Is there any sort of stop word removal happening for tensorflow pipeline? If yes where can i see the dictionary of stop words?
Did you take a look at the docs: Component Configuration
intent_featurizer_count_vectors
has configurable stop_words
parameter
yes I saw that it has configurable , but do we need to send the list of stop words from the pipeline?
please see the sklearn’s docs: sklearn.feature_extraction.text.CountVectorizer — scikit-learn 0.21.3 documentation for more details