Lucia was trying to generate some tables for a big single-mouse experiment.
The bug appears to be related to the longterm_id column getting assigned NaNs and dropping those columns.
The following code appears to correct the table translation (trivially forcing all longterm_ids to be present -- only possible in single-mouse experiments):
from jabs_utils.project_utils import *
tmp = BoutTable.from_file('Results_2024-07-29_bouts.csv')
tmp.data.longterm_idx = -1
bin_table = tmp.to_summary_table(5)
bin_table.to_file('Results_2024-07-29_summaries.csv')