Skip to content

cli.md

Latest commit

 

History

History
78 lines (55 loc) · 3.86 KB

cli.md

File metadata and controls

78 lines (55 loc) · 3.86 KB

HOME

command line interface

If you don't want to always write new scripts to build the graph or to find routes, you can simply use the command line interface.

building a graph from the terminal

this assumes that your data has been preprocessed into the correct format

from your terminal run:

multimodalrouter-build hubType1 transportMode1 pathToDataset1 --maxDist 123 --extraMetrics metric1 metric2 --drivingEnabled --Dir pathToSomeDir

args

  • hub type,transport mode,dataset path = the first arguments in your cli should be these three. They must follow this order to be parsed correctly
    • hub type = the name the hubs from this dataset will get
    • transport mode = the mode of transport unique to this hub
    • dataset path = ideally the absolute path to the dataset that contains this hub type
    • NOTE: you can add as many hub types and datasets at once as you want just make sure the order is always; type1 mode1 path1 type2 ...
  • --maxDist = the float value that limits the maximum length a driving edge can have (this is irrelevant id --enableDriving isn't set)
  • extraMetrics = a list of keys that the graph will scan your data for and add the values to any edge where the key exists in the data (NOTE: each key should be present in at least one dataset, the number of extra metrics is not limited)
  • --drivingEnabled = if this flag is added the graph will try to build connections between hubs for any hub that is closer than the --maxDist to anoter hub (check here to see if your data allows this)
  • --Dir = put the target directory of the save file here. (if this is not set the graph will be saved to a default dir)

example

create a new dir for the garph and edit the --Dir if you want the graph in a easy to access directory otherwise keep as is

NOTE: this example assumes you are still in the project dir. If not please adapt the dataset path accordingly

multimodalrouter-build airport plane ./docs/examples/flightRouter/data/fullDataset.csv --maxDist 100 --drivingEnabled 
--extraMetrics source_name destination_name

output:

Building graph...
Generating airport Hubs: 100hub [00:00, 186995.27hub/s]
Graph built and saved.

finding a route

multimodalrouter-route --start lat1 lng1 --end lat2 lng2 --allowedModes car plane --maxSegments 123 --verbose

NOTE: if you have higher dimensional coordinates you can simply add them here aswell

agrs

  • --start = the start coordinates for the route (doesn't have to be exact to the Hub it will find the closest hub to the point. Also the coordinates do not have to be in dgrees, but rather fit with the coordinate system of your data. Read more here)
  • --end = the end coordinates of the route: (same features as for --start apply)
  • --allowedModes = a filter that discards edges that use a transport mode that is not in this list. (default= ['car'])
  • --maxSegments = an interger value of the maximum number of hubs a route can have. (e.g.: to avoid very deep searches when long routes are not logically feasable) (default = 10)
  • --verbose = if this flag is set the output route will contain the edgeMetadata for every leg of the route (default=false when flag not set)

example

NOTE: this example assumes you build the graph from the previous example. if you changed the save path this script wont find it (this will hopefully be fixed in the future).

multimodalrouter-route --start 60.866699 -162.272996 --end 60.872747 -162.5247 --allowedModes plane car --maxSegments 10 --verbose

output:

Start: ATT
        Edge: (transportMode=plane, metrics={'distance': 13.641155702721523, 'source_name': 'Atmautluak Airport', 'destination_name': 'Kasigluk Airport'})
-> KUK