💛 playdoh

What is playdoh?

The playdoh package provides some applications that can be used easily by people who are not familiar with AI. This package offers the following features:

• Image Classification
• Sentiment Analysis
• Speech-to-Text
• Time Series Forecasting

Download models

The pretrained models required for using the Playdoh package are almost 1 GB in size, but we are not using the paid plan of git lfs. Please download the models from the link below and place them in the 'fastapi_backend/pretrained_model' folder.

Download models (Google Drive)

Prerequisite

• Docker
• Visual Studio Code

Development Stack

• Python
• Streamlit
• FastAPI
• Pytorch

Instruction for Dev

1. Clone playdoh and open it in VS Code.
2. Download pretrained models into fastapi_backend/pretrained_model from Google Drive
3. Build Docker images and run containers using Docker Compose with the following command:

   docker-compose up -d
   

4. Once set up, navigate to 127.0.0.1:80 or localhost:80 in your browser.
5. If you make any modifications to the code, either on the frontend or backend, restart the respective container derived from the corresponding image.

Task tutorial

Click the image below to view other tutorial videos.

Authors

• Dongryul Min @MinDongRyul

License

Copyright ©️ 2023 PHI This project is MIT licensed