This repository demonstrates the implementation of a deep learning model trained with Differential Privacy (DP) and deployed using a Flask API. The project utilizes TensorFlow Privacy (TFP) to ensure data privacy while maintaining model utility.This repository demonstrates the implementation of a deep learning model trained with Differential Privacy (DP) and deployed using a Flask API. The project utilizes TensorFlow Privacy (TFP) to ensure data privacy while maintaining model utility.
- Implementation of differentially private stochastic gradient descent (DPSGD) for model training
- Deployment of the trained model via a Flask API for inference
- Resources and links to further readings, tutorials, and relevant code libraries
DifferentialPrivacy ├── Dp.py └── app.py └── requirements.txt └── README.md
- Clone the repository and navigate to its directory:
git clone https://github.com/BVChandrahaas/DifferentialPrivacy.git cd DifferentialPrivacy
- Install the required dependencies:
pip install -r requirements.txt
- Train the Model Run the dp_training.py script to train the model with differential privacy:
python Dp.py
The trained model will be saved in the saved_model/ directory. Run the Flask App Start the Flask API server:
python app.py
By default, the app runs at http://127.0.0.1:5000.
- Papers
- Deep Learning with Differential Privacy by Abadi et al.
- Federated Learning with Differential Privacy by McMahan et al.
- Code Libraries
- TensorFlow Privacy
- PyTorch-DP
- Tutorials
- Differential Privacy Tutorial by Stanford Natural Language Processing Group
- Federated Learning with Differential Privacy by TensorFlow Team
We welcome contributions! Feel free to:
- Fork this repository
- Submit pull requests
- Report issues
Your contributions will help improve Vanilla Split Learning.
This project is licensed under the MIT License.
See LICENSE for details.