This is the code and the files used to create the Montreal crime tracking app by CBC/Radio-Canada.
The app does the following:
- Fetches crime data from the City of Montreal's open data portal (updated every three months).
- Aggregates crimes into a hexagonal grid of the Island of Montreal and uploads the resulting GeoJSON to Mapbox.
- Calculates montly numbers and time-of-day frequency for each type of crime and exports them as JSON files to be used with Chart.js
- Calculates significant changes in the last three months to the same period one year before to signal large increases or decreases in crime (this will be added to the page in a future update).
- Exports additional supporting data (averages, min/max, etc.) in a JSON file used by the app.
The source data files are in the src folder. The resulting data is in static
The app is written in Python 3.6 and uses the pandas and geopandas libraries for most of the analysis.
This app assumes the following:
- You have a Mapbox account and an API key, which is stored in a dict in a file called
secrets.py - You have created a tileset and have its ID stored in
secrets.py
First, create a virtual environment to keep everything tidy. I use Anaconda.
conda create -n crimeapp
Enter the environment:
source activate crimeapp
Install the dependencies. Geopandas is a tricky one to install depending on your system. I suggest installing it separately with conda.
conda install -c conda-forge geopandas
conda install --yes --file requirements.txt
If the second installation commaand fails, try with pip:
pip install -r requirements.txt
Then run the app:
python app.py