What it does

How we built it

The backend is built in python with flask. It queries the GE SmartCity API for real-time pedestrian data and passes that on to the iOS app. For requests in the future, it reads pre-generated data from a file. The data was generated with a version of Andrej Karpathy's LSTM ported to tensorflow and trained on a little more than a week of pedestrian traffic.

Challenges we ran into

The data returned by the SmartCity API isn't very consistent as it's still in the works. Also, despite being CS grad students, neither of us actually knows how to code.

Accomplishments that we're proud of

What we learned

What's next for Pedestrian Predictor

Built With

Share this project: