Bayesian Network Inference
Exact and approximate ways to answer probability questions on Bayesian networks.
A university project on Bayesian networks: graphs of variables and how they depend on each other.
What it does
- Defines how a Bayesian network is stored.
- Simplifies the network before answering (pruning and min-fill ordering).
- Gives exact answers with the jointree algorithm.
- Gives approximate answers with iterative join-graph propagation (IJGP).
- Tests both on five networks, from small (Cancer) to massive (Munin, over 1,000 nodes).
Built with
- Python
- Jupyter Notebook