Skip to content

Latest commit

 

History

History
69 lines (53 loc) · 3.08 KB

README.md

File metadata and controls

69 lines (53 loc) · 3.08 KB
Raw

Wizard's Apprentice (1.0.0)

The wizard needs your help - he needs to fight the evil dragon, but he forgot what the tool for his latest spell looks like. Take pictures of what his tool looks like and the wizard will award you with points - compete with your friends for top points.

Authors

Joyce Liao, Hannah Sindorf, Scott Currie

Overview

Users are given a randomized prompt consisting of a few keywords (e.g. "blue sports car and a person"). When they submit a photo that they believe fills those keywords, it is evaluated by google vision API and the returned tags are matched against the prompt. When the prompt is fulfilled, points are awarded and a new prompt is generated.

Users have a unique login where their own points are tracked.

Stretch goals

  • video game-styled UI with characters.
  • make it a social game. Users share the same prompt and only the winning user is awarded points. Photos are posted into a public feed.
  • look at photo headers to determine that the photo was taken the same day (to avoid cheating)
  • add a report feature in case it's obvious the photo is just a stock photo

Getting Started

Navigate to https://rebrand.ly/wizards_apprentice.

To run locally:

  1. Enter a new virtual environment: pipenv shell --python 3.6
  2. Install dependencies: pipenv install
  3. Create .env file in the root of the project.
  4. Create a postgresql database for the app
  5. Run Flask db init, migration, upgrade
  6. Run the app: flask run
  7. Navigate to http://localhost:5000

Architecture

Requires Python >= 3.6, Flask, PostgreSQL, Pillow

Frontend

Server-side rendered HTML templates using Jinja2.

Backend

Flask, PostgreSQL. Deployed on EC2 (Ubuntu 18.04) running Nginx and Gunicorn

Credits

Sample code/tutorials used

Assets Used

Change Log

12-27-2018 09:00:00 Project Started
12-27-2018 10:30:00 Flask app set up
12-27-2018 11:30:00 Google Vision API submissions working
12-27-2018 12:00:00 User registration/authorization set up
12-27-2018 12:00:00 Database layer implemented
12-27-2018 15:00:00 App deployed to AWS
12-27-2018 15:30:00 Submission evaluations working
12-27-2018 17:00:00 First tests of front end user experience
12-28-2018 09:00:00 Begin Day 2
12-28-2018 14:00:00 Improved submission evaluation logic
12-28-2018 15:30:00 Player history and all players history complete
12-28-2018 16:00:00 Basic styling complete 01-02-2019 09:00:00 Begin Day 3
01-02-2019 11:00:00 Time conversions from UTC to PDT working
01-02-2019 12:00:00 Color detection improved
01-02-2019 13:30:00 Testing on auth routes complete
01-02-2019 16:00:00 Admin page added