You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi! I am trying to use your package to integrate a dataset of RNA + ATAC, it is approximately 100k cells. I keep having issues with jobs getting OOM killed. So far I have tested up to 20 CPUs, 500GB RAM, but no luck, so I was wondering if you have done any testing and could give me an indication of how much memory I would need.
The text was updated successfully, but these errors were encountered:
What is the amount of features you're using there as input? One way to make it work has been using only a subset of highly variable features. You can also try lower precision, i.e. float32, which should be the default now.
Hi! I am trying to use your package to integrate a dataset of RNA + ATAC, it is approximately 100k cells. I keep having issues with jobs getting OOM killed. So far I have tested up to 20 CPUs, 500GB RAM, but no luck, so I was wondering if you have done any testing and could give me an indication of how much memory I would need.
The text was updated successfully, but these errors were encountered: