'''
This file is part of PM4Py (More Info: https://pm4py.fit.fraunhofer.de).
PM4Py is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
PM4Py is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with PM4Py. If not, see <https://www.gnu.org/licenses/>.
'''
from pm4py.algo.conformance.tokenreplay import algorithm as executor
from pm4py.algo.conformance.tokenreplay.variants import token_replay
from pm4py.evaluation.replay_fitness.parameters import Parameters
from pm4py.util import exec_utils
from pm4py.util.xes_constants import DEFAULT_NAME_KEY
[docs]def evaluate(aligned_traces, parameters=None):
"""
Gets a dictionary expressing fitness in a synthetic way from the list of boolean values
saying if a trace in the log is fit, and the float values of fitness associated to each trace
Parameters
------------
aligned_traces
Result of the token-based replayer
parameters
Possible parameters of the evaluation
Returns
-----------
dictionary
Containing two keys (percFitTraces and averageFitness)
"""
if parameters is None:
parameters = {}
no_traces = len(aligned_traces)
fit_traces = len([x for x in aligned_traces if x["trace_is_fit"]])
sum_of_fitness = sum([x["trace_fitness"] for x in aligned_traces])
perc_fit_traces = 0.0
average_fitness = 0.0
log_fitness = 0
total_m = sum([x["missing_tokens"] for x in aligned_traces])
total_c = sum([x["consumed_tokens"] for x in aligned_traces])
total_r = sum([x["remaining_tokens"] for x in aligned_traces])
total_p = sum([x["produced_tokens"] for x in aligned_traces])
if no_traces > 0:
perc_fit_traces = float(100.0 * fit_traces) / float(no_traces)
average_fitness = float(sum_of_fitness) / float(no_traces)
if total_c > 0 and total_p > 0:
log_fitness = 0.5 * (1 - total_m / total_c) + 0.5 * (1 - total_r / total_p)
return {"perc_fit_traces": perc_fit_traces, "average_trace_fitness": average_fitness, "log_fitness": log_fitness,
"percentage_of_fitting_traces": perc_fit_traces }
[docs]def apply(log, petri_net, initial_marking, final_marking, parameters=None):
"""
Apply token replay fitness evaluation
Parameters
-----------
log
Trace log
petri_net
Petri net
initial_marking
Initial marking
final_marking
Final marking
parameters
Parameters
Returns
-----------
dictionary
Containing two keys (percFitTraces and averageFitness)
"""
if parameters is None:
parameters = {}
activity_key = exec_utils.get_param_value(Parameters.ACTIVITY_KEY, parameters, DEFAULT_NAME_KEY)
token_replay_variant = exec_utils.get_param_value(Parameters.TOKEN_REPLAY_VARIANT, parameters,
executor.Variants.TOKEN_REPLAY)
cleaning_token_flood = exec_utils.get_param_value(Parameters.CLEANING_TOKEN_FLOOD, parameters, False)
remaining_in_fitness = exec_utils.get_param_value(token_replay.Parameters.CONSIDER_REMAINING_IN_FITNESS, parameters, True)
parameters_tr = {token_replay.Parameters.ACTIVITY_KEY: activity_key,
token_replay.Parameters.CONSIDER_REMAINING_IN_FITNESS: remaining_in_fitness,
token_replay.Parameters.CLEANING_TOKEN_FLOOD: cleaning_token_flood}
aligned_traces = executor.apply(log, petri_net, initial_marking, final_marking, variant=token_replay_variant,
parameters=parameters_tr)
return evaluate(aligned_traces)