It would be great if syntaxgym evaluate had an option for returning by-region surprisals in a tabular format (something easily readable as a data frame in R or pandas). Probably the most natural format would be where there are columns for item and condition, and then one column per region (in which the entries are the region's surprisal values).
Updated: I've added this functionality by adding a --tabular_surprisals option for syntaxgym compute-surprisals, on the tabular-surprisals branch. Not sure I did this in the most elegant way though.
It would be great if
syntaxgym evaluate
had an option for returning by-region surprisals in a tabular format (something easily readable as a data frame inR
orpandas
). Probably the most natural format would be where there are columns for item and condition, and then one column per region (in which the entries are the region's surprisal values).