Update app.py

app.py

@@ -33,7 +33,7 @@ def extract_agent_info(filename: str):
     return agent_type, model_name
 
 def get_available_games() -> List[str]:
-    """Extracts all unique game names from all SQLite databases."""
+    """Extracts all unique game names from all SQLite databases and includes 'Total Performance'."""
     db_files = find_or_download_db()
     game_names = set()
     
@@ -48,7 +48,9 @@ def get_available_games() -> List[str]:
         finally:
             conn.close()
     
-    return sorted(game_names) if game_names else ["No Games Found"]
+    game_list = sorted(game_names) if game_names else ["No Games Found"]
+    game_list.insert(0, "Total Performance")  # Ensure 'Total Performance' is always first
+    return game_list
 
 def extract_leaderboard_stats(game_name: str) -> pd.DataFrame:
     """Extract and aggregate leaderboard stats from all SQLite databases."""
@@ -59,23 +61,48 @@ def extract_leaderboard_stats(game_name: str) -> pd.DataFrame:
         conn = sqlite3.connect(db_file)
         agent_type, model_name = extract_agent_info(db_file)
         
-        query = "SELECT COUNT(DISTINCT episode) AS games_played, " \
-                "AVG(generation_time) AS avg_gen_time, SUM(reward) AS total_rewards " \
-                "FROM game_results WHERE game_name = ?"
+        if game_name == "Total Performance":
+            query = "SELECT game_name, COUNT(DISTINCT episode) AS games_played, " \
+                    "AVG(generation_time) AS avg_gen_time, SUM(reward) AS total_rewards " \
+                    "FROM game_results GROUP BY game_name"
+            df = pd.read_sql_query(query, conn)
+        else:
+            query = "SELECT COUNT(DISTINCT episode) AS games_played, " \
+                    "AVG(generation_time) AS avg_gen_time, SUM(reward) AS total_rewards " \
+                    "FROM game_results WHERE game_name = ?"
+            df = pd.read_sql_query(query, conn, params=(game_name,))
+        
+        # Calculate win rate against random bot
+        vs_random_query = """
+            SELECT COUNT(*) FROM game_results 
+            WHERE game_name = ? AND opponent = 'random_None' AND reward > 0
+        """
+        total_vs_random_query = """
+            SELECT COUNT(*) FROM game_results 
+            WHERE game_name = ? AND opponent = 'random_None'
+        """
+        wins_vs_random = conn.execute(vs_random_query, (game_name,)).fetchone()[0] or 0
+        total_vs_random = conn.execute(total_vs_random_query, (game_name,)).fetchone()[0] or 0
+        vs_random_rate = (wins_vs_random / total_vs_random * 100) if total_vs_random > 0 else 0
         
-        df = pd.read_sql_query(query, conn, params=(game_name,))
         df["agent_name"] = model_name
         df["agent_type"] = agent_type
+        df["vs_random"] = round(vs_random_rate, 2)
+        
         all_stats.append(df)
         conn.close()
     
     leaderboard_df = pd.concat(all_stats, ignore_index=True) if all_stats else pd.DataFrame()
+    
+    if leaderboard_df.empty:
+        leaderboard_df = pd.DataFrame(columns=["LLM Model", "# games", "moves/game", "illegal-moves", "win-rate", "vs Random"])
+    
     return leaderboard_df
 
 def generate_leaderboard_json():
     """Generate a JSON file containing leaderboard stats."""
     available_games = get_available_games()
-    leaderboard = extract_leaderboard_stats(available_games[0]).to_dict(orient="records")
+    leaderboard = extract_leaderboard_stats("Total Performance").to_dict(orient="records")
     json_file = "results/leaderboard_stats.json"
     with open(json_file, "w", encoding="utf-8") as f:
         json.dump({"timestamp": datetime.utcnow().isoformat(), "leaderboard": leaderboard}, f, indent=4)
@@ -83,10 +110,10 @@ def generate_leaderboard_json():
 
 with gr.Blocks() as interface:
     with gr.Tab("Leaderboard"):
-        gr.Markdown("# Leaderboard")
+        gr.Markdown("# LLM Model Leaderboard\nTrack performance across different games!")
         available_games = get_available_games()
-        leaderboard_game_dropdown = gr.Dropdown(available_games, label="Select Game", value=available_games[0])
-        leaderboard_table = gr.Dataframe()
+        leaderboard_game_dropdown = gr.Dropdown(available_games, label="Select Game", value="Total Performance")
+        leaderboard_table = gr.Dataframe(headers=["LLM Model", "# games", "moves/game", "illegal-moves", "win-rate", "vs Random"])
         generate_button = gr.Button("Generate Leaderboard JSON")
         download_component = gr.File(label="Download Leaderboard JSON")
         refresh_button = gr.Button("Refresh Leaderboard")