DewCrypt is a state-of-the-art model designed for detecting vulnerabilities in smart contracts, specifically within the Dew-Cloud framework. This repository contains the source code, datasets, and evaluation metrics for the DewCrypt model. The model has been rigorously evaluated against existing baseline algorithms and has undergone an extensive ablation study to assess the contributions of each component.
Instructions for setting up the DewCrypt environment.
# Clone the repository
git clone https://github.com/yourusername/DewCrypt-Project.git
# Navigate to the project directory
cd DewCrypt-Project
# Install dependencies
pip install -r requirements.txtHow to run the DewCrypt model.
# Run the main script
python main.py- Computational Efficiency: DewCrypt operates at a computational efficiency of 120 gs.
- Decision Time Latency: The model has a low decision time latency of 2.1 ms.
- Success Rate: DewCrypt boasts a success rate of 99%.
- Accuracy: The model has an accuracy rate of 98%.
DewCrypt has been evaluated against a range of baseline algorithms, including Oyente, Mythril, SODA, and AEGIS. The model demonstrates superior performance across all metrics. For more details, refer to the Comparative Analysis section.
An ablation study has been conducted to assess the contributions of each component in the DewCrypt model. The study confirms the scientific rationale behind incorporating these elements. For more details, refer to the Ablation Study section.
We welcome contributions to the DewCrypt project. Please refer to the Contributing Guidelines for more information.
This project is licensed under the MIT License - see the LICENSE.md file for details.
If you use DewCrypt in your research, please cite our paper. The citation details can be found in the Citation section.
For any questions, feedback, or suggestions regarding the PGT project, please reach out to the project maintainers:
- Rexford Sosu
- Email: [email protected]
- GitHub: @rexfordsosu
- LinkedIn: Rexford's LinkedIn
We appreciate your interest in the DewCrypt project and look forward to collaborating with you!