SODAOpt: Socio-Demographic and Textual Adaptive Fusion for Optimizing Developer Task Assignment

A transformer-based framework that integrates textual and socio-demographic embeddings to enhance task matching.

Installation

1. Clone the repository:

   git clone https://github.com/your-username/RecomText.git
   cd SODAOpt

2. Create and activate virtual environment:

   python -m venv venv
   source venv/bin/activate  # for Linux/Mac
   venv\Scripts\activate  # for Windows

3. Install dependencies:

   pip install -r requirements.txt

Data Preparation

1. Place your source data in the data/ directory:

   • train_events.csv
   • repo_info.csv
   • train_targets.csv
   • all_events.csv

2. Run the data preparation script:

   python -m data.baseline_socdem

Training

1. Configure parameters in configs/config.yaml

2. Start training via Jupyter notebook:

   jupyter notebook notebooks/train.ipynb

   Alternatively, use the training script:

   python train.py

Inference

Use the following code for getting recommendations:

допишу чуть позже

Project Structure

• data/ - data processing modules
• models/ - model architectures
• utils/ - utility functions
• notebooks/ - jupyter notebooks
• configs/ - configuration files

System Requirements

• Python 3.8+
• CUDA-compatible GPU (optional)

Features

• Multimodal embeddings combining text and IDs
• Contrastive learning approach
• Efficient batch processing
• Support for both CPU and GPU inference
• Configurable model architecture

License

MIT License

Contributing

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

Cite

add later