Enhance seed frame initialization with player validation

Updated the seed frame initialization functions for DraftKings and FanDuel to include validation against a set of valid players. This ensures that only players present in the Range of Outcomes collection are considered, improving data integrity. Additionally, refactored name mapping to use apply for better performance and clarity.

app.py

@@ -59,6 +59,11 @@ def init_DK_seed_frames(load_size):
         cursor = collection.find()
         raw_data = pd.DataFrame(list(cursor))
         names_dict = dict(zip(raw_data['key'], raw_data['value']))
+
+        # Get the valid players from the Range of Outcomes collection
+        collection = db["Player_Range_Of_Outcomes"]
+        cursor = collection.find({"site": "Draftkings", "slate": "Main Slate"})
+        valid_players = set(pd.DataFrame(list(cursor))['Name'].unique())
     
         collection = db["DK_NBA_seed_frame"] 
         cursor = collection.find().limit(load_size)
@@ -66,9 +71,13 @@ def init_DK_seed_frames(load_size):
         raw_display = pd.DataFrame(list(cursor))
         raw_display = raw_display[['PG', 'SG', 'SF', 'PF', 'C', 'G', 'F', 'FLEX', 'salary', 'proj', 'Team', 'Team_count', 'Secondary', 'Secondary_count', 'Own']]
         dict_columns = ['PG', 'SG', 'SF', 'PF', 'C', 'G', 'F', 'FLEX']
-        for col in dict_columns:
-            raw_display[col] = raw_display[col].map(names_dict)
-
+        # Map names
+        raw_display[dict_columns] = raw_display[dict_columns].apply(lambda x: x.map(names_dict))
+        
+        # Validate lineups against valid players
+        raw_display = validate_lineup_players(raw_display, valid_players, dict_columns)
+        
+        # Remove any remaining NaN values
         raw_display = raw_display.dropna()
         DK_seed = raw_display.to_numpy()
 
@@ -81,6 +90,11 @@ def init_DK_secondary_seed_frames(load_size):
         cursor = collection.find()
         raw_data = pd.DataFrame(list(cursor))
         names_dict = dict(zip(raw_data['key'], raw_data['value']))
+
+        # Get the valid players from the Range of Outcomes collection
+        collection = db["Player_Range_Of_Outcomes"]
+        cursor = collection.find({"site": "Draftkings", "slate": "Secondary Slate"})
+        valid_players = set(pd.DataFrame(list(cursor))['Name'].unique())
     
         collection = db["DK_NBA_Secondary_seed_frame"] 
         cursor = collection.find().limit(load_size)
@@ -88,9 +102,13 @@ def init_DK_secondary_seed_frames(load_size):
         raw_display = pd.DataFrame(list(cursor))
         raw_display = raw_display[['PG', 'SG', 'SF', 'PF', 'C', 'G', 'F', 'FLEX', 'salary', 'proj', 'Team', 'Team_count', 'Secondary', 'Secondary_count', 'Own']]
         dict_columns = ['PG', 'SG', 'SF', 'PF', 'C', 'G', 'F', 'FLEX']
-        for col in dict_columns:
-            raw_display[col] = raw_display[col].map(names_dict)
-
+        # Map names
+        raw_display[dict_columns] = raw_display[dict_columns].apply(lambda x: x.map(names_dict))
+        
+        # Validate lineups against valid players
+        raw_display = validate_lineup_players(raw_display, valid_players, dict_columns)
+        
+        # Remove any remaining NaN values
         raw_display = raw_display.dropna()
         DK_seed = raw_display.to_numpy()
 
@@ -103,6 +121,11 @@ def init_FD_seed_frames(load_size):
         cursor = collection.find()
         raw_data = pd.DataFrame(list(cursor))
         names_dict = dict(zip(raw_data['key'], raw_data['value']))
+
+        # Get the valid players from the Range of Outcomes collection
+        collection = db["Player_Range_Of_Outcomes"]
+        cursor = collection.find({"site": "Fanduel", "slate": "Main Slate"})
+        valid_players = set(pd.DataFrame(list(cursor))['Name'].unique())
     
         collection = db["FD_NBA_seed_frame"] 
         cursor = collection.find().limit(load_size)
@@ -110,9 +133,13 @@ def init_FD_seed_frames(load_size):
         raw_display = pd.DataFrame(list(cursor))
         raw_display = raw_display[['PG1', 'PG2', 'SG1', 'SG2', 'SF1', 'SF2', 'PF1', 'PF2', 'C1', 'salary', 'proj', 'Team', 'Team_count', 'Secondary', 'Secondary_count', 'Own']]
         dict_columns = ['PG1', 'PG2', 'SG1', 'SG2', 'SF1', 'SF2', 'PF1', 'PF2', 'C1']
-        for col in dict_columns:
-            raw_display[col] = raw_display[col].map(names_dict)
-            
+        # Map names
+        raw_display[dict_columns] = raw_display[dict_columns].apply(lambda x: x.map(names_dict))
+        
+        # Validate lineups against valid players
+        raw_display = validate_lineup_players(raw_display, valid_players, dict_columns)
+        
+        # Remove any remaining NaN values
         raw_display = raw_display.dropna()
         FD_seed = raw_display.to_numpy()
 
