Cycle Safe
Plan a safer bike route in Boston. Project built during the Insight Data Science Fellows Program
Tools
I used Docker and conda to run both the jupyter notebook and the app locally, and deploy to AWS Beanstalk. Docker is definitely overkill, but I wanted to use this project to learn more about it.
To launch the server:
docker-compose up --build flask-dash
Then open your browser at http://localhost:5000/
Data
The data is coming from the city of Boston and from the Massachusetts DOT.
Model
I currently use a logistic regression to classify a street segment into safe or risky.