TaxonomicClassifier

I’m running into error “Missing one or more files for TaxonomicClassiferTemporaryPickleDirFmt: ‘sklearn_version.json’”

I ran
conda install --override-channels -c defaults scikit-learn=0.22.1
qiime tools import --type TaxonomicClassifier --input-path gg_13_8_otus --output-path gg_13_8_otus.qza

the classifier downloaded is greengenes 13_8.

Thank you,
Cam

Hi @Chozinentropy - we are missing some information necessary in order to help you:

  • what version of QIIME 2 are you running

Also, I have some questions:

This step is not necessary, and has now mutated your conda env. Before we can help you, please reinstall QIIME 2 in a clean env (which uses the package versions that we have validated and verified can work together).

You are attempting to import many different formats (and types) of data into an incorrect semantic type. TaxonomicClassifiers are created, not imported:

https://docs.qiime2.org/2020.2/tutorials/feature-classifier/

Please take some time to review the QIIME 2 User Docs at https://docs.qiime2.org.

1 Like

ok, I’ve reinstalled the env ran everything to Taxonomic Analysis. Using the downloaded gg-13-8-99-515-806 I getting error
Plugin error from feature-classifier:

The scikit-learn version (0.20.2) used to generate this artifact does not match the current version of scikit-learn installed (0.22.1). Please retrain your classifier for your current deployment to prevent data-corruption errors.

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

I’m not sure how to handle this without changing sklearn install.

You need to download the version of the classifier that matches the version of QIIME 2 you are using. You haven't yet answered my question about what version of QIIME 2 you are using, but, I will assume 2020.2, since that version uses sklean 0.22.1. In that case, download the version listed here:

https://docs.qiime2.org/2020.2/data-resources/#taxonomy-classifiers-for-use-with-q2-feature-classifier

Again, I strongly encourage you to spend some time reviewing the QIIME 2 User Docs at https://docs.qiime2.org

:qiime2:

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