It would be good to have some indication of progress from hcluster().
Yeah, well, that might be a bit hard. The first step is to sort the distance matrix which is done using the system sort commands. I guess we could probably tell you what it’s doing, but I don’t think we’ll be able to give updates.
anything would help. I’ve been running over 2 days now. I know that the process is using resources, but I don’t know if it’s actually getting anything done.
Point taken - it probably won’t be in the next official release, but look for it in the next go around. Also, feel free to email us in a week or two for the update.
Here’s another thought.
the hcluster() job I’m running is pushing up against the limits on my RAM (I only have 8GB!). In the past, I think cluster() has failed due to memory limits (I saw this behavior with cluster(), not hcluster()), but appeared to keep running in a loop. I wonder if there’s a memory handling bug. Perhaps some more robust trap for memory failures would be useful?
That’s just a guess.