mtt_haum/code/13_pm-case-clusters.py

# 12_pm-case-clusters.py
#
# content: (1) Load data and create event log
#          (2) Infos for clusters
#          (3) Process maps for clusters
#
# input:  results/haum/eventlogs_pre-corona_item-clusters.csv
# output: results/haum/pn_infos_clusters.csv
#
# last mod: 2024-03-10

import pm4py
import pandas as pd

from python_helpers import eval_pm, pn_infos

#--------------- (1) Load data and create event logs ---------------

dat = pd.read_csv("results/haum/eventlogs_2019_case-clusters_new.csv", sep = ";")

event_log = pm4py.format_dataframe(dat, case_id = "case", activity_key = "event_new",
                                  timestamp_key = "date.start")

#--------------- (2) Infos for clusters ---------------

# Merge clusters into data frame
eval = pd.DataFrame(columns = ["fitness", "precision", "generalizability",
                               "simplicity", "sound", "narcs", "ntrans",
                               "nplaces", "nvariants", "mostfreq"])
for cluster in event_log.cluster.unique().tolist():
    eval = pd.concat([eval, pn_infos(event_log, "cluster", cluster)])
eval = eval.sort_index()

eval.to_csv("results/haum/pn_infos_clusters.csv", sep = ";")

#--------------- (3) Process maps for clusters ---------------

for cluster in event_log.cluster.unique().tolist():
    subdata = event_log[event_log.cluster == cluster]
    subnet, subim, subfm = pm4py.discover_petri_net_inductive(subdata, noise_threshold = .7)
    pm4py.save_vis_petri_net(subnet, subim, subfm,
       "results/processmaps/petrinet_cluster" + str(cluster) + "_cases.png")
    bpmn = pm4py.convert.convert_to_bpmn(subnet, subim, subfm)
    pm4py.vis.save_vis_bpmn(bpmn, "results/processmaps/bpmn_cluster" +
            str(cluster) + "_cases.png")