Package to deal with particles, the PDG particle data table, PDGIDs, etc.





GitHub Stars



Last Commit

12d ago










.. image:: :target:

Particle: PDG particle data and identification codes

|Scikit-HEP| |PyPI version| |Conda-forge version| |Zenodo DOI|

|GitHub Actions Status: CI| |Code Coverage| |Code style: black|


Particle provides a pythonic interface to the Particle Data Group <>_ (PDG) particle data tables and particle identification codes, with extended particle information and extra goodies.

The PDG defines the standard particle identification (ID) numbering scheme. The package provides the PDGID class implementing queries on those PDG IDs. The queries are also accessible through free standing functions mimicking, and expanding from, the HepPID/HepPDT C++ interface.

The Particle class wraps the information in the PDG particle data tables and provides an object-oriented interface and powerful search and look-up utilities.


Install particle like any other Python package:

.. code-block:: bash

pip install particle

or similar (use --user, virtualenv, etc. if you wish).

Strict dependencies

  • Python <>_ (2.7+, 3.5+)
  • importlib_resources backport <>_ if using Python < 3.7
  • attrs <>_ provides classes without boilerplate (similar to DataClasses in Python 3.7)
  • hepunits <>_ provides units for the Scikit-HEP packages


See the changelog <>__ for a history of notable changes.

Getting started: PDG IDs

.. code-block:: python

>>> from particle import PDGID
>>> pid = PDGID(211)
>>> pid
<PDGID: 211>
>>> pid.is_meson
>>> pid = PDGID(99999999)
>>> pid
<PDGID: 99999999 (is_valid==False)>

For convenience, all properties of the PDGID class are available as standalone functions that work on any SupportsInt (including Particle):

.. code-block:: python

>>> from particle.pdgid import is_meson
>>> is_meson(211)

These composable functions qualifying PDG IDs make it easy to classify particles. For the sake of example, quarkonia can be specified with the following user-defined functions:

.. code-block:: python

>>> is_heavy_flavor = lambda x: has_charm(x) or has_bottom(x) or has_top(x)
>>> is_quarkonium = lambda x: is_meson(x) and three_charge(x)==0 and is_heavy_flavor(x).

PDG ID literals provide (PDGID class) aliases for all particles loaded, with easily recognisable names. For example:

.. code-block:: python

>>> from particle.pdgid import literals as lid
>>> lid.pi_plus
<PDGID: 211>
>>> from particle.pdgid.literals import Lambda_b_0
>>> Lambda_b_0
<PDGID: 5122>
>>> Lambda_b_0.has_bottom

You can quickly display PDGID info from the command line with:

.. code-block:: bash

$ python -m particle pdgid 323
<PDGID: 323>
A              None
J              1.0
L              0
S              1
Z              None
abspid         323
charge         1.0
has_bottom     False

Similarly, classes exist to express identification codes used by MC programs, see information on converters below.

Getting started: Particles

You can use a variety of methods to get particles. If you know the PDG ID number you can get a particle directly, or you can use a search:

.. code-block:: python

>>> from particle import Particle
>>> Particle.from_pdgid(211)
<Particle: name="pi+", pdgid=211, mass=139.57061 ± 0.00024 MeV>
>>> Particle.findall('pi')[0]
<Particle: name="pi0", pdgid=111, mass=134.9770 ± 0.0005 MeV>

You can search for the properties using keyword arguments, which include pdg_name, name, mass, width, charge, three_charge, anti_flag, rank, I, J, G, P, quarks, status, mass_upper, mass_lower, width_upper, and width_lower. You can pass a callable or an exact match for any property. The argument particle can be set to True/False, as well, to limit the search to particles or antiparticles. You can also build the search yourself with the first positional argument, which accepts a callable that is given the particle object itself. If the first positional argument is a string, that will match against the particle's name. The alternative .find() requires only one match returned by the search, and will throw an error if more or less than one match is found.

Here are possible sophisticated searches:

.. code-block:: python

>>> # Print out all particles with asymmetric decay width uncertainties
>>> ps = Particle.findall(lambda p: p.width_lower != p.width_upper)
>>> for p in ps:
...     print(, p.pdgid, p.width_lower, p.width_upper)
>>> # Find all antiparticles with 'Omega' in the name
>>> Particle.findall('Omega', particle=False)   # several found
>>> # Find all antiparticles of name=='Omega'
>>> Particle.findall(name='Omega', particle=False)  # none found
>>> # Find all antiparticles of pdg_name=='Omega'
>>> Particle.findall(pdg_name='Omega', particle=False)  # only 1, of course
[<Particle: name="Omega~+", pdgid=-3334, mass=1672.5 ± 0.3 MeV>]
>>> # Find all neutral beauty hadrons
>>> Particle.findall(lambda p: p.pdgid.has_bottom and p.charge==0)
>>> # Find all strange mesons with c*tau > 1 meter
>>> from hepunits import meter
>>> Particle.findall(lambda p: p.pdgid.is_meson and p.pdgid.has_strange and p.ctau > 1 * meter, particle=True)
[<Particle: name="K(L)0", pdgid=130, mass=497.611 ± 0.013 MeV>,
 <Particle: name="K+", pdgid=321, mass=493.677 ± 0.016 MeV>]

Once you have a particle, any of the properties can be accessed, along with several methods. Though they are not real properties, you can access is_name_barred, and spin_type. You can also .invert() a particle.

