Open Travis-Barton opened 2 years ago
Side note, i noticed that there's a section inside SectionInfo.process_raw_features that looks for nameless sections
Should that include names like ' '
?
I included a screenshot that comes from the first line of X_test.jsonl
line['section']['sections']
Did you manage to get an interpretable listing?
Hey all,
Right now I'm manually dissecting the code and looking to create a vector of column names for the 2381 columns based on features.py. But is there an easier way to do this? I imagine people have wanted to have interpretable columns before. Is there a script to extract the actual column names for the datasets?