Corpus compact #112
Corpus compact
#112
|
What does the st.CorpusFromPandas().compact(st.AssociationCompactor(2000)) do exactly? and how does it function? |
Answered by
JasonKessler
May 19, 2022
Replies: 1 comment
|
It accumulates the highest scoring words (most associated) for each category, as selected by a set frequency rank in all classes, until no more than 2000 (or whatever the specified number is) words are collected. By default, Scaled F-Score is used, but any |
0 replies
Answer selected by
JasonKessler
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
It accumulates the highest scoring words (most associated) for each category, as selected by a set frequency rank in all classes, until no more than 2000 (or whatever the specified number is) words are collected.
By default, Scaled F-Score is used, but any
TermScorersuch asRankDifference,DeltaJSDivergence, etc. could be used,