You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
@Robinlovelace I've pushed a new function journey2 to this fork of the package.
I've done a quick rewrite of the journey function using the code from opentripplanner. It is very hacky, and doesn't return exactly the same results, because I had trouble understanding how you were parsing the json returned by cyclestreets.
But I'm getting a healthy 31 routes/second out of the function.
There is more parsing to do on the outputs, as it has whole routes and segments in the same data frame. But this seems like useful work to me?
The text was updated successfully, but these errors were encountered:
But I'm getting a healthy 31 routes/second out of the function.
Impressive! If you could put in a PR to the cyclestreets package that could be a good way to capture benefits from this work. Surprised the speed-up is so high: seemed we were getting only max ~10 routes/s on the batch routing UI. Parallelising it can speed it up but that's sending queries one-route-at-a-time.
@Robinlovelace I've pushed a new function
journey2
to this fork of the package.I've done a quick rewrite of the
journey
function using the code fromopentripplanner
. It is very hacky, and doesn't return exactly the same results, because I had trouble understanding how you were parsing the json returned by cyclestreets.But I'm getting a healthy 31 routes/second out of the function.
There is more parsing to do on the outputs, as it has whole routes and segments in the same data frame. But this seems like useful work to me?
The text was updated successfully, but these errors were encountered: