def test_applyAlphaMinerToCSV(self):
# to avoid static method warnings in tests,
# that by construction of the unittest package have to be expressed in such way
self.dummy_variable = "dummy_value"
# calculate and compare Petri nets obtained on the same log to verify that instances
# are working correctly
log1, net1, marking1, fmarking1 = self.obtainPetriNetThroughAlphaMiner(
os.path.join(INPUT_DATA_DIR, "running-example.csv"))
log2, net2, marking2, fmarking2 = self.obtainPetriNetThroughAlphaMiner(
os.path.join(INPUT_DATA_DIR, "running-example.csv"))
log1 = sorting.sort_timestamp(log1)
log1 = sampling.sample(log1)
log1 = index_attribute.insert_trace_index_as_event_attribute(log1)
log2 = sorting.sort_timestamp(log2)
log2 = sampling.sample(log2)
log2 = index_attribute.insert_trace_index_as_event_attribute(log2)
petri_exporter.export_net(
net1, marking1,
os.path.join(OUTPUT_DATA_DIR, "running-example.pnml"))
os.remove(os.path.join(OUTPUT_DATA_DIR, "running-example.pnml"))
self.assertEqual(len(net1.places), len(net2.places))
self.assertEqual(len(net1.transitions), len(net2.transitions))
self.assertEqual(len(net1.arcs), len(net2.arcs))
final_marking = petri.petrinet.Marking()
for p in net1.places:
if not p.out_arcs:
final_marking[p] = 1
aligned_traces = token_replay.apply_log(log1, net1, marking1,
final_marking)
self.assertEqual(aligned_traces, aligned_traces)
def test_importExportCSVtoCSV(self):
# to avoid static method warnings in tests,
# that by construction of the unittest package have to be expressed in such way
self.dummy_variable = "dummy_value"
event_log = csv_importer.import_event_stream(
os.path.join(INPUT_DATA_DIR, "running-example.csv"))
event_log = sorting.sort_timestamp(event_log)
event_log = sampling.sample(event_log)
event_log = index_attribute.insert_event_index_as_event_attribute(
event_log)
log = log_conv_fact.apply(event_log)
log = sorting.sort_timestamp(log)
log = sampling.sample(log)
log = index_attribute.insert_trace_index_as_event_attribute(log)
event_log_transformed = log_conv_fact.apply(
log, variant=log_conv_fact.TO_EVENT_STREAM)
csv_exporter.export(
event_log_transformed,
os.path.join(OUTPUT_DATA_DIR, "running-example-exported.csv"))
event_log_imported_after_export = csv_importer.import_event_stream(
os.path.join(OUTPUT_DATA_DIR, "running-example-exported.csv"))
log_imported_after_export = log_conv_fact.apply(
event_log_imported_after_export)
self.assertEqual(len(log), len(log_imported_after_export))
os.remove(os.path.join(OUTPUT_DATA_DIR,
"running-example-exported.csv"))
def test_importExportCSVtoCSV(self):
# to avoid static method warnings in tests,
# that by construction of the unittest package have to be expressed in such way
self.dummy_variable = "dummy_value"
df = pd.read_csv(os.path.join(INPUT_DATA_DIR, "running-example.csv"))
df = dataframe_utils.convert_timestamp_columns_in_df(df)
event_log = log_conversion.apply(
df, variant=log_conversion.TO_EVENT_STREAM)
event_log = sorting.sort_timestamp(event_log)
event_log = sampling.sample(event_log)
event_log = index_attribute.insert_event_index_as_event_attribute(
event_log)
log = log_conversion.apply(event_log)
log = sorting.sort_timestamp(log)
log = sampling.sample(log)
log = index_attribute.insert_trace_index_as_event_attribute(log)
event_log_transformed = log_conversion.apply(
log, variant=log_conversion.TO_EVENT_STREAM)
df = log_conversion.apply(event_log_transformed,
variant=log_conversion.TO_DATA_FRAME)
df.to_csv(os.path.join(OUTPUT_DATA_DIR,
"running-example-exported.csv"))
df = pd.read_csv(
os.path.join(OUTPUT_DATA_DIR, "running-example-exported.csv"))
df = dataframe_utils.convert_timestamp_columns_in_df(df)
event_log_imported_after_export = log_conversion.apply(
df, variant=log_conversion.TO_EVENT_STREAM)
log_imported_after_export = log_conversion.apply(
event_log_imported_after_export)
self.assertEqual(len(log), len(log_imported_after_export))
os.remove(os.path.join(OUTPUT_DATA_DIR,
"running-example-exported.csv"))
def test_importExportCSVtoXES(self):
# to avoid static method warnings in tests,
# that by construction of the unittest package have to be expressed in such way
self.dummy_variable = "dummy_value"
event_log = csv_importer.import_event_stream(os.path.join(INPUT_DATA_DIR, "running-example.csv"))
event_log = sorting.sort_timestamp(event_log)
event_log = sampling.sample(event_log)
event_log = index_attribute.insert_event_index_as_event_attribute(event_log)
log = log_transform.transform_event_stream_to_event_log(event_log)
log = sorting.sort_timestamp(log)
log = sampling.sample(log)
log = index_attribute.insert_trace_index_as_event_attribute(log)
xes_exporter.export_log(log, os.path.join(OUTPUT_DATA_DIR, "running-example-exported.xes"))
log_imported_after_export = xes_importer.import_log(
os.path.join(OUTPUT_DATA_DIR, "running-example-exported.xes"))
self.assertEqual(len(log), len(log_imported_after_export))
os.remove(os.path.join(OUTPUT_DATA_DIR, "running-example-exported.xes"))
def select_attributes_from_log_for_tree(log, max_cases_for_attr_selection=DEFAULT_MAX_CASES_FOR_ATTR_SELECTION,
max_diff_occ=DEFAULT_MAX_CASES_FOR_ATTR_SELECTION / 4):
"""
Select attributes from log for tree
Parameters
------------
log
Log
max_cases_for_attr_selection
Maximum number of cases to consider for attribute selection
max_diff_occ
Maximum number of different occurrences
Returns
------------
"""
if len(log) > max_cases_for_attr_selection:
filtered_log = sampling.sample(log, max_cases_for_attr_selection)
else:
filtered_log = log
event_attributes = get_all_event_attributes_from_log(filtered_log)
trace_attributes = get_all_trace_attributes_from_log(filtered_log)
event_attributes_values = {}
trace_attributes_values = {}
for attr in event_attributes:
event_attributes_values[attr] = set(get_attribute_values(log, attr).keys())
for attr in trace_attributes:
trace_attributes_values[attr] = set(get_trace_attribute_values(log, attr).keys())
numeric_event_attributes_to_consider = list()
string_event_attributes_to_consider = list()
numeric_trace_attributes_to_consider = list()
string_trace_attributes_to_consider = list()
for attr in event_attributes_values:
if type(list(event_attributes_values[attr])[0]) is int or type(list(event_attributes_values[attr])[0]) is float:
numeric_event_attributes_to_consider.append(attr)
elif type(list(event_attributes_values[attr])[0]) is str and len(event_attributes_values[attr]) < max_diff_occ:
string_event_attributes_to_consider.append(attr)
for attr in trace_attributes_values:
if type(list(trace_attributes_values[attr])[0]) is int or type(list(trace_attributes_values[attr])[0]) is float:
numeric_trace_attributes_to_consider.append(attr)
elif type(list(trace_attributes_values[attr])[0]) is str and len(trace_attributes_values[attr]) < max_diff_occ:
string_trace_attributes_to_consider.append(attr)
numeric_event_attributes_to_consider = check_event_attributes_presence(log,
numeric_event_attributes_to_consider)
string_event_attributes_to_consider = check_event_attributes_presence(log,
string_event_attributes_to_consider)
numeric_trace_attributes_to_consider = check_trace_attributes_presence(log,
numeric_trace_attributes_to_consider)
string_trace_attributes_to_consider = check_trace_attributes_presence(log,
string_trace_attributes_to_consider)
return string_trace_attributes_to_consider, string_event_attributes_to_consider, numeric_trace_attributes_to_consider, numeric_event_attributes_to_consider
def test_alphaMinerVisualizationFromXES(self):
# to avoid static method warnings in tests,
# that by construction of the unittest package have to be expressed in such way
self.dummy_variable = "dummy_value"
log, net, marking, fmarking = self.obtainPetriNetThroughAlphaMiner(
os.path.join(INPUT_DATA_DIR, "running-example.xes"))
log = sorting.sort_timestamp(log)
log = sampling.sample(log)
