Javanese Distilbert Small

🚀 Javanese DistilBERT Small

Javanese DistilBERT Small is a masked language model trained on Javanese Wikipedia articles, based on the DistilBERT architecture.

🚀 Quick Start

Javanese DistilBERT Small is a masked language model based on the DistilBERT model. It was trained on the latest (late December 2020) Javanese Wikipedia articles.

The model was originally HuggingFace's pretrained English DistilBERT model and is later fine - tuned on the Javanese dataset. It achieved a perplexity of 23.54 on the validation dataset (20% of the articles). Many of the techniques used are based on a Hugging Face tutorial notebook written by Sylvain Gugger, and fine - tuning tutorial notebook written by Pierre Guillou.

Hugging Face's Transformers library was used to train the model -- utilizing the base DistilBERT model and their Trainer class. PyTorch was used as the backend framework during training, but the model remains compatible with TensorFlow nonetheless.

✨ Features

• Based on the DistilBERT architecture, a lightweight and efficient model.
• Trained on Javanese Wikipedia articles, suitable for Javanese language tasks.
• Compatible with both PyTorch and TensorFlow.

📦 Installation

The model can be used via the Hugging Face Transformers library. You can install it using pip:

pip install transformers

💻 Usage Examples

Basic Usage

from transformers import pipeline

pretrained_name = "w11wo/javanese-distilbert-small"

fill_mask = pipeline(
    "fill-mask",
    model=pretrained_name,
    tokenizer=pretrained_name
)

fill_mask("Aku mangan sate ing [MASK] bareng konco-konco")

Advanced Usage

from transformers import DistilBertModel, DistilBertTokenizerFast

pretrained_name = "w11wo/javanese-distilbert-small"
model = DistilBertModel.from_pretrained(pretrained_name)
tokenizer = DistilBertTokenizerFast.from_pretrained(pretrained_name)

prompt = "Indonesia minangka negara gedhe."
encoded_input = tokenizer(prompt, return_tensors='pt')
output = model(**encoded_input)

📚 Documentation

Model

Evaluation Results

The model was trained for 5 epochs and the following is the final result once the training ended.

Disclaimer

⚠️ Important Note

Although the dataset originated from Wikipedia, the model may not always generate factual texts. Additionally, the biases which came from the Wikipedia articles may be carried over into the results of this model.

Author

Javanese DistilBERT Small was trained and evaluated by Wilson Wongso. All computation and development are done on Google Colaboratory using their free GPU access.

Citation

If you use any of our models in your research, please cite:

@inproceedings{wongso2021causal,
    title={Causal and Masked Language Modeling of Javanese Language using Transformer-based Architectures},
    author={Wongso, Wilson and Setiawan, David Samuel and Suhartono, Derwin},
    booktitle={2021 International Conference on Advanced Computer Science and Information Systems (ICACSIS)},
    pages={1--7},
    year={2021},
    organization={IEEE}
}

📄 License

This project is licensed under the MIT license.