-
Notifications
You must be signed in to change notification settings - Fork 254
improve memory performance of transitfeed.py #30
Comments
From [email protected] on May 29, 2008 02:24:02 I wasn't able to process my entire feed (52000 trips, 810000 stop times) in 360MB of |
From [email protected] on July 29, 2008 16:41:17 By the way, I have a fix under review ( http://codereview.appspot.com/2644 ) in the |
From [email protected] on October 07, 2008 16:06:05 I have released 1.1.6 which uses sqlite for the stop_times table. Slow, but works. Status: Fixed |
# This is the 1st commit message: add muni ac gtfs # This is the commit message google#2: only output routes # This is the commit message google#3: add initial version of kml route styler # This is the commit message google#4: add sample csv route styles # This is the commit message google#5: add route styling functionality to kmlwriter based on csv provided in /route_styles # This is the commit message google#6: add generated kml with styling and selection done based on csv # This is the commit message google#7: replace K and T with KT muni line in csv input # This is the commit message google#8: change zero routes assertion to warning due to naive automation scripts # This is the commit message google#9: strip whitespace on color in csv # This is the commit message google#10: add script for generating all maps # This is the commit message google#11: add extracted gtfs and most output # This is the commit message google#12: update excels # This is the commit message google#13: update automation script to enable starting midpoint in the list # This is the commit message google#14: update requirements file # This is the commit message google#15: set default skip ahead to none # This is the commit message google#16: add more kml outputs # This is the commit message google#17: add parallel processing # This is the commit message google#18: add muni bart current 15 kml # This is the commit message google#19: add muni current 15 geojson # This is the commit message google#20: add conversion to geojson # This is the commit message google#21: change pipe to and # This is the commit message google#22: add json outputs because google maps kml layer is garbage # This is the commit message google#23: remove once daily tripper for samtrans 120 route # This is the commit message google#24: convert samtrans current 15 kml to geojson # This is the commit message google#25: add regional new plan # This is the commit message google#26: updated kml manually # This is the commit message google#27: add json of new projects proposed # This is the commit message google#28: add Hidden_Shapes param to kmlwriter and route_styler # This is the commit message google#29: add Hidden_Shapes column to route style csvs # This is the commit message google#30: dict typo # This is the commit message google#31: support for none hidden_shape # This is the commit message google#32: nan support... # This is the commit message google#33: nan stuff # This is the commit message google#34: add hidden shapes to vallejo sfbay ferryroute # This is the commit message google#35: fix bug # This is the commit message google#36: set same stroke width for everyone # This is the commit message google#37: sfbay ferry modified # This is the commit message google#38: reverse boolean bug # This is the commit message google#39: Reload KML/Json output # This is the commit message google#40: remake kml json output
From [email protected] on June 28, 2007 03:23:39
processing feeds takes tons of memory and it seems to get worse with each
release. Internally we have some large GTFS files that won't load on a 2GB
machine.
We need to do some profiling to work out if we are doing something that
prevents python doing the right thing.
It might be worth trying to reload a medium sized feed several times. If we
delete the Schedule object each time then the python process should stop
growing.
Here's someone with debugging a possibly similar problem http://mail.python.org/pipermail/python-list/2007-June/445838.html but he didn't get a reply.
Original issue: http://code.google.com/p/googletransitdatafeed/issues/detail?id=30
The text was updated successfully, but these errors were encountered: