Text Tagging - Hateful Comments Detection

About   |   Technologies   |   Requirements   |   Starting   |   License   |   Author

🎯 About

Deployed a model mimicing cloud production server using Docker, Pytorch, HuggingFace, Git and FastAPI. Reduced the model size by quantization and inference time by singleton design pattern approach.

🚀 Technologies

The following tools were used in this project:

• Docker
• Python
• Git
• FastAPI
• Torch
• Hugging Face

✅ Requirements

Before starting 🏁, you need to have Git, Docker and Python installed.

🏁 Starting

# Clone this project
$ git clone https://github.com/human-ai2025/Text-Tagging

# Access
$ cd Text-Tagging


# After doing this, it will give you the model weights [make sure to use the same versions of tensorflow in docker and while buiding the model]

# For docker 
$ docker compose build -- up

# This will fire up the web serice for making predictions.
# The server will initialize in the "http://localhost:8000/predict"

📝 License

This project is under license from MIT. For more details, see the LICENSE file.

Made with ❤️ by Tarini Tanaya Mohapatra

 

Back to top