Skip to content

Latest commit

 

History

History
52 lines (40 loc) · 1.03 KB

README.md

File metadata and controls

52 lines (40 loc) · 1.03 KB

Requirements

Classifier

Dependencies
# conda update scikit-learn numpy
# pip install pybrain pandas geocoder

Scraper

Dependencies
# npm install -g bluebird request json2csv fast-csv jsdom

Note: You may need to set NODE_PATH=/usr/local/lib/node_modules

Run scraper. Responses are saved under responses
node fetch_listings.js
Combine responses. This creates a file data/listings.csv
node combine.js
Fetch additional data for a provider (eg. REMAX, SUTTON, C21). Responses are saved under extra_data/$PROVIDER
# touch ${PROVIDER}_errors.txt
# node fetch_details.js $PROVIDER
Firestation data

Run notebooks/firestations.ipynb

Fetch historical data
# ./fetch_hist_DUPROPRIO.sh
# node extract_DUPROPRIO.js
# node combine.js
Geocode historical listings. This creates a file data/scripts/geocoded_listings.csv
# python geocode_listings.py