pgmpy is a Python package for causal inference and probabilistic inference using Directed Acyclic Graphs (DAGs) and Bayesian Networks with a focus on modularity and extensibility. Implementations of various algorithms for Causal Discovery (a.k.a, Structure Learning), Parameter Estimation, Approximate (Sampling Based) and Exact inference, and Causal Inference are available.
Possible Workflows in pgmpy for Directed Acyclic Graphs (DAGs) and Bayesian Networks (BNs).¶
Supported Data Types¶
Casual Discovery / Structure Learning |
Parameter Estimation |
Causal Inference |
Probabilistic Inference |
|
|---|---|---|---|---|
Discrete |
Yes |
Yes |
Yes |
Yes |
Continuous |
Yes (only PC) |
Yes |
Yes (partial) |
Yes |
Hybrid |
Yes (only PC) |
No |
No |
No |
Time Series |
No |
Yes |
Yes (ApproximateInference) |
Yes |
Algorithms¶
Structure Learning |
Parameter Estimation |
Probabilistic Inference |
Causal Inference |
|---|---|---|---|
PC with variants |
Maximum Likelihood |
Variable Elimination |
do-operation |
Hill-Climb Search |
Bayesian Estimator |
Belief Propagation |
adjustment sets |
Tree Search |
Expectation Maximization |
MPLP |
|
Max-Min Hill-Climb |
Sampling methods |
||
Exhaustive Search |
Examples¶
Example notebooks: https://pgmpy.org/examples.html
Tutorial notebooks: https://pgmpy.org/tutorial.html
Citation¶
If you use pgmpy in your scientific work, please consider citing us:
Ankan, Ankur, Abinash, Panda. "pgmpy: Probabilistic Graphical Models using Python." Proceedings of the Python in Science Conference. SciPy, 2015.
Bibtex:
@inproceedings{Ankan2015,
series = {SciPy},
title = {pgmpy: Probabilistic Graphical Models using Python},
ISSN = {2575-9752},
url = {http://dx.doi.org/10.25080/Majora-7b98e3ed-001},
DOI = {10.25080/majora-7b98e3ed-001},
booktitle = {Proceedings of the Python in Science Conference},
publisher = {SciPy},
author = {Ankan, Ankur and Panda, Abinash},
year = {2015},
collection = {SciPy}
}
