Hi guys,
I have a project with 16s RNA microbiom data. While I was doing a taxonomy with Silva 132 99% OTUs from 515F/806R region of sequences, I got this error 
Plugin error from feature-classifier:
Unable to allocate 6.19 GiB for an array with shape (830193664,) and data type float64
I have searched the solution on this forum but I seem that anyone got this before 
Debug info has been saved to /tmp/qiime2-q2cli-err-51nqz5j_.log
Hi @diep.vy,
Welcome to the 
forum!
More information a bout your system and the command ran are really important, as is the full debug log. We try to be pretty good here, but we don’t necessarily know everyone’s specifics unless you tell us!
That said, it looks like a memory error; your data is too big to train. Try decreasing the number of reads per batch (--p-reads-per-batch). It will probably increase. your runtime but it may be more memory efficient.
Best,
Justine
Thank you for the reply,
I used Mac with 16GB ram, I have tried to decreasing the number of the reads per batch but it not work
Are there any solution for this problem except more memory ?
This is my file to do a taxonomy, that not a big data right ?!
rep-seqs-2.qza (28.2 KB)
Hi @diep.vy,
Well, one thought is that you’ll probably get better performance if you use a native installation on your mac.
A second is that. it doesn’t look like you ran quality filtering prior to running deblur, which might explain a large number of sequences.
Best,
Justine
