Skip to content

hayoung1214/Mandoo

ย 
ย 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 

Repository files navigation

QUIZRIX(ํ—ˆ๋ฝ ํ•„์š”)

  • ํ•™์Šต์ง€ ์ด๋ฏธ์ง€ ๋ฐ์ดํ„ฐ์—์„œ ๋ฌธํ•ญ์ •๋ณด๋ฅผ ์ถ”์ถœํ•ด ๋ฐ์ดํ„ฐ๋ฒ ์ด์Šค์— ์ €์žฅ ํ›„ ํ•ด๋‹น ๋‚ด์šฉ์„ ์‚ฌ์šฉ์ž์˜ ๋ธŒ๋ผ์šฐ์ €์—์„œ ์ ‘์† ๊ฐ€๋Šฅํ•˜๋„๋กํ•˜๋Š” ํ”Œ๋žซํผ
  • Quizrix is a platform with a user-friendly interface designed to upload and view workbook problems via classifying different components of the question from a problem image and saving them to a database.
  1. ํ•™์Šต ๋ฌธ์ œ์˜ ์ด๋ฏธ์ง€๋ฅผ ์—…๋กœ๋“œ ๋ฐฉ์‹์œผ๋กœ ๋“ฑ๋กํ•œ๋‹ค.
    The user uploads the image of the problem.

quizrix_imageupload

2. ํ•™์Šต์ง€ ์ด๋ฏธ์ง€ ๋ฐ์ดํ„ฐ์—์„œ ๋”ฅ๋Ÿฌ๋‹์„ ์ด์šฉํ•˜์—ฌ ๋ฌธํ•ญ์ •๋ณด๋ฅผ ์ถ”์ถœํ•œ๋‹ค.
The components of the problem are classified through a deep-learning model. (๋ชจ๋ธ ์™„์„ฑ ํ›„ ์ˆ˜์ •)

quizrix_imageupload


 ์ตœ์ข…๋ณธ gif ์ถ”๊ฐ€

๋ณธ ํ”„๋กœ์ ํŠธ๋Š” ์ฝ”๋“œ๋„› 'QUIZRIX' ์‚ฌ์—…์˜ ์ผ๋ถ€ ํ”„๋กœํ† ํƒ€์ž… ์ œ์ž‘์„ ์œ„ํ•ด ์ง„ํ–‰๋˜์—ˆ์Šต๋‹ˆ๋‹ค.


This project was carried out for building a prototype for "Quizrix" of Codnut.

System Architecture ->๊ฒ€์‚ฌ ๋ฐ›๊ณ  ์ตœ์ข… ์˜ฌ๋ ค๋‘˜๊ฒŒ์š”

Description

์›นํŽ˜์ด์ง€์— ๋ฌธ์ œ ์ด๋ฏธ์ง€๋ฅผ ์ฒจ๋ถ€ํ•˜๋ฉด ๋ฌธํ•ญ / ๋ณด๊ธฐ / ์„ ์ง€๋กœ ์„ธ๋ถ„ํ™”ํ•˜์—ฌ ๋ฐ์ดํ„ฐ๋ฒ ์ด์Šค์— ์ €์žฅํ•ฉ๋‹ˆ๋‹ค.
The problem is classified and separated into three components; question, content, and answer, and then they save into the database.

ํ˜„์žฌ ๊ฐœ๋ฐœ ์ค‘์— ์žˆ์Šต๋‹ˆ๋‹ค.


Tech Stack


๋ถ„๋ฅ˜ ๊ธฐ์ˆ 
๊ฐœ๋ฐœํ™˜๊ฒฝ NHN S3 Docker
Front-end react Javascript Bootstrap axios Styled-components
Back-end Flask Python Gunicorn Postman Swagger
DB MongoDB
Deep learning Pytorch Colab
Etc Nginx github VScode Google Cloud Platform AWS Prometheus Grafana Redis

AI

flow

AI Flow


๊ฒฐ๊ณผ Results

  • Model 1 model1

  • Model 2
    model2

Frontend: React

file tree install


Backend: Flask

file tree install


Installation

Deploy

Github

  • Clone Repository

    git clone https://github.com/2021-Team-E/Mandoo.git

Front-end setting

  • package.json

    {
      ...
    
      "proxy": "http://<ip>:<server_port>",
    
      ...
    }
  • config.js

    export const USER_SERVER = "http://<ip>:<server_port>";

Back-end Setting

  • app.py

    mongo = MongoClient('mongo_db', 27017)
  • detection.py

    #์„œ๋ฒ„ ํ™˜๊ฒฝ์—์„œ์˜ tesseract.exe ๊ฒฝ๋กœ๋กœ ์„ค์ •
    pytesseract.pytesseract.tesseract_cmd="/usr/bin/tesseract"
  • requirements.txt
    Dockerfile์—์„œ ๋”ฐ๋กœ ์„ค์น˜ํ•˜๋Š” ๋ผ์ด๋ธŒ๋Ÿฌ๋ฆฌ๋ฅผ ์ฃผ์„์ฒ˜๋ฆฌํ•ฉ๋‹ˆ๋‹ค.

    # requirements.txt
    
    ...
    
    # tesseract-ocr
    # pytesseract
    
    โ€ฆ
    
  • s3.py <โ€” Make new file name 's3.py' in ./Backend

    AWS_ACCESS_KEY = <AWS ACCESS KEY>
    AWS_SECRET_KEY = <AWS SECRET KEY>
    BUCKET_NAME = <AWS S3 bucket name>

Docker ๐Ÿณ

docker-compose up โ€”build

Docker

๋ฐฐํฌ

API

gif ์ด๋ฏธ์ง€ ๋„ฃ๋Š”๊ฑฐ/ ์ˆ˜์ •ํ•˜๋Š”๊ฑฐ

๋ฉค๋ฒ„ ์†Œ๊ฐœ


Members

์ด๋ฆ„ ๊ฐœ๋ฐœ๋ถ„์•ผ ๋‹ด๋‹น
์ตœ์œค์žฌ Front-end, Back-end Web development,API Design, Cloud
์ด์ฑ„๋ฆผ Front-end Web development
๋ฐ•์‹ ์˜ Front-end Web development
์ดํ•˜์˜ Back-end API Design
Ryan Lee Deep learning Algorithm
๋ฐ•๊ทผ์šฐ Devops Cloud, Docker

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 86.7%
  • JavaScript 11.3%
  • Other 2.0%