UPDATE - The application demo is no longer live. See the instructions below on how to download it locally. You may also contact me, and I can quickly get the demo up and running again on a public server.
This project is a submission to the Yelp Dataset Challenge . Using Yelp's public dataset, I trained a neural network that is designed to comment on pictures of food, as humans do on Yelp's site.
The system is was live here - check it out! If you're interested in how it works, be sure to check out the technical report I wrote to accompany this project.
The model is built in keras, and the application is served using the flask microframework.
If you'd like to run a demo locally, clone the the server directory and enter that directory. You'll then need to install the application's software requirements.
- Open a
virtualenv(optional)
# start an environment
virtualenv your-env
# activate it
source your-env/bin/activate- Install the required packages (python3)
python3 -m pip install -r requirements.txt
- Then, run the application
python3 app.pyAnd you're done! go to your localhost, port 5000, and you'll see the application is live.