Open MarleneJacobsen1 opened 1 year ago
For documentation, this issue has been resolved. Posting below correspondence sent via email. The file scoring.py has been updated accordingly.
"Dear Marlene,
Thank you for using emfdscore and our apologies for the slightly confusing output. To answer your question about the relevance of the "moral" column when calling the single vice virtue function, there was a small bug in our code which was instantiating the "moral" column name when 0 moral words were detected, thus giving values of 0s and NaNs. The "moral" column has no significant meaning here. This column is instead usually reflected within the MFD and MFD2 outputs (not for eMFD).
For the sake of efficiency (and to save your time from reconstructing the emfdscore environment from scratch), our recommendation for your analyses would be to simply ignore this column and utilize the other variables as normal. Nonetheless, we have updated the code in the master repository and this error should not happen for future installations of emfdscore. Thank you for drawing our attention to this and good luck for your research! "
Hi,
I just generated the vice-virtue scores for single probabilities. Everything looks fine, the only thing I am confused about is that there is an empty column in my output file, right behind the moral_nonmoral_ration. It's just called "moral" and contains no information (apart from one article that generated no output, it says 0.0 in every column, including the "moral" one. I suppose that's because the article was too long, it's got over 6000 words). What strikes me is that the example output in the tutorial does not have the "moral" column at all.
Any ideas why it's there, what it's supposed to indicate and if I should do something about it?
I'd be glad about your advice @fhopp @jacobtfisher @rene-weber @Sungbinyouk
Thanks so much and have a nice day,
Marlene