The scikit-learn version (0.19.1) used to generate this artifact does not match the current version of scikit-learn installed (0.20.2). Please retrain your classifier for your current deployment to prevent data-corruption errors

Now, I tried Greengene 13_8 99% OTUs full-length sequences and used the reads-per-batch parameter and i got this error:

qiime feature-classifier classify-sklearn \

–i-classifier gg-13-8-99-nb-classifier.qza
–i-reads rep-seqs.qza
–o-classification taxonomy.qza
–p-reads-per-batch 0
Plugin error from feature-classifier:

The scikit-learn version (0.19.1) used to generate this artifact does not match the current version of scikit-learn installed (0.20.2). Please retrain your classifier for your current deployment to prevent data-corruption errors.

Debug info has been saved to /tmp/qiime2-q2cli-err-yzq1yhkl.log

I tried to train classifier following this tutorial. then I ran the command again:

qiime feature-classifier classify-sklearn \

–i-classifier gg-13-8-99-nb-classifier.qza
–i-reads rep-seqs.qza
–o-classification taxonomy.qza
–p-reads-per-batch 500


but unfortunately same error keep coming over and over…:

Plugin error from feature-classifier:

The scikit-learn version (0.19.1) used to generate this artifact does not match the current version of scikit-learn installed (0.20.2). Please retrain your classifier for your current deployment to prevent data-corruption errors.

Debug info has been saved to /tmp/qiime2-q2cli-err-d7blqz0o.log

you need to re-download the classifier from here: https://docs.qiime2.org/2019.1/data-resources/#taxonomy-classifiers-for-use-with-q2-feature-classifier

because you are using an older version of the classifier that does not match the latest version of the software.

1 Like

which one do I re-download? I have deleted and re-install three classifiers multiple times which are:

and all three of them is giving the same errors:
either -

Plugin error from feature-classifier:

The scikit-learn version (0.19.1) used to generate this artifact does not match the current version of scikit-learn installed (0.20.2). Please retrain your classifier for your current deployment to prevent data-corruption errors.

Debug info has been saved to /tmp/qiime2-q2cli-err-ctjopi14.log

OR -

MemoryError

Now I tried to download this 13_8 (most recent) and tryto train my classifier using these two files from it (99_otus.fasta and 99_otu_taxonomy.txt) , after importing them into artifact files using my (rep-seqs.qza) but I am stuck at this step it doesn’t move forward:

qiime feature-classifier extract-reads
–i-sequences 99_otus.qza
–p-f-primer GTGCCAGCMGCCGCGGTAA
–p-r-primer GGACTACHVGGGTWTCTAAT
–p-trunc-len 120
–p-min-length 100
–p-max-length 400
–o-reads ref-seqs.qza

Note (maybe helping) : samples were sequenced using 454 GS Junior instrument model…
(18 samples / single end)

yes go that route, train your own classifier. That is the best way to make sure that you are using the correct version during classification.

“doesn’t move forward” sounds to me like this is working as intended! This step should take time, possibly several hours.

I am training my own classifiers. After training, I am assigning taxonomy using -

qiime feature-classifier classify-sklearn
–i-classifier classifier.qza
–i-reads rep_seqs.qza
–o-classification taxonomy.qza

and its running for almost 2 days now. Is it normal time for this step?
or its really stuck?
P.S. I am using silva132_99 database

Yep! Be patient, it’ll be over soon!

1 Like

4 posts were split to a new topic: Issue with incompatible scikit-learn versions installed

A post was split to a new topic: scikit-learn version warning