bayesnet_tool

Design Space Exploration with Bayesian Networks

A Streamlit-based tool for designing and analyzing Bayesian networks with interactive visualization and probabilistic querying.

Requirements

• Python 3.11 (due to dependency constraints with pgmpy and numpy/pandas)

Installation

1. Clone the repository

git clone https://github.com/spinjet/bayesnet_tool
cd bayesnet_tool

2. Install the package with dependencies

pip install .

For development (editable mode):

pip install -e .

Note: Ensure you have Python 3.11 installed and active before running the above commands.

Running the Application

Once the virtual environment is activated, start the Streamlit application:

streamlit run run.py

The application will open in your default browser at http://localhost:8501.

Features

• Data Loading: Import and preprocess CSV datasets
• Missing Data Management: Handle NaN values with configurable thresholds
• Structure Learning: Discover Bayesian network structure from data
• Discretisation: Convert continuous variables to discrete states
• Training: Learn network parameters from data
• Visualization: Interactive network graph with topological layout
• Probabilistic Queries: Calculate conditional probabilities and inferences