Navigation Menu

Skip to content


Repository files navigation

R Interface to the PM4Py Process Mining Library


The goal of the R package 'pm4py' is to provide a bridge between bupaR and the Python library 'PM4Py'.


You can install the released CRAN version of pm4py with:


You can install the development version of pm4py from the dev branch with:


Then, automatically install the pm4py package in a virtual or Conda environment:


See the 'reticulate' documentation for more information on the available options or how to specify an existing Python environment:

PM4Py Version

To facilitate getting stable results and to reduce the number of regressions due to API changes in PM4Py, this package is built against a fixed PM4Py version that is defined in the file R/version.R. We also adopt the versioning schema of the PM4Py project for this R package. So, the R package version 1.1.19 will install the PM4Py version 1.1.19.

In case of fixes required to the R package itself, for example, for bugs or adopting new features, we will add a suffix -rev to the version to indicate the change. Of course, nothing prevents you from manually overriding the synchronisation between the PM4Py version and the R PM4Py package version using the parameter version as follows:

pm4py::install_pm4py(version = "1.2.7")



# Most of the data structures are converted in their bupaR equivalents

# As Inductive Miner of PM4PY is not life-cycle aware, keep only `complete` events:
patients_completes <- patients[patients$registration_type == "complete", ]

# Discovery with Inductive Miner
pn <- discovery_inductive(patients_completes)

# This results in an auto-converted bupaR Petri net and markings

# Render with bupaR

# Render with  PM4PY and DiagrammeR
viz <- reticulate::import("pm4py.visualization.petrinet")

# Convert back to Python
py_pn <- r_to_py(pn$petrinet)

# Render to DOT with PMP4Y
dot <- viz$factory$apply(py_pn)$source
grViz(diagram = dot)

# Compute alignment
alignment <- conformance_alignment(patients_completes, pn$petrinet, pn$initial_marking, pn$final_marking)

# # Alignment is returned in long format as data frame

# Evaluate model quality
quality <- evaluation_all(patients_completes, pn$petrinet, pn$initial_marking, pn$final_marking)