/gabbar

Guarding OSM from invalid or suspicious edits!

Primary LanguageJupyter NotebookMIT LicenseMIT

gabbar

image

EXPERIMENTAL: UNDER DEVELOPMENT

Guarding OSM from invalid or suspicious edits, Gabbar is an alpha package of a pre-trained binary problematic/not problematic classifier that was trained on manually labelled changesets from OpenStreetMap.

image

https://en.wikipedia.org/wiki/Gabbar_Singh_(character)

Install

pip install gabbar

Setup

# Setup a virtual environment with Python 3.
mkvirtualenv --python=$(which python3) gabbar_py3

# Install in locally editable (``-e``) mode.
pip install -e .[test]

# Install node dependencies.
npm install

Get a prediction

# Get a prediction for a changeset.
$ python gabbar/scripts/cli.py 47734592
{"prediction": "good", "timestamp": "2017-04-26 01:05:00.441977", "version": "0.2.4"}

Run tests

# Run tests.
npm run test

Performance

Performance of the model on both labelled and unlabelled changesets is tracked in metrics.csv

Hyperlinks