Where I play with GTFS, some notebooks also published in https://cjer.github.io
To run these notebooks you need Python 3.5+ with all the necessary packages, Jupyter Notebook (and an internet connection for some data fetches).
- Install Anaconda3 - this will give you all the basic scipy libraries (pandas, matplotlib, seaborn...) and Jupyter notebook. I'm using version 4.4, can't vouch for 5+. Get the right version here, make sure you're installing Anaconda3 and not 2
- You might need to update some libraries (
conda update libname [-c conda-forge]
) - Install partridge
- (for notebook 3) Install networkx
- (for notebook 3) Install peartree In the near-ish future, I'll try to publish a docker container and/or a spec-file for creating an environment more easily. A binder might pop-up at anytime too.
After you have everything installed, run a local jupyter notebook (shortcut should be provided with the Anaconda installation), pointing it to the folder you cloned the repo to. And voila - you're on your own.
Note that the data is based on the time of download, so some parameters might need to be changed (e.g. the date for filtering the Partridge GTFS feed).
Please let me know if something's off or missing with anything.