a20r / mod

Machine learning for mobility on demand

mod

Preprocessing data for mobility on demand

Installing dependencies

This assumes you have Python 2.7 and pip

$ ./install.sh

Getting some data

$ scripts/download_data.sh

This will create a data/ directory if one does not already exist. It downloads essential data for creating the experimental data needed to run the ridesharing experiments

To create a small dataset

Running the preprocessing on the entire dataset takes a long time. For debugging and experimentation, I recommend using only a small portion of the file. The command below will make a new file in data/ called trip_data_short.csv. This file contains 10000 lines of trip_data_5.csv

$ head -n 10000 data/trip_data_5.csv > data/trip_data_short.csv

Create data for experiments

$ scripts/create_mod_files.sh [output dir for new files]