Error in Pre-fitted sklearn-based taxonomy classifier

Hi Everyone! I am pretty new in Qiime2 and I have managed to work through my very first 16S and ITS samples all by myself… until now… I achieved to generate the outputs from the denoising through qiime dada2 denoise-paired method. Now I want to use the feature-classifier using the pre-fitted sklearn-based taxonomy classifier with a classifier file downloaded from the Qiime2 data resources. I downloaded the following file and saved it in the folder I am currently working in: Naive Bayes classifiers trained on:

qiime feature-classifier classify-sklearn --i-reads PE-rep-seqs.qza --i- classifier silva-132-99-515-806-nb-classifier.qza --o-classification taxonomic-classification.qza

And everytime the error message was:

Plugin error from feature-classifier:
Unable to allocate array with shape (796852224,) and data type float64
Debug info has been saved to …

I don’t know what else to try or what errors I have in what I am doing.
Hope anyone could help and thank you!!!
Cheers!

1 Like

Hi, Melisa! :wave:

Congrats on working through your first analysis!

This is an out of memory error:

Out of memory ( OOM ) is an often undesired state of computer operation where no additional memory can be allocated for use by programs or the operating system. Such a system will be unable to load any additional programs, and since many programs may load additional data into memory during execution, these will cease to function correctly. This usually occurs because all available memory, including disk swap space, has been allocated.

If you search the forum for "unable to allocate array" you will find some solutions to this problem, such as this post for example:

Good luck!
Andrew

1 Like

Thanks Andrew! I will try with the solution in the post you mention. Anyway I tried the commands in another computer and the error was different. It said that the version of the classifier I was trying to use was different than the version of the plugin used to train the classifier. In any case I think I will have to try to train my own classifier and I am completely lost there!

This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.