Javabert

🚀 JavaBERT Model Card

JavaBERT is a BERT - like model pretrained on Java software code, offering Fill - Mask capabilities.

🚀 Quick Start

Use the code below to get started with the model.

from transformers import pipeline
pipe = pipeline('fill-mask', model='CAUKiel/JavaBERT')
output = pipe(CODE) # Replace with Java code; Use '[MASK]' to mask tokens/words in the code.

✨ Features

• Pretrained on Java software code.
• Capable of Fill - Mask tasks.

📦 Installation

No installation steps are provided in the original document.

💻 Usage Examples

Basic Usage

from transformers import pipeline
pipe = pipeline('fill-mask', model='CAUKiel/JavaBERT')
output = pipe(CODE) # Replace with Java code; Use '[MASK]' to mask tokens/words in the code.

📚 Documentation

Model Details

Model Description

A BERT - like model pretrained on Java software code.

• Developed by: Christian - Albrechts - University of Kiel (CAUKiel)
• Shared by [Optional]: Hugging Face
• Model type: Fill - Mask
• Language(s) (NLP): en
• License: Apache - 2.0
• Related Models: A version of this model using an uncased tokenizer is available at [CAUKiel/JavaBERT - uncased](https://huggingface.co/CAUKiel/JavaBERT - uncased).
  • Parent Model: BERT
• Resources for more information:
  • Associated Paper

Uses

Direct Use

Fill - Mask

Out - of - Scope Use

The model should not be used to intentionally create hostile or alienating environments for people.

Bias, Risks, and Limitations

Significant research has explored bias and fairness issues with language models (see, e.g., [Sheng et al. (2021)](https://aclanthology.org/2021.acl - long.330.pdf) and Bender et al. (2021)). Predictions generated by the model may include disturbing and harmful stereotypes across protected classes; identity characteristics; and sensitive, social, and occupational groups.

Recommendations

Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.

Training Details

Training Data

The model was trained on 2,998,345 Java files retrieved from open source projects on GitHub. A bert - base - cased tokenizer is used by this model.

Training Procedure

Training Objective

A MLM (Masked Language Model) objective was used to train this model.

Environmental Impact

Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).

Citation

BibTeX:

@inproceedings{De_Sousa_Hasselbring_2021,
  address={Melbourne, Australia},
  title={JavaBERT: Training a Transformer - Based Model for the Java Programming Language},
  rights={https://ieeexplore.ieee.org/Xplorehelp/downloads/license - information/IEEE.html},
  ISBN={9781665435833},
  url={https://ieeexplore.ieee.org/document/9680322/},
  DOI={10.1109/ASEW52652.2021.00028},
  booktitle={2021 36th IEEE/ACM International Conference on Automated Software Engineering Workshops (ASEW)},
  publisher={IEEE},
  author={Tavares de Sousa, Nelson and Hasselbring, Wilhelm},
  year={2021},
  month=nov,
  pages={90–95} }

🔧 Technical Details

No detailed technical information is provided in the original document.

📄 License

The model is licensed under the Apache - 2.0 license.