Update learning_hub/adaptive_hub.py

learning_hub/adaptive_hub.py

@@ -1,6 +1,6 @@
 # ==============================================================================
 # 🧠 learning_hub/adaptive_hub.py 
-# (V55.0 - GEM-Architect: The Judge Logic)
+# (V55.1 - GEM-Architect: Fixed & Polished)
 # ==============================================================================
 
 import json
@@ -18,10 +18,8 @@ class StrategyDNA:
         self.filters = filters
         self.guard_settings = guard_settings if guard_settings else {}
         
-        # ✅ الإضافة الجديدة: سجل أداء الباكتست الذي أنتج هذه الإعدادات
-        # هذا هو "السجل الرياضي" للبطل الحالي
         self.backtest_performance = backtest_performance if backtest_performance else {
-            "net_profit": -9999.0, # قيمة منخفضة جداً للبداية
+            "net_profit": -9999.0, 
             "win_rate": 0.0,
             "total_trades": 0,
             "date_recorded": "N/A"
@@ -36,7 +34,7 @@ class StrategyDNA:
             "ob_settings": self.ob_settings,
             "filters": self.filters,
             "guard_settings": self.guard_settings,
-            "backtest_performance": self.backtest_performance, # ✅ حفظ السجل
+            "backtest_performance": self.backtest_performance, 
             "stats": self.stats
         }
 
@@ -46,7 +44,7 @@ class AdaptiveHub:
         self.dna_file_key = "learning/strategic_dna_v5_struct.json" 
         self.current_market_regime = "RANGE"
         self.strategies: Dict[str, StrategyDNA] = {}
-        print("🧠 [AdaptiveHub V55.0] The Judge Core Initialized.")
+        print("🧠 [AdaptiveHub V55.1] Core Initialized.")
 
     async def initialize(self):
         try:
@@ -65,7 +63,7 @@ class AdaptiveHub:
 
     def _create_default_dna(self):
         default_guards = {"hydra_crash": 0.85, "hydra_giveback": 0.70, "legacy_v2": 0.95, "legacy_v3": 0.95}
-        # القيم الافتراضية ليس لها سجل باكتست، لذا ستكون سهلة الهزيمة
+        
         self.strategies["BULL"] = StrategyDNA("BULL", {"titan": 0.50, "structure": 0.30, "sniper": 0.20}, {"wall_ratio_limit": 0.60, "imbalance_thresh": 0.5}, {"l1_min_score": 0.55, "l3_conf_thresh": 0.60}, default_guards)
         self.strategies["BEAR"] = StrategyDNA("BEAR", {"titan": 0.30, "structure": 0.40, "sniper": 0.30}, {"wall_ratio_limit": 0.30, "imbalance_thresh": 0.7}, {"l1_min_score": 0.75, "l3_conf_thresh": 0.75}, default_guards)
         self.strategies["RANGE"] = StrategyDNA("RANGE", {"titan": 0.40, "structure": 0.40, "sniper": 0.20}, {"wall_ratio_limit": 0.40, "imbalance_thresh": 0.6}, {"l1_min_score": 0.65, "l3_conf_thresh": 0.65}, default_guards)
@@ -79,25 +77,16 @@ class AdaptiveHub:
                 val["ob_settings"], 
                 val["filters"], 
                 val.get("guard_settings", {}),
-                val.get("backtest_performance", None) # ✅ تحميل السجل
+                val.get("backtest_performance", None)
             )
         self.current_market_regime = data.get("current_regime", "RANGE")
 
-    # 🔥🔥🔥 الدالة الجديدة: الحكم (The Judge) 🔥🔥🔥
     def submit_challenger(self, regime: str, new_config: dict, new_stats: dict) -> bool:
-        """
-        تقارن بين المتحدي الجديد والبطل الحالي.
-        تعيد True إذا فاز الجديد وتم التحديث، و False إذا تم رفضه.
-        """
         if regime not in self.strategies: return False
         
         champion = self.strategies[regime]
         old_stats = champion.backtest_performance
         
-        # معايير التحكيم:
-        # 1. الربح الصافي هو الملك.
-        # 2. إذا تساوى الربح، نختار نسبة الفوز الأعلى.
-        
         new_profit = new_stats.get('net_profit', -100)
         old_profit = old_stats.get('net_profit', -9999)
         
@@ -106,25 +95,18 @@ class AdaptiveHub:
         print(f"   🥊 Challenger: Profit ${new_profit:.2f} | WinRate {new_stats.get('win_rate', 0):.1f}%")
         
         is_winner = False
-        
-        # القاعدة 1: الربح الصافي يجب أن يكون أعلى بوضوح (أو النظام جديد كلياً)
-        if new_profit > old_profit:
-            is_winner = True
-        # القاعدة 2: إذا الربح متقارب جداً، نفضل نسبة الفوز الأعلى
-        elif abs(new_profit - old_profit) < 0.5 and new_stats.get('win_rate', 0) > old_stats.get('win_rate', 0):
-            is_winner = True
+        if new_profit > old_profit: is_winner = True
+        elif abs(new_profit - old_profit) < 0.5 and new_stats.get('win_rate', 0) > old_stats.get('win_rate', 0): is_winner = True
             
         if is_winner:
             print(f"   ✅ [JUDGE] Challenger WINS! Updating DNA.")
-            # تحديث الجينات
             champion.model_weights['titan'] = new_config['w_titan']
             champion.model_weights['structure'] = new_config['w_struct']
             champion.filters['l1_min_score'] = new_config['thresh']
-            # تحديث سجل البطل
             champion.backtest_performance = new_stats
             return True
         else:
-            print(f"   🛡️ [JUDGE] Champion retains title. Challenger rejected.")
+            print(f"   🛡️ [JUDGE] Champion retains title.")
             return False
 
     async def register_trade_outcome(self, trade_data: Dict[str, Any]):
@@ -159,6 +141,23 @@ class AdaptiveHub:
         SystemLimits.L3_CONFIDENCE_THRESHOLD = active_dna.filters.get("l3_conf_thresh", 0.65)
         SystemLimits.L4_OB_WALL_RATIO = active_dna.ob_settings.get("wall_ratio_limit", 0.4)
 
+    # ✅ الدالة المضافة لحل الخطأ
+    def get_status(self) -> str:
+        """
+        تعيد ملخصاً نصياً لحالة النظام الحالية (للعرض في السجلات أو الواجهة).
+        """
+        dna = self.strategies.get(self.current_market_regime)
+        if not dna: return "System Initializing..."
+        
+        thresh_ratio = dna.filters.get('l1_min_score', 0)
+        titan_w = dna.model_weights.get('titan', 0)
+        struct_w = dna.model_weights.get('structure', 0)
+        
+        return (f"Regime: {self.current_market_regime} | "
+                f"L1 Thresh: {thresh_ratio:.0%} | "
+                f"Titan: {titan_w:.2f} | "
+                f"Struct: {struct_w:.2f}")
+
     async def _save_state_to_r2(self):
         if not self.r2: return
         try: