Skip to content

Fill out the form, predict the diabetes stage of the form filler using a persistent pre-trained model, and visualize real-time data in the backend

License

Notifications You must be signed in to change notification settings

RogelioKG/GlucoGuard

Repository files navigation

GlucoGuard

Version 0.2.1 Python 3.11.4 Last Updated Coverage Licence

Brief

填寫表單並透過持久化的預訓練模型,預測當前填表者的糖尿病階段,
使用 PostgreSQL 儲存填表者資料,並使用 plotly 做為後臺實時資料可視化。

glucoguard

Requirements

  • flask
  • python-dotenv
  • psycopg2-binary
  • Flask-SQLAlchemy
  • pandas
  • xgboost
  • joblib
  • plotly
  • coverage
  • pytest
  • tqdm

Running the Flask Application

若你已設置好虛擬環境,並開啟 Docker 應用程式,你可以這樣運行此應用

Caution

由於存在第三方 JS 腳本 / CSS 樣式表,強烈建議連網運行,否則將導致功能或格式缺損

  • Windows

    • 本地運行
      scripts\dev_run.bat
    • Docker 容器運行
      scripts\docker_dev_run.bat

Flask CLI

若為本地運行,可以使用以下命令列進行測試

  • 清空資料庫
    flask db clear
  • 生成 500 筆填表者 mock data
    flask db generate --volunteer 500

Testing

  • 單元測試 (並產生 .coverage)
    scripts\test_run.bat
  • 覆蓋率報告 (需存在 .coverage)
    scripts\test_report.bat

Changelog

See notable changes to this project in CHANGELOG.md.

About

Fill out the form, predict the diabetes stage of the form filler using a persistent pre-trained model, and visualize real-time data in the backend

Topics

Resources

License

Stars

Watchers

Forks

Packages

No packages published