Rate My Plate

Identify the environmental impact of food.

Developed for the CSSBristol Boeing Hackathon.

Overview

Tweet an image of food to our account, and we recognise the food and tweets back the environmental impact of that food.

Icons designed by Madebyoliver from Flaticon

Development Setup

Dependancies

• Install Python version: 3.5 (required for tensorflow)
• Install Pip
• Make sure pip is updated pip install --upgrade pip

Setup

• Clone repo git clone https://github.com/harrymt/boeing-hackathon.git
• Navigate to the directory cd boeing-hackathon
• Run pip install -r requirements.txt to install all Python dependancies
• Setup the Database python db_setup.py
• Run the web server locally by running python application.py

API Credentials

• Create your own file 'credentials.py' with a consumer_key, consumer_secret, access_token, and access_token_secret, which you can generate via the Twitter Application Management (https://apps.twitter.com)

# credentials.py

consumer_key = ''
consumer_secret = ''
access_token = ''
access_token_secret = ''

Server Side Extra Setup

Clone the Repo

• git clone https://github.com/harrymt/boeing-hackathon.git
• Then git pull whenever there are changes

Setup a Cron Job

• crontab -e to show the list of cron jobs.
• Then add * * * * * python twitterbot.py to the cronfile.
• Then add crontab -l to check and see if it worked

Setup a GitHook

• Create a pull.php file with 1 line, <?php exec('cd boeing-hackathon && git pull'); ?>

Technologies used

• Python
• Flask
• Tweepy
• TensorFlow
• Bit.ly
• Leaflet
• Word Net
• Mat Plot Lib
• Hosted on AWS