I trained the classifier using greengenes2 database with used primer (341F, 785R for V3-V4 region). It only took 1-2 hours (felt like super fast). Many people used to mention that this process took several hours, but not in my case. While it's advantageous for me to complete it quickly, I'm wondering whether my classifier works well or not.
Therefore, I tested the classifier following the Qiime2 tutorial (Training feature classifiers with q2-feature-classifier) using the representative sequences from the Moving Pictures tutorial. And then I compared my taxonomy.qzv file with the one in the tutorial. However, I'm not sure about how to assess whether my classifier is performing adequately. When I viewed the file using Qiime2, it appeared to be similar, but how can I verify this? Should I be concerned about minor differences, or are they not that important?
Just in case, I attach my code below.
#Extract reference reads
qiime feature-classifier extract-reads
#Train the classifier
qiime feature-classifier fit-classifier-naive-bayes\
--i-reference-reads Greengenes2_ref_seqs.qza \
#Test the classifier
qiime feature-classifier classify-sklearn
qiime metadata tabulate
Thankyou for helping me. Have a nice day!