@@ -125,6 +152,11 @@ def init_FD_secondary_seed_frames(load_size):
         cursor = collection.find()
         raw_data = pd.DataFrame(list(cursor))
         names_dict = dict(zip(raw_data['key'], raw_data['value']))
+
+        # Get the valid players from the Range of Outcomes collection
+        collection = db["Player_Range_Of_Outcomes"]
+        cursor = collection.find({"site": "Fanduel", "slate": "Secondary Slate"})
+        valid_players = set(pd.DataFrame(list(cursor))['Name'].unique())
     
         collection = db["FD_NBA_Secondary_seed_frame"] 
         cursor = collection.find().limit(load_size)
@@ -132,9 +164,13 @@ def init_FD_secondary_seed_frames(load_size):
         raw_display = pd.DataFrame(list(cursor))
         raw_display = raw_display[['PG1', 'PG2', 'SG1', 'SG2', 'SF1', 'SF2', 'PF1', 'PF2', 'C1', 'salary', 'proj', 'Team', 'Team_count', 'Secondary', 'Secondary_count', 'Own']]
         dict_columns = ['PG1', 'PG2', 'SG1', 'SG2', 'SF1', 'SF2', 'PF1', 'PF2', 'C1']
-        for col in dict_columns:
-            raw_display[col] = raw_display[col].map(names_dict)
-
+        # Map names
+        raw_display[dict_columns] = raw_display[dict_columns].apply(lambda x: x.map(names_dict))
+        
+        # Validate lineups against valid players
+        raw_display = validate_lineup_players(raw_display, valid_players, dict_columns)
+        
+        # Remove any remaining NaN values
         raw_display = raw_display.dropna()
         FD_seed = raw_display.to_numpy()
 
@@ -328,15 +364,11 @@ with tab1:
                 if sim_slate_var1 == 'Main Slate':
                     st.session_state.working_seed = init_DK_seed_frames(sharp_split)
                     dk_id_dict = dict(zip(dk_raw.Player, dk_raw.player_id))
-
-                    st.session_state.dk_raw = validate_lineup_players(dk_raw, set(st.session_state.working_seed.Player), ['PG', 'SG', 'SF', 'PF', 'C', 'G', 'F', 'FLEX'])
                     raw_baselines = dk_raw
                     column_names = dk_columns
                 elif sim_slate_var1 == 'Secondary Slate':
                     st.session_state.working_seed = init_DK_secondary_seed_frames(sharp_split)
                     dk_id_dict = dict(zip(dk_secondary.Player, dk_secondary.player_id))
-
-                    st.session_state.dk_secondary = validate_lineup_players(dk_secondary, set(st.session_state.working_seed.Player), ['PG', 'SG', 'SF', 'PF', 'C', 'G', 'F', 'FLEX'])
                     raw_baselines = dk_secondary
                     column_names = dk_columns
 
@@ -344,15 +376,11 @@ with tab1:
                 if sim_slate_var1 == 'Main Slate':
                     st.session_state.working_seed = init_FD_seed_frames(sharp_split)
                     fd_id_dict = dict(zip(fd_raw.Player, fd_raw.player_id))
-
-                    st.session_state.fd_raw = validate_lineup_players(fd_raw, set(st.session_state.working_seed.Player), ['PG1', 'PG2', 'SG1', 'SG2', 'SF1', 'SF2', 'PF1', 'PF2', 'C1'])
                     raw_baselines = fd_raw
                     column_names = fd_columns
                 elif sim_slate_var1 == 'Secondary Slate':
                     st.session_state.working_seed = init_FD_secondary_seed_frames(sharp_split)
                     fd_id_dict = dict(zip(fd_secondary.Player, fd_secondary.player_id))
-
-                    st.session_state.fd_secondary = validate_lineup_players(fd_secondary, set(st.session_state.working_seed.Player), ['PG1', 'PG2', 'SG1', 'SG2', 'SF1', 'SF2', 'PF1', 'PF2', 'C1'])
                     raw_baselines = fd_secondary
                     column_names = fd_columns
 
@@ -694,14 +722,12 @@ with tab2:
                         st.session_state.working_seed = init_DK_seed_frames(sharp_split_var)
                         dk_id_dict = dict(zip(dk_raw.Player, dk_raw.player_id))
 
-                        st.session_state.dk_raw = validate_lineup_players(dk_raw, set(st.session_state.working_seed.Player), ['PG', 'SG', 'SF', 'PF', 'C', 'G', 'F', 'FLEX'])
                         raw_baselines = dk_raw
                         column_names = dk_columns
                     elif slate_var1 == 'Secondary Slate':
                         st.session_state.working_seed = init_DK_secondary_seed_frames(sharp_split_var)
                         dk_id_dict = dict(zip(dk_secondary.Player, dk_secondary.player_id))
 
-                        st.session_state.dk_secondary = validate_lineup_players(dk_secondary, set(st.session_state.working_seed.Player), ['PG', 'SG', 'SF', 'PF', 'C', 'G', 'F', 'FLEX'])
                         raw_baselines = dk_secondary
                         column_names = dk_columns
                 
@@ -725,14 +751,12 @@ with tab2:
                         st.session_state.working_seed = init_FD_seed_frames(sharp_split_var)
                         fd_id_dict = dict(zip(fd_raw.Player, fd_raw.player_id))
 
-                        st.session_state.fd_raw = validate_lineup_players(fd_raw, set(st.session_state.working_seed.Player), ['PG1', 'PG2', 'SG1', 'SG2', 'SF1', 'SF2', 'PF1', 'PF2', 'C1'])
                         raw_baselines = fd_raw
                         column_names = fd_columns
                     elif slate_var1 == 'Secondary Slate':
                         st.session_state.working_seed = init_FD_secondary_seed_frames(sharp_split_var)
                         fd_id_dict = dict(zip(fd_secondary.Player, fd_secondary.player_id))
-
-                        st.session_state.fd_secondary = validate_lineup_players(fd_secondary, set(st.session_state.working_seed.Player), ['PG1', 'PG2', 'SG1', 'SG2', 'SF1', 'SF2', 'PF1', 'PF2', 'C1'])
+                    
                         raw_baselines = fd_secondary
                         column_names = fd_columns