Trying to make the code faster

workshops.py

@@ -43,6 +43,13 @@ class Schedule:
     self.timeslots[time].remove(person)
 
 
+  def print(self): 
+      print(f"# timeslots filled: {self.num_timeslots_filled}")
+      print(f"# workshops: {self.total_num_workshops}")
+      for time,instructors in self.timeslots.items(): 
+          print(f"{time}: {', '.join(instructors)}")
+
+
 # Returns True if the person can teach during the slot, and False otherwise
 def can_teach(person: str, slot: list, capacity: int) -> bool: 
     if len(slot) == capacity or len(slot) > capacity: 
@@ -103,10 +110,13 @@ def initialize_timeslots(df) -> dict:
 
 
 # Recursive function that generates all possible schedules
-def find_all_schedules(people: list, availability: dict, schedule_obj: Schedule, capacity: int, schedules: list, max_list: list) -> None: 
-    if schedule_obj.num_timeslots_filled > max_list[0] or schedule_obj.num_timeslots_filled == max_list[0]: 
+def find_all_schedules(people: list, availability: dict, schedule_obj: Schedule, capacity: int, schedules: list, max_timeslots_list: list, max_workshops_list: list) -> None: 
+    if schedule_obj.num_timeslots_filled > max_timeslots_list[0] or schedule_obj.num_timeslots_filled == max_timeslots_list[0]: 
         schedules.append(copy.deepcopy(schedule_obj))
-        max_list[0] = schedule_obj.num_timeslots_filled
+        max_timeslots_list[0] = schedule_obj.num_timeslots_filled
+        # Keep track of total number of workshops taught
+        if schedule_obj.total_num_workshops > max_workshops_list[0] or schedule_obj.total_num_workshops == max_workshops_list[0]: 
+            max_workshops_list[0] = schedule_obj.total_num_workshops
     
     # Base case
     if len(people) == 0: 
@@ -123,21 +133,20 @@ def find_all_schedules(people: list, availability: dict, schedule_obj: Schedule,
 
             # Explore (assign everyone else to timeslots based on that decision)
             if len(people) == 1: 
-                find_all_schedules([], availability, schedule_obj, capacity, schedules, max_list)
+                find_all_schedules([], availability, schedule_obj, capacity, schedules, max_timeslots_list, max_workshops_list)
 
             else: 
-                find_all_schedules(people[1:len(people)], availability, schedule_obj, capacity, schedules, max_list)
+                find_all_schedules(people[1:len(people)], availability, schedule_obj, capacity, schedules, max_timeslots_list, max_workshops_list)
 
             # Unchoose (remove that person from the timeslot)
             schedule_obj.remove(person, time)
         # NOTE: this will not generate a full timeslot, but could still lead to a good schedule
         else: 
             if len(people) == 1: 
-                find_all_schedules([], availability, schedule_obj, capacity, schedules, max_list)
+                find_all_schedules([], availability, schedule_obj, capacity, schedules, max_timeslots_list, max_workshops_list)
             else: 
-                find_all_schedules(people[1:len(people)], availability, schedule_obj, capacity, schedules, max_list)
-            
-
+                find_all_schedules(people[1:len(people)], availability, schedule_obj, capacity, schedules, max_timeslots_list, max_workshops_list)
+    
     return
 
 
@@ -256,8 +265,10 @@ def classify_schedules(people: list, schedules: list, partial_names: list, total
     
     all_names = pref_dict.keys() 
     
-    # Evaluate each schedule
-    overall_max = 0
+    
+    ## Evaluate each schedule ##
+    overall_max = 0 # changes throughout the function 
+
     for sched in schedules: 
         if sched.num_timeslots_filled != max_timeslots_filled:
             continue
@@ -307,27 +318,25 @@ def classify_schedules(people: list, schedules: list, partial_names: list, total
 
 
 # Parameters: schedules that have the max number of timeslots filled 
+# Max number of workshops taught in filled timeslots
 # Returns: a list of all schedules that have the max number of workshops
     # To make it less overwhelming, it will return {cutoff} randomly 
-def get_best_schedules(schedules: list, cutoff: str) -> list: 
+def get_best_schedules(schedules: list, cutoff: str, max_workshops: int) -> list: 
     cutoff = int(cutoff)
-    overall_max = 0 
-    best_schedules = {}
+    best_schedules = []
     for sched in schedules: 
-        if sched.total_num_workshops not in best_schedules: 
-            best_schedules[sched.total_num_workshops] = []
-        best_schedules[sched.total_num_workshops].append(sched.timeslots)
-        if sched.total_num_workshops > overall_max: 
-            overall_max = sched.total_num_workshops
-    all_best_schedules = best_schedules[overall_max]
+        if sched.total_num_workshops != max_workshops: 
+            continue 
+        best_schedules.append(sched.timeslots)
+
     if cutoff == -1: 
-        return all_best_schedules
+        return best_schedules
     else: 
-        if len(all_best_schedules) > cutoff: 
+        if len(best_schedules) > cutoff: 
             # Sample without replacement
-            return random.sample(all_best_schedules, cutoff)
+            return random.sample(best_schedules, cutoff)
         else:
-            return all_best_schedules
+            return best_schedules
 
 
 # Big wrapper function that calls the other functions
@@ -338,6 +347,7 @@ def main(df, capacity:int, num_results: int, og_slots: list):
     res = convert_df(df)
     people = res[0]
     availability = res[1]
+    print(availability)
 
     partial_names = []
 
@@ -345,21 +355,22 @@ def main(df, capacity:int, num_results: int, og_slots: list):
 
     schedules = []
     schedule_obj = Schedule(timeslots)
-    max_list = [0]
+    max_timeslots_list = [0]
+    max_workshops_list = [0]
  
-    find_all_schedules(people, availability, schedule_obj, capacity, schedules, max_list)
+    find_all_schedules(people, availability, schedule_obj, capacity, schedules, max_timeslots_list, max_workshops_list)
 
     total_timeslots = len(timeslots)
 
 
-    res = classify_schedules(people, schedules, partial_names, total_timeslots, max_list[0])
+    res = classify_schedules(people, schedules, partial_names, total_timeslots, max_timeslots_list[0])
     valid_schedules = res[0]
     decent_schedules = res[1]
 
     
     # Return schedules
     if len(valid_schedules) > 0: 
-        best_schedules = get_best_schedules(valid_schedules, num_results)
+        best_schedules = get_best_schedules(valid_schedules, num_results, max_workshops_list)
         res = make_df(best_schedules, descrip_dict, og_slots)
         new_df = res[0]
         count = res[1]
@@ -369,11 +380,11 @@ def main(df, capacity:int, num_results: int, og_slots: list):
             results = "Good news! I was able to make multiple complete schedules."
         
     else:
-        best_schedules = get_best_schedules(decent_schedules, num_results)
+        best_schedules = get_best_schedules(decent_schedules, num_results, max_workshops_list)
         res = make_df(best_schedules, descrip_dict, og_slots)
         new_df = res[0]
         count = res[1]
-        beginning = "I wasn't able to make a complete schedule, but here"
+        beginning = "Here"
         if count == 1: 
             results = f"{beginning} is the best option."
         else: 
@@ -387,7 +398,6 @@ def main(df, capacity:int, num_results: int, og_slots: list):
     new_df.to_csv(path, index=False)
     return results, new_df, path
 
-