Update ml_engine/processor.py

ml_engine/processor.py

@@ -1,4 +1,4 @@
-# ml_engine/processor.py (Updated to V6.2 - Normalized MC Score)
+# ml_engine/processor.py (Updated to V6.3 - Exhaustion Penalty Score)
 import pandas as pd
 import numpy as np
 from datetime import datetime
@@ -112,14 +112,15 @@ class MLProcessor:
             print(f"❌ Error creating DataFrame: {e}")
             return pd.DataFrame()
 
-    # 🔴 --- START OF CHANGE (V6.2 - NORMALIZE MC SCORE) --- 🔴
+    # 🔴 --- START OF CHANGE (V6.3 - EXHAUSTION PENALTY) --- 🔴
     def _calculate_enhanced_final_score(self, analysis):
-        """(محدث V6.2) تطبيع درجة مونت كارلو لمعاقبة النتائج العشوائية."""
+        """(محدث V6.3) تطبيق 'عامل جزاء الإرهاق' على النتيجة النهائية."""
         try:
             base_score = analysis.get('final_score', 0)
             pattern_confidence = analysis.get('pattern_analysis', {}).get('pattern_confidence', 0)
             strategy_confidence = analysis.get('strategy_confidence', 0)
             
+            # --- 1. حساب درجة مونت كارلو (كما في V6.2) ---
             mc_distribution = analysis.get('monte_carlo_distribution')
             monte_carlo_score = 0
             
@@ -134,13 +135,12 @@ class MLProcessor:
                     if normalized_var > 0.05: risk_penalty = 0.5           
                     elif normalized_var > 0.03: risk_penalty = 0.8
                     
-                    # (تعديل: تطبيع الاحتمالية)
-                    # (0.5 -> 0), (0.75 -> 0.5), (1.0 -> 1.0)
                     normalized_prob_score = max(0.0, (prob_gain - 0.5) * 2)
                     monte_carlo_score = normalized_prob_score * risk_penalty
                 else:
                     monte_carlo_score = 0 
             
+            # --- 2. حساب درجة الحيتان (كما في V6.2) ---
             whale_confidence = 0
             whale_data = analysis.get('whale_data')
             if whale_data and whale_data.get('data_available'):
@@ -148,6 +148,7 @@ class MLProcessor:
                  if signal.get('action') != 'HOLD' and signal.get('confidence', 0) >= 0.5:
                       whale_confidence = signal.get('confidence', 0)
 
+            # --- 3. حساب النتيجة الموزونة الأولية (كما في V6.2) ---
             components = []
             weights = []
             
@@ -161,7 +162,28 @@ class MLProcessor:
             total_weight = sum(weights)
             if total_weight == 0: return 0
             enhanced_score = sum(comp * weight for comp, weight in zip(components, weights)) / total_weight
-            return min(max(enhanced_score, 0.0), 1.0)
+            
+            # --- 4. (جديد V6.3) تطبيق 'عامل جزاء الإرهاق' ---
+            exhaustion_penalty_factor = 1.0
+            
+            # (جلب البيانات المطلوبة من القاموس)
+            price_change_24h = analysis.get('price_change_24h', 0)
+            rsi_1d = analysis.get('advanced_indicators', {}).get('1d', {}).get('rsi', 50)
+            rsi_4h = analysis.get('advanced_indicators', {}).get('4h', {}).get('rsi', 50)
+
+            if price_change_24h > 60 and (rsi_1d > 80 or rsi_4h > 80):
+                exhaustion_penalty_factor = 0.4 # (عقوبة قاسية - 60%)
+            elif price_change_24h > 40 and (rsi_1d > 75 or rsi_4h > 75):
+                exhaustion_penalty_factor = 0.6 # (عقوبة متوسطة - 40%)
+            
+            if exhaustion_penalty_factor < 1.0:
+                # (اختياري: للطباعة التشخيصية)
+                # print(f"   ⚠️ [Processor] {analysis.get('symbol')} Exhaustion Penalty Applied! Score {enhanced_score:.2f} * {exhaustion_penalty_factor} (24h: {price_change_24h:+.1f}%, RSI 1D: {rsi_1d:.1f})")
+                pass
+
+            final_penalized_score = enhanced_score * exhaustion_penalty_factor
+            
+            return min(max(final_penalized_score, 0.0), 1.0)
             
         except Exception as e:
             print(f"❌ Error calculating enhanced score: {e}")
@@ -169,7 +191,7 @@ class MLProcessor:
     # 🔴 --- END OF CHANGE --- 🔴
 
     async def process_and_score_symbol(self, raw_data):
-        # (This function remains unchanged)
+        """(محدث V6.3) إضافة price_change_24h إلى القاموس"""
         try:
             symbol = raw_data['symbol']
             ohlcv_data = raw_data.get('ohlcv')
@@ -179,11 +201,17 @@ class MLProcessor:
             reasons = raw_data.get('reasons_for_candidacy', [])
             final_score = layer1_score
             successful_timeframes = raw_data.get('successful_timeframes', 0)
+            
+            # 🔴 --- START OF CHANGE (V6.3) --- 🔴
+            price_change_24h = raw_data.get('price_change_24h', 0)
+            # 🔴 --- END OF CHANGE --- 🔴
+
             return {
                 'symbol': symbol, 'current_price': current_price, 'final_score': final_score,
                 'enhanced_final_score': final_score, 'reasons_for_candidacy': reasons,
                 'layer1_score': layer1_score, 'ohlcv': ohlcv_data,
-                'successful_timeframes': successful_timeframes
+                'successful_timeframes': successful_timeframes,
+                'price_change_24h': price_change_24h # (إضافة الحقل هنا)
             }
         except Exception as error:
             print(f"❌ Error in basic symbol processing {raw_data.get('symbol', 'unknown')}: {error}")
@@ -241,4 +269,4 @@ def safe_json_parse(json_string):
             return json.loads(s)
         except json.JSONDecodeError: return None
 
-print("✅ ML Processor loaded - (Patched to pass LearningHub)")
+print("✅ ML Processor loaded - V6.3 (Exhaustion Penalty Score)")