log = index_attribute.insert_trace_index_as_event_attribute(log)
petri_exporter.apply(net, marking, os.path.join(OUTPUT_DATA_DIR, "running-example.pnml"))
os.remove(os.path.join(OUTPUT_DATA_DIR, "running-example.pnml"))
gviz = pn_viz.graphviz_visualization(net)
self.assertEqual(gviz, gviz)
final_marking = petri.petrinet.Marking()
for p in net.places:
if not p.out_arcs:
final_marking[p] = 1
aligned_traces = token_replay.apply(log, net, marking, fmarking)
self.assertEqual(aligned_traces, aligned_traces)
def sample_cases(log: Union[EventLog, pd.DataFrame],
num_cases: int) -> Union[EventLog, pd.DataFrame]:
"""
(Random) Sample a given number of cases from the event log.
Parameters
---------------
log
Event log / Pandas dataframe
num_cases
Number of cases to sample
Returns
---------------
sampled_log
Sampled event log (containing the specified amount of cases)
"""
if isinstance(log, EventLog):
from pm4py.objects.log.util import sampling
return sampling.sample(log, num_cases)
elif isinstance(log, pd.DataFrame):
from pm4py.objects.log.util import dataframe_utils
return dataframe_utils.sample_dataframe(
log, parameters={"max_no_cases": num_cases})
from grm import GRM, preprocessing
from pm4py.objects.log.util import sampling
# import data
data_raw = preprocessing.import_data("data",
"BPI2020_PermitLog.csv",
separator=";",
quote='',
case_id="Case ID",
activity="Activity",
time_stamp="Complete Timestamp",
target="(case) Overspent")
# Create new GRM model object
hyper_params = {'num_epochs': 1}
grm_model = GRM.GRM(data_raw, params=hyper_params)
# Train GGNN model
grm_model.train()
# Evaluation of the GGNN model
evaluation_metrics = grm_model.testing_log()
print(evaluation_metrics)
# Visualization as DFG (with sample of evaluation data)
multi_instance_log = sampling.sample(data_raw, n=100)
grm_model.visualize_dfg(save_file=False,
log=multi_instance_log,
file_name="multi")
from grm import preprocessing, GRM
from grm.util import get_activities
from pm4py.algo.filtering.log.attributes import attributes_filter
from pm4py.util import constants
from pm4py.objects.log.util import sampling
model_path = '../best_models/sp2020/2020-05-06-05-40_best_model.pickle'
logfile = "sp2020.csv"
name_of_case_id = "CASE_ID"
name_of_activity = "ACTIVITY"
name_of_timestamp = "TIMESTAMP"
name_of_label = "REPAIR_IN_TIME_5D"
log = preprocessing.import_data("data", logfile, separator=";", quote='"', case_id=name_of_case_id,
activity=name_of_activity,
time_stamp=name_of_timestamp, target=name_of_label)
activities = get_activities(log)
grm_model = GRM.GRM(log, activities, restore_file=model_path)
log = attributes_filter.apply(log, [0],
parameters={constants.PARAMETER_CONSTANT_ATTRIBUTE_KEY: "label", "positive": True})
log = sampling.sample(log, n=5000)
grm_model.visualize_dfg(save_file=True, log=log, file_name="sp2020_", variant="all")
import os
from pm4py.objects.log.adapters.pandas import csv_import_adapter
from pm4py.objects.conversion.log import factory as conversion_factory
dataframe = csv_import_adapter.import_dataframe_from_path(os.path.join(
"pmdata/", "running-example.csv"),
sep=',')
dataframe
dataframe.head()
dataframe.summary()
log = conversion_factory.apply(dataframe)
from pm4py.objects.log.exporter.csv import factory as csv_exporter
csv_exporter.export(event_stream, "data/outputFile1.csv")
#sorting
log
from pm4py.objects.log.util import sorting
log = sorting.sort_timestamp(log)
log
from pm4py.objects.log.util import sorting
sorted_log = sorting.sort_lambda(log,
lambda x: x.attributes["concept:name"],
reverse=False)
sorted_log
#sampling
from pm4py.objects.log.util import sampling
sampled_log = sampling.sample(log, n=50)
sampled_log
#links
#http://www.processmining.org/event_logs_and_models_used_in_book
def run_experiment(data_raw,
hyper_params=None,
k=10,
ml_flow_uri="databricks",
ml_flow_exp="/Shared/grm-review",
ml_flow_run_name_prefix="Experiment",
save_artifact=True):
"""
Performs experiment.
