Penn State Dining Predictions

This project is deployed on http://dining-predictions.herokuapp.com/

How to get started with this project

Pre-requisites - pip, npm, Heroku.

Setting up Flask.

1. Clone this repo
2. Download PyCharm as an IDE - you can get an Ultimate version for students.
3. Open this project in PyCharm.
4. Run: pip install -r requirements.txt on terminal.
5. Hit run on PyCharm - Your backend server should be ready to use on localhost:5000!
6. Using the built templates, you can see the client side on the 5000 port as well!

Setting up Vue

For dev mode:

1. Navigate to the frontend directory, Execute npm install on a Terminal.
2. Execute, npm run serve. - Your client side server is ready to use on: localhost:8080 (copy & paste into browser) and you should see the website running in dev mode with the form page as the main page.

For production mode:

1. run npm run build in the frontend directory and copy/replace all the files and folders in the frontend directory to the root directory of the project. This will be your new build - ready to be deployed with your flask server!

Heroku Deployment

1. Procfile, requirements.txt, Pipfile are required for heroku production. Please read more about how to deploy to Heroku on https://github.com/gtalarico/flask-vuejs-template
2. Please set up your account on https://signup.heroku.com/
3. Add and Commit all your changes to git and then execute git push heroku for Heroku deployment.

Machine Learning Model

1. 1.model is the model file we've generated and are using to get predictions.
2. 1.model is a result of an XGBoost model that was trained on the data found in supervised_learning directory of this project. To learn more about XGBoost - visit https://xgboost.readthedocs.io/en/latest/. We've used the sklearn API for XGBoost in this project.
3. The model is built in Test.ipynb in the same directory.
4. Descriptions of data files used:

• processed_buffet.csv: Data from the buffet area of Pollock Commons
• mix.csv: Data from The Mix at Pollock Commons
• edge.csv: Data from the Edge at Pollock Commons
• merged_data.csv: processed data for the model to use.

For any more information or access to this repo, please raise an issue on GitHub and I will respond promptly.