20220722 Event Cullers for the GUI

Added N Occurrences event culler

Constants.py

@@ -1,3 +1,3 @@
 # PyGAAP Constants
 version = "1.0.0 alpha 1"
-versiondate = "2022.07.21"
+versiondate = "2022.07.22"

Developer_Manual.md

@@ -97,7 +97,13 @@ edit_known_authors(.., mode)                            #called when a button in
 
 # Adding a new module <a name="new_mod"></a>
 Each module is a class in or imported to the file containing modules of the same type. These types are canonicizers (```Canonicizer.py```), event drivers (```EventDriver.py```), event cullers (```EventCulling.py```), analysis methods (```AnalysisMethod.py```), and distance functions (```DistanceFunction.py```). If a module is not in one of those py files (i.e. imported into them), it is an "external module". For the purpose of this manual, external modules do not include libraries or files not directly called by the API (e.g. sklearn).
-Add package dependencies and their version numbers to ```./requirements.txt```, if applicable.
+Add package dependencies and their version numbers to ```./requirements.txt```, if applicable.\
+As a readability consideration, it's recommended that the files for external modules be prefixed with the following:\
+```cc``` for canonicizers\
+```ed``` for event drivers\
+```ec``` for event cullers\
+```am``` for analysis methods\
+```df``` for distance functions.
 
 ## <span style="color:#aaeeff">Class variables</span> <a name="class_variables"></a>
 Class variables are declared within the class definition.

PyGAAP.py

@@ -10,7 +10,7 @@ def main():
 		cliMain()
 	else:
 		# import backend.GUI.GUI
-		from .backend.GUI import GUI2
+		from backend.GUI import GUI2
 		app = GUI2.PyGAAP_GUI()
 		app.run()
 

backend/API.py

@@ -133,17 +133,21 @@ def runAnalysis(self, analysisMethodString, distanceFunctionString):
 		analysis.train(knownDocs)
 		return unknownDoc, analysis.analyze(unknownDoc)
 
-	def prettyFormatResults(self, canonicizers, eventDrivers, analysisMethod, distanceFunc, unknownDoc, results):
+	def prettyFormatResults(self, canonicizers, eventDrivers, eventCullers, analysisMethod, distanceFunc, unknownDoc, results):
 		'''Returns a string of the results in a pretty, formatted structure.'''
 		# Build a string the contains general information about the experiment.
 		formattedResults = str(unknownDoc.title) + ' ' + str(unknownDoc.filepath) + "\nCanonicizers:\n"
 		for canonicizer in canonicizers:
 			formattedResults += '\t' + canonicizer + '\n'
 		if type(eventDrivers) == list:
 			for eventDriver in eventDrivers:
-				formattedResults += '\t' + eventDriver + '\n'
+				formattedResults += "Event Drivers:\n\t" + eventDriver + '\n'
 		else:
 			formattedResults += "Event Driver:\n\t" + eventDrivers + '\n'
+
+		for eventCuller in eventCullers:
+			formattedResults += "Event Culler:\n\t" + eventCuller + '\n'
+
 		formattedResults += "Analysis Method:\n\t" + analysisMethod + " with " + distanceFunc + '\n'
 
 		# Sort the dictionary in ascending order by distance values and build the results listing.

backend/CLI.py

@@ -54,8 +54,9 @@ def cliMain():
 			api.runEventDriver(eventDriver)
 
 			# Run the analysis and get the results in a formatted string.
+			# TODO implement event culling for the CLI
 			unknownDoc, results = api.runAnalysis(analysisMethod, distanceFunc)
-			formattedResults = api.prettyFormatResults(canonicizers, eventDriver, analysisMethod, distanceFunc, unknownDoc, results)
+			formattedResults = api.prettyFormatResults(canonicizers, eventDriver, "", analysisMethod, distanceFunc, unknownDoc, results)
 
 			# Create the directories that the results will be stored in.
 			outPath = os.path.join(Path.cwd(), "tmp", '&'.join(canonicizers).replace('|', '_').replace(':', '_'), eventDriver.replace('|', '_').replace(':', '_'), analysisMethod + '-' + distanceFunc)

backend/GUI/GUI_run_experiment.py

@@ -67,9 +67,9 @@ def run_pre_processing(
 			doc.setEventSet(event_set, append=True)
 			#print("evset----------------", doc.eventSet[:15])
 
-		if len(self.backend_API.modulesInUse["EventCulling"]) != 0:
-			raise NotImplementedError
-			#?._global_parameters = self.backend_API.global_parameters
+		for ec in self.backend_API.modulesInUse["EventCulling"]:
+			ec._global_parameters = self.backend_API.global_parameters
+			doc.setEventSet(ec.process(doc.eventSet))
 		return doc
 
 	def run_experiment(self):
@@ -182,6 +182,7 @@ def run_experiment(self):
 				formatted_results = \
 					self.backend_API.prettyFormatResults(self.module_names["canonicizers_names"],
 													self.module_names["event_drivers_names"],
+													self.module_names["event_cullers_names"],
 													self.module_names["am_df_names"][am_df_index][0],
 													self.module_names["am_df_names"][am_df_index][1],
 													d,

generics/EventCulling.py

@@ -19,7 +19,7 @@ def __init__(self):
 			self._variable_options = dict()
 
 	@abstractmethod
-	def process(self, procText):
+	def process(self, eventSet):
 		'''To be determined'''
 		pass
 
@@ -33,20 +33,21 @@ def displayDescription():
 		'''Returns the display description for the event culler.'''
 
 
-class EmptyEventCuller(EventCulling):
-	_variable_options={"test_param1": {"options": list(range(2, 8)), "default": 0, "type": "OptionMenu"},
-					   "test_param2": {"options": list(range(10, 15)), "default": 4, "type": "OptionMenu"},
-					   "test_param3": {"options": ["op1", "op2"], "default": 0, "type": "OptionMenu"}}
+class N_Occurrences(EventCulling):
+	_variable_options = {"Frequency": {"options": list(range(1, 5)), "default": 0, "type": "OptionMenu"}}
 
-	test_param1 = _variable_options["test_param1"]["options"][_variable_options["test_param1"]["default"]]
-	test_param2 = _variable_options["test_param2"]["options"][_variable_options["test_param2"]["default"]]
-	test_param3 = _variable_options["test_param3"]["options"][_variable_options["test_param3"]["default"]]
+	Frequency = _variable_options["Frequency"]["options"][_variable_options["Frequency"]["default"]]
 
-	def process(self, procText):
-		pass
+	def process(self, eventSet: list):
+		freq = dict()
+		for e in eventSet:
+			if freq.get(e) == None: freq[e] = 1
+			else: freq[e] += 1
+		new_events = [ev for ev in eventSet if (freq.get(ev)!=None and freq.get(ev) > 1)]
+		return new_events
 
 	def displayName():
-		return "Test event culler"
+		return "N occurrences"
 
 	def displayDescription():
-		return "An empty event culler class."
+		return "Remove features that are encountered N or fewer times."