changes in benchmark.py and experiment_test.json to include output path and specify event log names

config_files/algorithm/experiment_test.json

@@ -46,6 +46,7 @@
     "pipeline_step": "benchmark_test",
     "benchmark_test": "discovery",
     "input_path":"data/test_2",
+    "output_path":"output",
     "miners" : ["inductive", "heuristics", "imf", "ilp", "sm"]
   }
 ]

tag/benchmark.py

@@ -15,6 +15,7 @@ from pm4py.algo.evaluation.generalization import algorithm as generalization_eva
 from pm4py.algo.evaluation.simplicity import algorithm as simplicity_evaluator
 from pm4py.objects.bpmn.obj import BPMN
 from pm4py.objects.log.importer.xes import importer as xes_importer
+from pm4py.objects.log.obj import EventLog
 from tag.utils.io_helpers import dump_features_json
 from tqdm import tqdm
 from utils.param_keys import INPUT_PATH, OUTPUT_PATH
@@ -44,11 +45,10 @@ class BenchmarkTest:
         if True:
              num_cores = multiprocessing.cpu_count() if len(
                         event_logs) >= multiprocessing.cpu_count() else len(event_logs)
-             #self.benchmark_wrapper(event_logs[0], miners=self.params[MINERS])# TESTING
+            #  self.benchmark_wrapper(event_logs[0], miners=self.params[MINERS])# TESTING
              with multiprocessing.Pool(num_cores) as p:
                  print(f"INFO: Benchmark starting at {start.strftime('%H:%M:%S')} using {num_cores} cores for {len(event_logs)} files...")
-                 p.map(partial(self.benchmark_wrapper, miners = self.params[MINERS])
-                                       , event_logs)
+                 p.map(partial(self.benchmark_wrapper, miners = self.params[MINERS]), event_logs)
 
              # Aggregates metafeatures in saved Jsons into dataframe
              self.root_path = self.params[INPUT_PATH]
@@ -88,7 +88,11 @@ class BenchmarkTest:
 
         benchmark_results = pd.DataFrame()
         # TODO: Use iteratevely generated name for log name in dataframe for passed unnamed logs instead of whole log. E.g. gen_el_1, gen_el_2,...
-        results = {'log': event_log.replace(".xes", "")}
+        if isinstance(event_log, str):
+            results = {'log': event_log.replace(".xes", "")}
+        else:
+            results = {"log": event_log}
+            
         for miner in miners:
             miner_cols = [f"fitness_{miner}", f"precision_{miner}", f"fscore_{miner}", f"size_{miner}", f"cfc_{miner}", f"pnsize_{miner}"]# f"generalization_{miner}",f"simplicity_{miner}"]
             start_miner = dt.now()