tokenizer-Bafia / README.md
FredMike23's picture
Upload README.md with huggingface_hub
13f4358 verified
# bafia Tokenizer for NLP tasks
## Model Description
This tokenizer was developed for bafia, a language from the fula[ksf] family of languages in Cameroon. The tokenizer is based on the WordPiece model architecture and has been fine-tuned to handle the unique phonetic and diacritical features of the Fulfulde language.
- **Developed by**: DS4H-ICTU Research Group in Cooperation with the
- **Language(s)**: bafia (bafia[ksf] language from Cameroon)
- **License**: Apache 2.0 (or specify if different)
- **Model Type**: Tokenizer (WordPiece)
## Model Sources
- **Repository**: [Your repository URL]
- **Paper**: [Link to related paper if available]
- **Demo**: [Optional: link to demo]
## Uses
- **Direct Use**: This tokenizer is designed for NLP tasks such as Named Entity Recognition (NER), translation, and text generation in the bafia language.
- **Downstream Use**: Can be used as a foundation for models processing bafia text.
## Bias, Risks, and Limitations
- **Biases**: The tokenizer might not perfectly capture linguistic nuances due to the limited size of the bafia corpus.
- **Out-of-Scope Use**: The tokenizer may not perform well for non-bafia languages.
## Training Details
- **Training Data**: Extracted from bafia Bible text corpus (bafia_DATASET.xlsx).
- **Training Procedure**: Preprocessing of text involved normalization of diacritics, tokenization using WordPiece, and post-processing to handle special tokens.
- **Training Hyperparameters**:
- Vocabulary Size: 19076
- Special Tokens: "[UNK]", "[PAD]", "[CLS]", "[SEP]", "[MASK]", "[BOS]", "[EOS]"
## Evaluation
- **OOV Rate**: 0.00%
- **Tokenization Efficiency**: Average tokens per sentence: 27.585227817745803
- **Special Character Handling**: Successfully handles diacritics and tone markers in bafia.
## Environmental Impact
- **Hardware Type**: Google Colab GPU
- **Hours Used**: 4 hours (training time)
- **Cloud Provider**: Google Cloud
- **Carbon Emitted**: Estimated using [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700) calculator
## Citation
If you use this tokenizer in your work, please cite it using the following format:
```
@misc{bafia_tokenizer,
title = {bafia Tokenizer},
author = {Ing. Zingui Fred Mike},
year = {2024},
publisher = {Hugging Face},
url = {https://huggingface.co/FredMike23/tokenizer-Bafia}
}
```
## Contact Information
For more information, contact the developers at: philiptamla@gmail.com