add bonus

app.py

@@ -299,93 +299,24 @@ def display_predictions():
         else:
             st.write("No predictions for today's matches yet.")
 
-def redistribute_lost_points(match_id) -> dict:
-    # Load already processed matches
-    done_matches = load_bonus(BONUS_JSON)
-
-    if match_id in done_matches:
-        return {}
-
-    users = load_users(USERS_JSON)
-    predictions = []
-
-    for file in PREDICTIONS_FOLDER.glob(f"prediction_{match_id}_*.json"):
-        with open(file, "r") as f:
-            for line in f:
-                predictions.append(json.loads(line.strip()))
-
-    outcomes = load_data(OUTCOMES_JSON)
-    outcome = next((m for m in outcomes if m["match_id"] == match_id), None)
-    if not outcome:
-        st.error("Match outcome not found.")
-        return {}
-
-    correct_winner = outcome["winner"]
-    correct_motm = outcome["man_of_the_match"]
-
-    user_losses = {}
-    for pred in predictions:
-        user = pred["user_name"]
-        bid = pred["bid_points"]
-        if (pred["predicted_winner"] != correct_winner) or (pred["predicted_motm"] != correct_motm):
-            user_losses[user] = user_losses.get(user, 0) + bid
-
-    top_5_users = sorted(users.items(), key=lambda x: x[1]["points"], reverse=True)[:5]
-    top_5_usernames = [user for user, _ in top_5_users]
-
-    lost_points_by_top5 = sum(user_losses.get(user, 0) for user in top_5_usernames)
-    if lost_points_by_top5 == 0:
-        return {}
-
-    rest_users = [user for user in users if user not in top_5_usernames]
-    total_points_rest_users = sum(users[u]["points"] for u in rest_users if users[u]["points"] > 0)
-
-    bonus_map = {}
-    for user in rest_users:
-        user_points = users[user]["points"]
-        if user_points <= 0:
-            continue
-        share_ratio = user_points / total_points_rest_users
-        bonus = round(share_ratio * lost_points_by_top5)
-        users[user]["points"] += bonus
-        bonus_map[user] = bonus
-
-    with open(USERS_JSON, "w") as f:
-        json.dump(users, f, indent=2)
-
-    # Record this match as done
-    done_matches.append(match_id)
-    with open(BONUS_JSON, "w") as f:
-        json.dump(done_matches, f, indent=2)
-
-    return bonus_map
 
 def display_leaderboard():
     if st.button("Show Leaderboard"):
         try:
             # # Load the 'leaders' configuration
             dataset = load_dataset("Jay-Rajput/DIS_IPL_Leads", split='train')
-            
-            # Load outcomes data from HF instead of local file
-            # outcome_dataset = load_dataset("Jay-Rajput/DIS_IPL_Outcomes", split='train')
-            # latest_match_id = outcome_dataset[-1]["match_id"] if outcome_dataset else None
-
-            # # Redistribute points only once per match
-            # bonus_map = {}
-            # if latest_match_id:
-            #     bonus_map = redistribute_lost_points(latest_match_id)
 
             users_data = []
             if dataset:
                 for user, points_dict in dataset[0].items():
                     points = points_dict.get("points", 0)
                     last_5_results = " ".join(points_dict.get("last_5_results", ["⚪"] * 5))  # Default: 5 white circles
-                    # bonus = bonus_map.get(user, 0)
+                    bonus = points_dict.get("redistributed_bonus", 0)
                     users_data.append({
                         'User': user,
                         'Points': points,
+                        'Redistribution Bonus': bonus,
                         'Last 5 Bids': last_5_results
-                        # 'Redistribution Bonus': bonus
                     })
             else:
                 st.warning("No leaderboard data found.")
@@ -399,7 +330,7 @@ def display_leaderboard():
             leaderboard['Rank'] = range(1, len(leaderboard) + 1)
             
             # Select and order the columns for display
-            leaderboard = leaderboard[['Rank', 'User', 'Points', 'Last 5 Bids']]
+            leaderboard = leaderboard[['Rank', 'User', 'Points', 'Redistribution Bonus', 'Last 5 Bids']]
             
             st.dataframe(leaderboard, hide_index=True)
         except Exception as e:
@@ -498,6 +429,68 @@ def fetch_latest_predictions(match_id):
         return pd.DataFrame()
 
 
+def redistribute_lost_points(match_id):
+    predictions = fetch_latest_predictions(match_id)
+    users = load_dataset("Jay-Rajput/DIS_IPL_Leads", split="train")
+    users_df = pd.DataFrame(users)
+
+    # Build current leaderboard (after score updates)
+    leaderboard = []
+    for user_name in users_df.columns:
+        points = users_df[user_name][0]['points']
+        leaderboard.append((user_name, points))
+
+    leaderboard.sort(key=lambda x: x[1], reverse=True)
+
+    top_5 = leaderboard[:5]
+    others = leaderboard[5:]
+
+    # Fetch match outcome
+    outcomes_df = load_dataset("Jay-Rajput/DIS_IPL_Outcomes", split="train").to_pandas()
+    match_row = outcomes_df[outcomes_df['match_id'] == match_id].iloc[0]
+    winning_team = match_row['winning_team']
+
+    # Calculate lost points from top 5 users who predicted incorrectly
+    total_lost_points = 0
+    lost_points_per_user = {}
+
+    for user_name, _ in top_5:
+        if user_name in predictions['user_name'].values:
+            pred = predictions[predictions['user_name'] == user_name].iloc[0]
+            if pred['predicted_winner'] != winning_team:
+                lost_points = 200 + pred['bid_points']
+                total_lost_points += lost_points
+                lost_points_per_user[user_name] = lost_points
+
+    if total_lost_points == 0 or not others:
+        return  # Nothing to redistribute
+
+    # Total points of eligible users (position 6 to last)
+    total_eligible_points = sum([points for (_, points) in others])
+    if total_eligible_points == 0:
+        return
+
+    # Distribute lost points proportionally
+    for user_name, user_points in others:
+        share_ratio = user_points / total_eligible_points
+        bonus = int(total_lost_points * share_ratio)
+
+        # Update bonus in leads
+        users_df[user_name][0]['points'] += bonus
+        users[user_name][0]['points'] = users_df[user_name][0]['points']
+
+        # Track redistributed bonus (initialize or accumulate)
+        prev_bonus_df = users_df[user_name][0].get("redistributed_bonus", 0)
+        prev_bonus_dict = users[user_name][0].get("redistributed_bonus", 0)
+        users_df[user_name][0]["redistributed_bonus"] = prev_bonus_df + bonus
+        users[user_name][0]["redistributed_bonus"] = prev_bonus_dict + bonus
+
+    # Push updated dataset
+    users.to_json(USERS_JSON)
+    updated_dataset = Dataset.from_pandas(users_df)
+    updated_dataset.push_to_hub("Jay-Rajput/DIS_IPL_Leads", split="train")
+
+
 def update_leaderboard_and_outcomes(match_id, winning_team, man_of_the_match, outcome_only=False):
 
     # Load existing match outcomes
@@ -570,6 +563,7 @@ def update_leaderboard_and_outcomes(match_id, winning_team, man_of_the_match, ou
         users.to_json(USERS_JSON)
         updated_dataset = Dataset.from_pandas(users_df)
         updated_dataset.push_to_hub("Jay-Rajput/DIS_IPL_Leads", split="train")
+        redistribute_lost_points(match_id)
 
     outcomes.to_json(OUTCOMES)
     outcomes.push_to_hub("Jay-Rajput/DIS_IPL_Outcomes", split="train")