wip experiment with real targets

config_files/algorithm/experiment_real_targets.json

@@ -0,0 +1,41 @@
+[
+   {
+    "pipeline_step": "event_logs_generation",
+    "output_path": "output",
+    "generator_params": {
+      "experiment": {
+        "input_path": "data/BaselineED_feat.csv",
+        "objectives":["ratio_variants_per_number_of_traces","ratio_most_common_variant","ratio_top_10_variants","epa_normalized_variant_entropy","epa_normalized_sequence_entropy","epa_normalized_sequence_entropy_linear_forgetting","epa_normalized_sequence_entropy_exponential_forgetting"]},
+        "config_space": {
+        "mode": [5, 20],
+        "sequence": [0.01, 1],
+        "choice": [0.01, 1],
+        "parallel": [0.01, 1],
+        "loop": [0.01, 1],
+        "silent": [0.01, 1],
+        "lt_dependency": [0.01, 1],
+        "num_traces": [10, 10001],
+        "duplicate": [0],
+        "or": [0]
+      },
+      "n_trials": 200,
+      "plot_reference_feature": ""
+    }
+  },
+  {
+    "pipeline_step": "feature_extraction",
+    "input_path": "output/generated/BaselineED_feat/7_ense_enseef_enself_enve_rmcv_rt10v_rutpt/",
+    "input_path": "output/features/BaselineED_feat/7_ense_enseef_enself_enve_rmcv_rt10v_rutpt/",
+    "feature_params": {"feature_set":["ratio_variants_per_number_of_traces","ratio_most_common_variant","ratio_top_10_variants","epa_normalized_variant_entropy","epa_normalized_sequence_entropy","epa_normalized_sequence_entropy_linear_forgetting","epa_normalized_sequence_entropy_exponential_forgetting"]},
+    "output_path": "output/plots",
+    "real_eventlog_path": "data/BaselineED_feat.csv",
+    "plot_type": "boxplot"
+  },
+  {
+    "pipeline_step": "benchmark_test",
+    "benchmark_test": "discovery",
+    "input_path": "output/generated/BaselineED_feat/7_ense_enseef_enself_enve_rmcv_rt10v_rutpt/",
+    "output_path":"output",
+    "miners" : ["heu", "imf", "ilp"]
+  }
+]

gedi/generator.py

@@ -91,10 +91,14 @@ class GenerateEventLogs():
 
         self.params = params.get(GENERATOR_PARAMS)
         experiment = self.params.get(EXPERIMENT)
+
         if experiment!= None:
             tasks, output_path = get_tasks(experiment, self.output_path)
             self.output_path = output_path
 
+        if 'ratio_variants_per_number_of_traces' in tasks.columns:#HOTFIX
+            tasks=tasks.rename(columns={"ratio_variants_per_number_of_traces": "ratio_unique_traces_per_trace"})
+
         if tasks is not None:
             num_cores = multiprocessing.cpu_count() if len(tasks) >= multiprocessing.cpu_count() else len(tasks)
             #self.generator_wrapper([*tasks.iterrows()][0])# For testing
@@ -111,6 +115,10 @@ class GenerateEventLogs():
                 self.configs = [self.configs]
             temp = self.generate_optimized_log(self.configs[0])
             self.log_config = [temp]
+            #TODO: Replace hotfix
+            if self.params[EXPERIMENT].get('ratio_unique_traces_per_trace'):#HOTFIX
+                self.params[EXPERIMENT]['ratio_variants_per_number_of_traces']=self.params[EXPERIMENT].pop('ratio_unique_traces_per_trace')
+
             save_path = get_output_key_value_location(self.params[EXPERIMENT],
                                              self.output_path, "genEL")+".xes"
             write_xes(temp['log'], save_path)
@@ -136,6 +144,10 @@ class GenerateEventLogs():
             log_config = self.generate_optimized_log(self.configs)
 
         identifier = 'genEL'+str(identifier)
+        #TODO: Replace hotfix
+        if self.objectives.get('ratio_unique_traces_per_trace'):#HOTFIX
+            self.objectives['ratio_variants_per_number_of_traces']=self.objectives.pop('ratio_unique_traces_per_trace')
+
         save_path = get_output_key_value_location(self.objectives,
                                          self.output_path, identifier)+".xes"