Training taxonomy classifier command killed

Technical Support
1

Hey everyone

I trust this message meets you well. I am new qiime2 user. When I try to train my taxonomy classifier, the process stops after 10 minutes with killed written below.
Below are the commands which I used and the error I encountered after adding --verbose to my command.
Kindly advise about this.

(qiime2-2021.8) afrah@afrah-IdeaPad-3-15IML05:~/Desktop/sf_qf$ qiime feature-classifier fit-classifier-naive-bayes \

--i-reference-reads gg_13_5.qza \

--i-reference-taxonomy gg_13_5_taxonomy.qza \

--o-classifier classifier.qza

Killed

(qiime2-2021.8) afrah@afrah-IdeaPad-3-15IML05:~/Desktop/sf_qf$ qiime feature-classifier fit-classifier-naive-bayes \

--i-reference-reads gg_13_5.qza \

--i-reference-taxonomy gg_13_5_taxonomy.qza \

--o-classifier classifier.qza --verbose

/home/afrah/anaconda3/envs/qiime2-2021.8/lib/python3.8/site-packages/q2_feature_classifier/classifier.py:102: UserWarning: The TaxonomicClassifier artifact that results from this method was trained using scikit-learn version 0.24.1. It cannot be used with other versions of scikit-learn. (While the classifier may complete successfully, the results will be unreliable.)

warnings.warn(warning, UserWarning)

Killed

2

Hello Afrah,

When I have seen the 'Killed' message before, it was when I was running Qiime on a compute cluster / supercomputer. I had submitted my Qiime command to a slurm queue, and the supercomputer 'killed' (canceled/removed) my job because it had ran too long or used too much memory.

Are you running Qiime on a computer using a slurm or torque queue? If so, how much time and memory / RAM are you listing on that submission script?

2 Likes
3

Hello Colin,

Thank you so much for your reply. I am not using a supercomputer, my device is Lenovo Ideapad 3 with 8GB RAM and I installed qiime2 via ubuntu in a linex window, which is created by partition and dedicated 100GB for qiime2 operation. The killed message appears after 10 minutes when I run this command. I am running only 8 samples at once.
Also, another thing to be noted is this note which appeared after I added --verbose to my command.
UserWarning: The TaxonomicClassifier artifact that results from this method was trained using scikit-learn version 0.24.1. It cannot be used with other versions of scikit-learn. (While the classifier may complete successfully, the results will be unreliable)
The qiime2 in my device has 0.24.2. the version of scikit-learn, I am wondering how to resolve this issue.

4

Hi @Afrah - thanks for the additional details! You are almost certainly running out of RAM (memory) - do you have access to a larger machine?

2 Likes
5

Hi @thermokarst Thanks for your reply. I have no access to larger machine at the moment, if I change my machine's RAM, will it work?

6

Hi @Afrah - if you are able to upgrade your RAM then it might work, although getting access to another machine will possibly be easier, and likely cheaper.

Another option is to use a different classifier - one that has been pre-trained for you. You can find a few here:

https://docs.qiime2.org/2021.8/data-resources/

2 Likes
7

Hi @thermokarst - Thanks alot for your help.

2 Likes
8

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