Closed RETprojects closed 1 year ago
I made some changes to the preprocess function in predict.py so that the clustering only depends on the words from the text that are classified as verbs by NLTK.
How does this change impact our f-scores? If it doesn't improve them, we should probably keep it on a separate branch.
I made some changes to the preprocess function in predict.py so that the clustering only depends on the words from the text that are classified as verbs by NLTK.