Important
Machine Learning/AI is not utilized in this project! It's nothing but simple math to parse collected data.
A data analysis and case study created by Kuroji Fusky that collects and parses drawing data from a furry artist and illustrator Paco Panda.
The main gist of the project is that it uses a web scraper to get new drawing submission from his socials; then cleans up the data and adds it to the database. And using that to build this website and the underlying APIs on it!
It uses a combination of BeautifulSoup4 and Selenium WebDriver to scrap drawing data from Paco's socials and adds it to a Redis database that pushes the image binary and the JSON data along with it.
Once we got the scraped data, it uses FastAPI to provide a REST API and does the data science stuff on the fly.
- Node.js 20 or higher (LTS recommended)
- Python 3.11 or higher
- pnpm
- Docker
WIP
cd client
pnpm install
cd server
python -m venv venv
# Linux
source venv/Scripts/activate
# Windows
.\venv\Scripts\activate
pip install -r requirements.txtTBA
The project began on October 31st 2021. The possible inspiration from this project is through McBroken (basically a McDonald's broken ice cream machine site) and it'd be interesting to see in all of his drawings to see said data, and its various datasets.
This project collects the following:
- The title and date of the piece
- Number of character(s) appearances, including species and names
- Media type (either drawn digital or traditional)
- Programs/mediums used (i.e. Photoshop, Procreate, etc.)
- The source where I got the data (i.e. FurAffinity, DeviantArt, InkBunny, Weasyl, etc.)
