Sofa Search
Thousands of sofas out there. Find your match in 20.
SofaSearch is a web application that helps users look for their dream sofa
Try me on Heroku! - http://sofa-search.herokuapp.com
Hack The North 2017 submission by:
- Lawrence Pang
- Nicholas Vadivelu
- Victor Yu
Tech Stack
- Python + Flask backend
- Machine learning tools Keras + Tensorflow
- Web page designed with HTML, CSS (with CSS transitions), & jQuery
Dependencies
- Flask==0.12.2
- requests==2.18.4
- numpy==1.13.1
- pillow==2.7.0
- keras==2.0.8
- tensorflow==1.2.1
Design Details
- A scraper retrieves sofa data from an Ikea website: including pictures, price, and link
- A sofa is presented to a user
- User can rate a sofa with a star number scale
- User has the option to purchase - taking them to the site so they can purchase
- Save a list of sofas the user is interested in through a ‘save’ button
- This process goes on until the user discovers their one true sofa
Local Deployment
- Install Python 3.6.2 (https://www.python.org/downloads/)
- Install virtualenv with
pip install virtualenv - Navigate to your project root and create a virtualenv with
virtualenv my_project - Activate your virtual environment with
source my_project/bin/activate - Install dependencies in the environment with
pip install -r requirements.txt - Start the application with
heroku local web - View the application on localhost:5000
- Stop running the application: Ctrl-C
- Exit virtual environment:
deactivate