awiegel / PaSyPy

A python-based tool using parameter synthesis to find safe and unsafe regions of the parameter space

PaSyPy

A python-based tool using parameter synthesis to find safe and unsafe regions of the parameter space on nonlinear real arithmetic.

For a detailed description of this project, read the publication.

Table of contents

• 1. Installation
• 2. Functionality / Usage
• 3. Suggestions / Bug reports

1. Installation

To install this tool, clone the directory and install all dependencies:

• z3-solver
• matplotlib
• scikit-learn
• numpy

All dependencies are also included in requirements.txt and can be installed by
pip install -r requirements.txt.

2. Functionality / Usage

To start this tool, simply execute __main__.py.

Functionality:

• SMT-LIB Support: Read .smt2 files or define your own formula in the text field.
• Find safe and unsafe regions: Separate the whole region of the parameter space into safe (green) and unsafe (red) regions. Select from different splitting heuristics to obtain optimal results.
• Visualization: Get a visualization for found safe (green) and unsafe (red) regions. Choose which parameters you want to see.
• Information: Lots of different information about your computation.
• Settings: Several settings available.

3. Suggestions / Bug reports

If you encounter any problems while using this tool or have any suggestions or feedback,
feel free to contact me.
Read the Known challenges section in the documentation first for a listing of all problems and suggestions.