Based on the topic classifier, which one of the following is the correct order of training or prediction?

Question options:

String data $>$ CountVectorizer$().$transform$($String data$)>$ use trained model for prediction

String data with target $>$ select feature $>$ train model

String data $>$ CountVectorizer$().$fit$\_$transform$($String data$)>$ use trained model for prediction

String data $>$ select feature $>$ use trained model for prediction