pgmpy

pgmpy is a python library for working with Probabilistic Graphical Models.

Documentation and list of algorithms supported is at our official site http://pgmpy.org/
Examples on using pgmpy: https://github.com/pgmpy/pgmpy/tree/dev/examples
Basic tutorial on Probabilistic Graphical models using pgmpy: https://github.com/pgmpy/pgmpy_notebook

Our mailing list is at https://groups.google.com/forum/#!forum/pgmpy .

We have our community chat at gitter.

Dependencies

pgmpy has following non optional dependencies:

• Python 2.7 or Python 3
• NetworkX 1.9.1
• Scipy 0.12.1
• Numpy 1.9.2
• Cython 0.21
• Pandas 0.15.1

Download

Currently pgmpy is not hosted on pypi or conda. You can either clone the git repo with:

git clone https://github.com/pgmpy/pgmpy

or download a zip from: https://github.com/pgmpy/pgmpy/archive/dev.zip

Installation

To install the dependencies switch to the pgmpy directory using:

$ cd /path/to/pgmpy

In the directory run either of the following:

Using pip

$ pip install -r requirements.txt  # or requirements-dev.txt if you want to run unittests

or conda

$ conda install --file requirements.txt  # or requirements-dev.txt

Then install using:

sudo python setup.py install

If you face any problems during installation let us know, via issues, mail or at our gitter channel.

Development

Code

You can check the latest sources from our github repository use the command:

git clone https://github.com/pgmpy/pgmpy.git

Contributing

Issues can be reported at our issues section or via mail, or gitter. We will try our best to solve the issue at the earliest.

Before opening a pull request , have look at our contributing guide

Contributing guide contains some points that will make our life's easier in reviewing and merging your PR.

If you face any problems in pull request, feel free to ask them at mail or gitter.

If you have any new features, please have a discussion on the issue tracker or the mailing list before working on it.

Testing

After installation, you can launch the test form pgmpy source directory (you will need to have the nose package installed):

$ nosetests -v

to see the coverage of existing code use following command

$ nosetests --with-coverage --cover-package=pgmpy

Documentation and usage

Everything is at: http://pgmpy.org/

You can also build the documentation in your local system. We use sphinx to help us building documentation from our code.

$ cd /path/to/pgmpy/docs
$ make html

Then the docs will be in _build/html

Example:

Here is a small snippet of pgmpy API

>>> from pgmpy.models import BayesianModel
>>> from pgmpy.factors.discrete import TabularCPD
>>> student = BayesianModel()
>>> # instantiates a new Bayesian Model called 'student'

>>> student.add_nodes_from(['diff', 'intel', 'grade'])
>>> # adds nodes labelled 'diff', 'intel', 'grade' to student

>>> student.add_edges_from([('diff', 'grade'), ('intel', 'grade')])
>>> # adds directed edges from 'diff' to 'grade' and 'intel' to 'grade'

>>> """
... diff cpd:
...
... +-------+--------+
... |diff:  |        |
... +-------+--------+
... |easy	|	0.2	 |
... +-------+--------+
... |hard	|	0.8	 |
... +-------+--------+
... """

>>> diff_cpd = TabularCPD('diff', 2, [[0.2], [0.8]])

>>> """
... intel cpd:
...
... +-------+--------+
... |intel: |        |
... +-------+--------+
... |dumb	|	0.5	 |
... +-------+--------+
... |avg	|	0.3	 |
... +-------+--------+
... |smart	|	0.2	 |
... +-------+--------+
... """

>>> intel_cpd = TabularCPD('intel', 3, [[0.5], [0.3], [0.2]])

>>> """
... grade cpd:
...
... +------+-----------------------+---------------------+
... |diff: |          easy         |         hard        |
... +------+------+------+---------+------+------+-------+
... |intel:| dumb |  avg |  smart  | dumb | avg  | smart |
... +------+------+------+---------+------+------+-------+
... |gradeA| 0.1  | 0.1  |   0.1   |  0.1 |  0.1 |   0.1 |
... +------+------+------+---------+------+------+-------+
... |gradeB| 0.1  | 0.1  |   0.1   |  0.1 |  0.1 |   0.1 |
... +------+------+------+---------+------+------+-------+
... |gradeC| 0.8  | 0.8  |   0.8   |  0.8 |  0.8 |   0.8 |
... +------+------+------+---------+------+------+-------+
... """

>>> grade_cpd = TabularCPD('grade', 3,
					     [[0.1,0.1,0.1,0.1,0.1,0.1],
                         [0.1,0.1,0.1,0.1,0.1,0.1], 
                         [0.8,0.8,0.8,0.8,0.8,0.8]],
					     evidence=['intel', 'diff'],
					     evidence_card=[3, 2])

>>> student.add_cpds(diff_cpd, intel_cpd, grade_cpd)

>>> # Finding active trail
>>> student.active_trail_nodes('diff')
{'diff', 'grade'}

>>> # Finding active trail with observation
>>> student.active_trail_nodes('diff', observed='grade')
{'diff', 'intel'}

License

pgmpy is released under MIT License. You can read about our lisence at here