scikit-learn version error

I try to use silva classifier as provided in the “Taxonomy classifiers for use with q2-feature-classifier” tutorial. when i run the command:
qiime feature-classifier classify-sklearn --i-classifier silva-119-99-515-806-nb-classifier.qza --i-reads rep-seqs.qza --o-classification taxonomy-silva.qza

I get this error:
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.

so i try to downgrade the scikit-learn by this command:
conda install --override-channels -c defaults scikit-learn=0.19.1

But this command removed the

And when when I try use qiime feature-classifier I get:
Error: No such command "feature-classifier".

So, my question is: How can i install the scikit-learn 0.19.1 without remove the “feature-classifier” option.
Or, there is an another way to slassify my data?
Thank you

You can or retrain your classifier with Qiime2-2019.1 (preferable) or install older version of Qiime (Qiime2-2018.11)

1 Like

or download the latest taxonomy classifiers! It looks like you shared a very old link (version 2018.2). Download the latest to be compatible with the latest release of QIIME 2.