:param data_raw: raw data from event log file.
:param hyper_params: set of hyper-parameters.
:param k: index of k-fold cross-validation.
:param ml_flow_uri: ??
:param ml_flow_exp: ??
:param ml_flow_run_name_prefix: ??
:param save_artifact: set False if >1GB
:return: none.
"""
# init ml flow
mlflow.set_tracking_uri(ml_flow_uri)
mlflow.set_experiment(ml_flow_exp)
# load event log
activities = get_activities(data_raw)
num_activities = len(activities)
with mlflow.start_run(run_name=ml_flow_run_name_prefix + "_" +
str(uuid.uuid1())) as run:
if hyper_params:
for key, value in hyper_params.items():
log_param(key, value)
log_param("k", k)
log_metric("number of activities", num_activities)
results_measures = dict()
i = 0
# Perform k-fold cross-validation
kf = KFold(n_splits=k, shuffle=True)
for train_idx, test_idx in kf.split(data_raw):
i += 1
data_training = [data_raw[j] for j in train_idx]
data_testing = [data_raw[j] for j in test_idx]
with mlflow.start_run(nested=True,
run_name="run_%d" % i) as run_cv:
print("Starting Run " + str(i))
# Create new GGNN model object
grm_model = GRM.GRM(data_training,
activities,
restore_file=None,
params=hyper_params)
# Train GGNN model
grm_model.train()
# Perform evaluation
measures = grm_model.testing_log(data_testing)
for key in measures.keys():
log_metric(key, measures[key], i)
if key in results_measures:
pass
else:
results_measures[key] = []
results_measures[key].append(measures[key])
print(key + " of run " + str(i) + ": " +
str(round(measures[key], 3)))
if save_artifact is True:
log_artifact(grm_model.best_model_file)
log_artifact('../results/cm.pdf')
for key in results_measures.keys():
overall_measure = mean(results_measures[key])
log_metric(key, overall_measure)
print("Overall " + key + ": " + str(overall_measure))
overall_st_dev = stdev(results_measures["accuracy"])
log_metric("st_dev", overall_st_dev)
print("Standard deviation: " + str(overall_st_dev))
""" Relevance visualisation for one instance """
# Extract one random instance from the log
single_instance_log = sampling.sample(data_raw, n=1)
# Visualization as direct follower graph (DFG) with evaluation data
filenames = grm_model.visualize_dfg(save_file=True,
log=single_instance_log,
file_name="single")
for file in filenames:
log_artifact(file)
""" Relevance visualisation for 1000 instances """
# Extract 1000 instances from the event log
multi_instance_log = sampling.sample(data_raw, n=1000)
# Visualization as DFG (with evaluation data)
for file in grm_model.visualize_dfg(save_file=True,
log=multi_instance_log,
file_name="multi"):
log_artifact(file)
