RateYourStyle was an online fashion community to help users discover their personal style. As of 1/31/2025, RYS is no longer active; this repo serves as an archive.
Key features of RYS included:
- User uploaded outfit pics
- AI assisted outfit descriptions
- Aggregating user clothes into a virtual closet
- Aggregating data about user clothes to give statstics like cost per wear, most worn items
- Searchable discover page for users to discover each others outfits and closets
- Ability to leave reviews and comments on each others outfits
- Ability to request style feedback from specific users
- Sign up via Oauth with Gmail
- Sign in via OTP or username/password combo
After enrolling in Georgia Tech's OMSCS program, RYS was also used to complete a class project for CS-6460: Educational Technology (final paper here). Two features were added as part of that project:
- Educational modules about color science
- Personal color anlaysis web app
RYS was designed with microservice architecture in mind. The main components include:
- server: handles outfit uploading, getting user outfits, posting reviews
- server-auth: handles user authentication and account sign up
- server-search (not included): hanldes search on discover page
- server-image: handles uploading images to GCP, and getting outfit descriptions from OpenAI
- ui: creates the frontend for YoutubeBacklinks
To build and run RYS locally, you need the following:
- Docker installed
- A GCP account and a GCP project created. See https://console.cloud.google.com
- OpenAI dev account
- Gmail SMTP
- Under the secrets/ subdirectory, create the following files:
- localdev_sa.json # service account key to GCP with GCS access
- openai.key
- Use
docker-compose up --buildto bring up the entire stack