⛏pyromorphite💎

Process mining in python.

Get the Source Code

Pyromorphite is actively developed on GitHub, where the code is always available.

You can clone the public repository:

$ git clone git://github.com/xcavation/pyromorphite.git 

Once you have a copy of the source, you can embed it in your own Python package, or install it into your site-packages easily:

$ cd requests
$ pip install .

Quickstart

This is a quick introduction to Pyromorphite. Before proceeding, make sure that Pyromorphite is installed.

Import a Log 📜

Reading in event log files in Pyromorphite is super easy. It supports XES, CSV, and Excel files.

XES Files

Begin by importing the Pyromorphite module:

>>> import pyromorphite as pm

We'll try now to get an xes file from a web repository. For this dataset:

>>> log = pm.read_xes("https://data.4tu.nl/repository/uuid:c1e9137e-2877-410d-a76a-21ce7f97a239/DATA1")

Similarly we would do if we would like to read a local file. Let's assume that under /path/to/file.xes lies our file. We can then do:

>>> log = pm.read_xes("/path/to/file.xes")

CSV Files

Although not part of Pyromorphite, reading a csv file can be done with the pandas library.

Begin by import the pandas module:

>>> import pandas as pd

We'll try now to get an xes file from a web repository. For this dataset:

>>> log = pd.read_csv("https://data.4tu.nl/repository/uuid:d5ccb355-ca67-480f-8739-289b9b593aaf/DATA")

Construct a Bag 🎒

Having parsed a log into a pandas DataFrame we can simply extract the traces of events, with togehter with their frequency in the log.

>>> bag = pm.as_bag(log)

If we consider that everybody might use different column names, we can specify the case, timestamp and activity columns when creating the bag in the following way:

>>> bag = pm.as_bag(log, case='CI Name (aff)', time='Actual Start', activity='Change Type')

Does this Trace Conform to the Event Log?

For this task we are going to use a model called Log Skeleton.

>>> bag = pm.as_bag(pm.read_xes(os.path.join("B1.xes")))
>>> traces = [("a1", "a4", "a5", "a7")]
>>> pred = pm.skeleton.classify(bag, traces)
>>> [False]

NOTE: Log B1.xes is equal to the multiset that can be found here.