Taxonomy classifier memory error zsh: killed

Technical Support
#1

Hi guys,

I’m having an issue I’m trying to solve for the last 3 days, unsuccessfully (qiime2, v2018.11).
I’m running:

qiime feature-classifier classify-sklearn --i-classifier silva-132-99-515-806-nb-classifier.qza --i-reads combined_rep_seqs.qza --o-classification taxonomy.qza

and after 10 minutes, I’m getting this error:

> zsh: killed qiime feature-classifier classify-sklearn --i-classifier --i-reads

My machine has 35 GB memory and 4 GB of Swap, and it seems that the process drains the memory, and then just killed. Any thoughts? it’s only occurred once, and only with this command.
I also tried to use smaller rep-seqs.qza file (400 KB instead of 1900 KB), but still getting the same error.

Thank you,
Barak

#2

Yeah, you don’t have enough memory to complete this command successfully. If you are unable to move to a machine with more memory, you can try setting --p-pre-dispatch to 1*n_jobs, but that might not help. The problem here is the Silva database - its huge.

1 Like
#4

Hi @thermokarst,

I changed the n-reads-per-batch to 10000 and it worked with the SALIVA!

Thanks,
Barak