Kuan Butts
Kuan Butts
Can't seem to draw blobs in the tutorial. Getting the following error: ``` Traceback (most recent call last): File "", line 1, in File "/usr/local/lib/python2.7/site-packages/SimpleCV-1.3-py2.7.egg/SimpleCV/Features/Features.py", line 123, in show self.draw(color,...
There's a bug in earlier version of numpy. See discussion stemming from this ticket: https://github.com/kuanb/peartree/issues/172
As pointed out by @rxl204, it would be handy to limit the number of transfers that can be made when identifying possible routes (e.g. no more than 2 transfers). Similarly,...
Right now peartree is crude in that it assumes/averages the boarding cost for a specific line for both directions that it serves. This is not ideal, as it poorly represents...
Logged during tests: ``` direction_subset = stop_times_by_stop[constraint_2] /home/travis/build/kuanb/peartree/peartree/parallel.py:127: UserWarning: Boolean Series key will be reindexed to match DataFrame index. ``` Is there a better way to do this in pandas?...
An improved transit JSON structure would support key additional attributes needed including: - dwell times by stop - speeds broken out by segments No need to reinvent the wheel here....
Attribute of networkx node should include the stop id (or ids) that it represents. The attribute should be a list type.
In posts such as this one (http://kuanbutts.com/2018/04/01/spectral-cluster-transit/) and elsewhere, I have suggested that coalescing networks can improve speed. This is true, but accuracy is sacrificed on graphs like a transit...
Edge time generation is where most of runtime is spent. The below image demonstrates the breakdown of time by each route in the AC Transit GTFS feed:  Total runtime...
This step is often performed by users, but can be confusing. Create a "most common" path for importing in walk networks (perhaps from OSMnx) and support its conversion to peartree.