PGM
Probabilistic graphs: Representation, Learning, and Inference
Features
-
Representation
- Bayesian Network Representation
- Linked List BN Representation
- Linked List MN Representation
- Conditional Estimation
- Marginal Estimation
- Joint Estimation
-
Inference
- Metropolis-Hastings algorithm
- Gibbs Sampling on 2d grid
- Generalized Gibbs Sampling
- Message Parsing and BP
- Loopy BP
- VE
- Causal Interventions
-
search methods
- DFS
- BFS
-
Additional
- Finding Active Trails
- Max clique size and clique node
- Calculate tree-width
- Learning
-
Miscellaneous
- Random BN and MN generation
Installation
pip install ppgm
Contact
- Avinash Kori (koriavinash1@gmail.com)