LFF-POS-Tagging

LFF POS Tagging is a PyTorch implementation of a BiLSTM-CRF sequence labeler that fuses morphological, orthographic, and character-level features—designed to boost tagging accuracy for rare and out-of-vocabulary words in Indonesian corpora.

🚀 Overview

LFF POS Tagging is a PyTorch-based implementation of a BiLSTM-CRF model with Linguistic Feature Fusion. It fuses multiple linguistic feature sets—morphological, orthographic, and character-level—to improve tagging accuracy, especially for rare and out-of-vocabulary words.

✨ Features

• Linguistic Feature Fusion: Dynamically weights and combines morphology, orthography, and character features.
• BiLSTM-CRF Architecture: State-of-the-art sequence labeling framework.
• Evaluation Suite: Computes accuracy and macro F1-score on Indonesian POS corpora.
• Configurable Pipeline: Easy to plug in new feature extractors or datasets.

🛠 Requirements

• Python 3.8+
• PyTorch 1.10+
• NumPy, pandas, scikit-learn
• (Optional) GPU with CUDA support

💾 Installation

1. Clone the repository

git clone https://github.com/your-username/LFF-POS-Tagging.git
cd LFF-POS-Tagging

2. Clone the repository

python -m venv venv
source venv/bin/activate   # Linux/macOS
venv\Scripts\activate 

3. Install dependencies

pip install -r requirements.txt 

⚙️ Usage

1. Preprocess your data:

python scripts/preprocess.py --input raw_corpus.txt --output data/

2. Train the model:

python train.py \
  --train data/train.json \
  --dev data/dev.json \
  --config configs/lff_config.yaml

3. Evaluate and infer:

python evaluate.py --model checkpoints/best.pt --test data/test.json
python predict.py --model checkpoints/best.pt --input "Saya suka NLP."

📊 Evaluation

After training, evaluation metrics will be logged to logs/ and a summary printed:

• Accuracy
• Macro F1-Score

🤝 Contributing

1. Fork the repo
2. Create a feature branch (git checkout -b feature/YourFeature)
3. Commit your changes (git commit -m "Add YourFeature")
4. Push to the branch (git push origin feature/YourFeature)
5. Open a Pull Request Please follow the existing code style and add tests for new functionality.

📄 License

This project is released under the MIT License. See LICENSE for details.