argo-marketplace / RadRoads

Using OSMnx to find Rad Streets in any city! - https://medium.com/a-r-g-o/rad-roads-using-osmnx-68fc8b15f046
MIT License
11 stars 4 forks source link

Consolidating RadRoads #14

Closed vr00n closed 6 years ago

vr00n commented 6 years ago

Hi everyone - I'm looking for a summary from someone on how an outside user can implement RadRoads?

I see many notebooks here but have no clear idea on how to implement RadRoads on some city.

My recommendation is to consolidate the work into a single python file and provide a requirements.txt so that any user can git clone this repo and input a city name and get RadRoad outputs:

Example:

git clone RadRoads repo
cd RadRoads
python RadRoads.py Tempe, AZ
RadRoads Output for Tempe, AZ
Top 5 longest roads
Top 5 curviest Roads
Top 5 Straightest Roads

A stretch goal could be to provide a single python notebook to illustrate plotting techniques using RadRoads outputs.

Thanks

How to submit an issue?

vr00n commented 6 years ago

Also im not able to find whether sinuosity is being used to determine curviest / straightest roads. Can someone tell me what the plan is here?

jazzzchan commented 6 years ago

Answering the sinuosity, I think our methodology for curviness now actually is the sinuosity methodology. We imported the Vincenty and devided that by the straight distance which is exactly the function of sinuosity. Are you looking for the "sinuosity" as keyword or..?

jazzzchan commented 6 years ago

to expand more, vincenty is the geographical distance , so by dividing the straight distance, we conduct the curviness

jazzzchan commented 6 years ago

for circuity, it is the difference between two nodes and divided by the straight distance, which i don't think it's what we are doing? Please correct me if my understanding was wrong.

vr00n commented 6 years ago

Thanks @jazzzchan can you please create a new issue for discussing approaches to Straight/Curvy roads? ...and how do you feel about this issue (consolidating to a single script). Can it be done? Thanks!

ywnch commented 6 years ago

I think I had a paragraph discussing the concept of sinuousity and how it is adopted in terms of roads (instead of rivers). Please let me know if you think we should elaborate more on that part.

I think we will be working on wrapping up the current stage of work throughout next week. I think there are still some calibrations to do.

I would say, in future work, the definition of curviness may have more than one indicators as different curviness measures can mean different things, as discussed in the article.

vr00n commented 6 years ago

@ywnch please refer #15 for curviness discussion.

Once you consolidate functions_basic.ipynb to a .py file - we can close this issue

THANK YOU!