Open blakerosenthal opened 3 years ago
If its an easy switch, I'd say go ahead and do it if we have the budget. Would be nice to only have one trip.csv file. My post processing function would be easy to modify for this.
I'm not too concerned about the file size since the individual are already large and can't be completely shown using excel.
I did some light data science work to examine how large the table would be. Based on the fake table I created, it seems like "trips.csv" would come in at around ~3.5gb. It wouldn't load at all in excel; not even partially lol. Pandas would basically required to do any kind of post processing, but I think that's fine.
Okay, great. Wanted to make sure I wouldn't break anything if I made that change. I have a hard time scanning the outputs folder sometimes with all those trip files, so combining them seemed convenient. I'll wrap that into my next round of fixes.
Currently we output a separate trip file indexed
pzone, azone, trips
for each segment intrips.yaml
, which is kind of cluttery. Would it be better to have atrips.csv
file with a header likepzone, azone, segment1, segment2, ...
?Pros:
Cons:
@joshua-reynolds?