Environment

pip install -r requirements.txt

File Structure

DP_PROJECT
│   README.md
│   requirements.txt
│
├── algorithm
│   ├── laplace_module.py       # contains the core `Laplace` class implementation
│   ├── sum_dp_module.py        # contains the core `SumDP` class implementation
│   └── __init__.py
│
├── error_evaluation
│   ├── laplace_eva.py          # error-analysis baseline code
│   ├── SumDP_eva.py            # error‐analysis code
│   └── __init__.py
│
├── notebook_test
│   ├── errors_compare.ipynb    # comparison of Laplace vs. SumDP errors
│   └── SumDP_error.ipynb       # detailed SumDP usage examples
│
├── datasets
│   ├── insurance.csv
│   ├── loan_data.csv
│   ├── readme.md
│   └── StudentsPerformance.csv
│
└── docs
    ├── Sum estimation under differential privacy (proof and formula design).md
    ├── Sum estimation under differential privacy.pdf
    ├── Paper explanation.md
    └── Paper explanation.pdf

• sum_dp_module.py: Implements the SumDP class (the core algorithm).
• error_evaluation/SumDP_eva.py: Contains the error‐analysis implementation.
• Notebooks
  • SumDP_error.ipynb: Detailed code examples for calling and testing SumDP.
  • errors_compare.ipynb: End‐to‐end reproduction of both Laplace and SumDP error experiments.

Reproduction

To reproduce all experiments, open the notebooks in the notebook_test folder:

• errors_compare.ipynb
• SumDP_error.ipynb

Ensure that your Jupyter kernel is using the environment installed via:

pip install -r requirements.txt