There are lots of printing choices for particles: describe(), programmatic_name, latex_name, html_name, HTML printing outs in notebooks, and of course repr and str support.

You can get the .pdgid from a particle, as well. Sorting particles will put lowest abs(PDGID) first.

Particle literals provide (Particle class) aliases for the particles loaded, with easily recognisable names. For example:

.. code-block:: python

>>> from particle import literals as lp
>>> lp.pi_plus
<Particle: name="pi+", pdgid=211, mass=139.57061 ± 0.00024 MeV>
>>> from particle.literals import Lambda_b_0
>>> Lambda_b_0
<Particle: name="Lambda(b)0", pdgid=5122, mass=5619.60 ± 0.17 MeV>
>>> Lambda_b_0.J

You can quickly search for particles from the command line with (note: quotes may be used/needed but only double quotes work as expected on Windows):

.. code-block:: bash

$ python -m particle search "K*0"
<Particle: name="K*(892)0", pdgid=313, mass=895.55 ± 0.20 MeV>
<Particle: name="K*(1680)0", pdgid=30313, mass=1718 ± 18 MeV>
<Particle: name="K*(1410)0", pdgid=100313, mass=1421 ± 9 MeV>

If you only select one particle, either by a search or by giving the PDG ID number, you can see more information about the particle:

.. code-block:: bash

$ python -m particle search 311
Name: K0             ID: 311          Latex: $K^{0}$
Mass  = 497.611 ± 0.013 MeV
Width = -1.0 MeV
Q (charge)        = 0       J (total angular) = 0.0      P (space parity) = -
C (charge parity) = ?       I (isospin)       = 1/2      G (G-parity)     = ?
    SpinType: SpinType.PseudoScalar
    Quarks: dS
    Antiparticle name: K~0 (antiparticle status: Barred)

Advanced: Loading custom tables ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

You can control the particle data tables if you so desire. You can append a new data table using the following syntax:

.. code-block:: python

>>> from particle import Particle
>>> Particle.load_table('new_particles.csv', append=True)

You can also replace the particle table entirely with append=False (the default).

If you want a non-default data file distributed with the package just proceed as follows:

.. code-block:: python

>>> from particle import data
>>> Particle.load_table(data.basepath / "particle2018.csv"))
>>> Particle.load_table(data.basepath / "nuclei2020.csv"), append=True)  # I still want nuclei info
>>> Particle.table_names()  # list the loaded tables

Advanced: Conversion ^^^^^^^^^^^^^^^^^^^^

You can convert and update the particle tables with the utilities in particle.particle.convert. This requires the pandas package, and is only tested with Python 3. Run the following command for more help:

.. code-block:: bash

$ python3 -m particle.particle.convert --help

Getting started: Converters

You can use mapping classes to convert between particle MC identification codes and particle names. See the particle.converters modules for the available mapping classes. For example:

.. code-block:: python

>>> from particle.converters import Pythia2PDGIDBiMap
>>> from particle import PDGID, PythiaID
>>> pyid = Pythia2PDGIDBiMap[PDGID(9010221)]
>>> pyid
<PythiaID: 10221>

>>> pdgid = Pythia2PDGIDBiMap[PythiaID(10221)]
>>> pdgid
<PDGID: 9010221>

This code makes use of classes similar to PDGID, which hold particle identification codes used by MC programs. Possible use cases are the following:

.. code-block:: python

>>> from particle import Particle
>>> from particle import Geant3ID, PythiaID
>>> g3id = Geant3ID(8)
>>> p = Particle.from_pdgid(g3id.to_pdgid())
>>> p = Particle.find(pdgid=g3id.to_pdgid())

>>> pythiaid = PythiaID(211)
>>> p = Particle.from_pdgid(pythiaid.to_pdgid())

>>> p = Particle.find(pdgid=pythiaid.to_pdgid())


Support for this work was provided by the National Science Foundation cooperative agreement OAC-1450377 (DIANA/HEP) and OAC-1836650 (IRIS-HEP). Any opinions, findings, conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.

.. |Scikit-HEP| image:: :target:

.. |PyPI version| image:: :target:

.. |Conda-forge version| image:: :target:

.. |Zenodo DOI| image:: :target:

.. |GitHub Actions Status: CI| image:: :target:

.. |Code Coverage| image:: :target:

.. |Code style: black| image:: :target:

.. |Binder| image:: :target:

Rate & Review

Great Documentation0
Easy to Use0
Highly Customizable0
Bleeding Edge0
Responsive Maintainers0
Poor Documentation0
Hard to Use0
Unwelcoming Community0
No reviews found
Be the first to rate


No alternatives found


No tutorials found
Add a tutorial