st22-beautify-berlin

Deep Learning

Deployment

• Dedicated repository: https://github.com/allakulov/techlabs_flask_app
• Heroku app: https://electricbox-dl.herokuapp.com/
• Prediction endpoint API: https://electricbox-dl.herokuapp.com/predict

Model

• Colab notebook: https://github.com/TechLabs-Berlin/st22-beautify-berlin/blob/main/cnn_model.ipynb
• Images: https://drive.google.com/drive/folders/18fl971UFKniAe6sH1oiclqtraRP1UB1k?usp=sharing
• Labels: https://docs.google.com/spreadsheets/d/18sq8x2CHwXdAo_dGpg7U7yKt5J5Ix4RT8gH2lUyzw98/edit?usp=sharing

Prototype

• Prototype: https://www.figma.com/file/8ivTAUNeGlCChKno5PKCTR/Beautify-Berlin-Team

Design

• Purple color: rgba(100, 75, 150, 1)
• Font: https://fonts.google.com/specimen/Inter
• Logo font: https://fonts.google.com/specimen/Pacifico
• Button colors:
  • Green: rgba(75, 150, 87, 1)
  • Red: rgba(150, 75, 75, 1)
• Status colors:
  • Art: rgba(100, 75, 150, 1)
  • Clean: rgba(0, 128, 0, 1)
  • Dirty: rgba(255, 0, 0, 1)
  • Poster: rgba(22, 159, 234, 1)
  • None: rgba(0, 0, 0, 1)

Web Development

Useful links

• React: https://reactjs.org/
• MUI: https://mui.com/
• Firebase: https://firebase.google.com/?gclid=Cj0KCQjwuO6WBhDLARIsAIdeyDLHnvntuPxKy3sDouwmNmePnScW5jL9pdsytfLIYxRN0IzQjsXkmNYaAumyEALw_wcB&gclsrc=aw.ds

How to run the code

1. Clone the repository.
2. Go to the directory /beautifyberlin.
3. Run "npm start" (starts the development server on port 3000).

It opens the browser at http://localhost:3000 automatically.