose pull` to make sure all images are latest?
@parkererickson-tg any thoughts?
It might be that the cluster is not big enough for the ldbc dataset. We will take a look at it today. @parkererickson-tg any thought on this? Just to confirm,...
ideally the larger the better. I would say at least 16G of memory