Update ml_engine/monte_carlo.py

ml_engine/monte_carlo.py

@@ -1,4 +1,4 @@
-# ml_engine/monte_carlo.py (Updated to V6.2 - Stablecoin Guard)
+# ml_engine/monte_carlo.py (Updated to V10.0 - Added Simple Sim for Ranker)
 import numpy as np
 import pandas as pd
 from arch import arch_model
@@ -24,10 +24,8 @@ def _sanitize_results_for_json(results_dict):
         return [_sanitize_results_for_json(v) for v in results_dict]
     elif isinstance(results_dict, np.ndarray):
         return results_dict.tolist()
-    # (Removed np.float_ which is deprecated in NumPy 2.0)
     elif isinstance(results_dict, (np.float64, np.float32)):
         return float(results_dict)
-    # (Removed np.int_ which is deprecated in NumPy 2.0)
     elif isinstance(results_dict, (np.int64, np.int32)):
         return int(results_dict)
     else:
@@ -37,10 +35,74 @@ def _sanitize_results_for_json(results_dict):
 class MonteCarloAnalyzer:
     def __init__(self):
         self.simulation_results = {}
-    
+        
+    # 🔴 --- START OF NEW FUNCTION (V10.0) --- 🔴
+    # (هذه الدالة هي التي يحتاجها الرانكر V9.8)
+    def generate_1h_price_distribution_simple(self, closes_np: np.ndarray, target_profit_percent=0.005) -> dict:
+        """
+        (V10.0) نسخة سريعة جداً (غير متزامنة) مخصصة للرانكر.
+        تقبل numpy array مباشرة وتُرجع الميزات المطلوبة فقط.
+        """
+        try:
+            # (نحتاج 100 شمعة كما في Colab)
+            if len(closes_np) < 30: 
+                return {'mc_prob_gain': 0.5, 'mc_var_95_pct': 0.0, 'error': True}
+            
+            current_price = closes_np[-1]
+            if current_price <= 0: 
+                return {'mc_prob_gain': 0.5, 'mc_var_95_pct': 0.0, 'error': True}
+
+            log_returns = np.log(closes_np[1:] / closes_np[:-1])
+            log_returns = log_returns[~np.isnan(log_returns) & ~np.isinf(log_returns)]
+            
+            if len(log_returns) < 20: 
+                return {'mc_prob_gain': 0.5, 'mc_var_95_pct': 0.0, 'error': True}
+
+            mean_return = np.mean(log_returns)
+            std_return = np.std(log_returns)
+            
+            if std_return < 1e-5: # (واقي العملة المستقرة)
+                return {'mc_prob_gain': 0.5, 'mc_var_95_pct': 0.0, 'error': True}
+
+            num_simulations = 1000 # (سريعة، كما في Colab)
+            t_df = 10
+            jump_lambda = 0.05
+            jump_mean = 0.0
+            jump_std = std_return * 3.0
+
+            drift = (mean_return - 0.5 * std_return**2)
+            diffusion = std_return * np.random.standard_t(df=t_df, size=num_simulations)
+            jump_mask = np.random.rand(num_simulations) < jump_lambda
+            jump_sizes = np.random.normal(jump_mean, jump_std, size=num_simulations)
+            jump_component = np.zeros(num_simulations)
+            jump_component[jump_mask] = jump_sizes[jump_mask]
+            
+            simulated_log_returns = drift + diffusion + jump_component
+            simulated_prices = current_price * np.exp(simulated_log_returns)
+
+            percentiles = np.percentile(simulated_prices, [5])
+            VaR_95_price = percentiles[0]
+            # (كنسبة مئوية، مع واقي من القسمة على صفر)
+            VaR_95_value = (current_price - VaR_95_price) / (current_price + 1e-9) 
+            
+            target_price = current_price * (1 + target_profit_percent)
+            probability_of_gain = np.mean(simulated_prices >= target_price)
+            
+            return {
+                'mc_prob_gain': probability_of_gain,
+                'mc_var_95_pct': VaR_95_value,
+                'error': False
+            }
+            
+        except Exception:
+            # (إرجاع قيم محايدة في حالة الفشل)
+            return {'mc_prob_gain': 0.5, 'mc_var_95_pct': 0.0, 'error': True}
+    # 🔴 --- END OF NEW FUNCTION (V10.0) --- 🔴
+
+
     async def generate_1h_price_distribution(self, ohlcv_data, target_profit_percent=0.005):
         """
-        (المرحلة 1 - سريعة)
+        (المرحلة 1 - سريعة) - (للاستخدامات القديمة إن وجدت)
         """
         try:
             if not ohlcv_data or '1h' not in ohlcv_data or len(ohlcv_data['1h']) < 30:
@@ -74,13 +136,9 @@ class MonteCarloAnalyzer:
             mean_return = np.mean(log_returns)
             std_return = np.std(log_returns)
             
-            # 🔴 --- START OF CHANGE (V6.2 - STABLECOIN GUARD) --- 🔴
-            # واقي العملة المستقرة: إذا كان الانحراف المعياري (التقلب) شبه صفري، أوقف التحليل
-            if std_return < 1e-5: # (1e-5 هو 0.00001)
-                print(f"   [MC Guard] {ohlcv_data.get('symbol', 'Symbol')} - Zero volatility detected. Stopping MC.")
+            if std_return < 1e-5: # (واقي العملة المستقرة)
                 self.simulation_results = {'error': 'Zero volatility detected (Stablecoin?)'}
                 return None
-            # 🔴 --- END OF CHANGE --- 🔴
 
             num_simulations = 5000
             t_df = 10
@@ -129,7 +187,6 @@ class MonteCarloAnalyzer:
                 'raw_simulated_prices': simulated_prices[:100]
             }
             
-            # (Sanitize Phase 1 results as well)
             return _sanitize_results_for_json(self.simulation_results)
             
         except Exception as e:
@@ -140,7 +197,7 @@ class MonteCarloAnalyzer:
 
     async def generate_1h_distribution_advanced(self, ohlcv_data, target_profit_percent=0.005):
         """
-        (المرحلة 2+3 - متقدمة)
+        (المرحلة 2+3 - متقدمة) - (للاستخدام في الطبقة 2)
         """
         try:
             if not ohlcv_data or '1h' not in ohlcv_data or len(ohlcv_data['1h']) < 50:
@@ -161,27 +218,20 @@ class MonteCarloAnalyzer:
             df['log_returns'] = np.log(df['close'] / df['close'].shift(1)).fillna(0)
             log_returns_series = df['log_returns'].replace([np.inf, -np.inf], 0)
 
-            # 🔴 --- START OF CHANGE (V6.2 - STABLECOIN GUARD) --- 🔴
-            # واقي العملة المستقرة: التحقق من التقلب قبل بدء التحليل
             std_return_check = np.std(log_returns_series.iloc[-30:])
-            if std_return_check < 1e-5: # (1e-5 هو 0.00001)
-                print(f"   [MC Guard Adv] {ohlcv_data.get('symbol', 'Symbol')} - Zero volatility detected. Stopping GARCH/LGBM.")
+            if std_return_check < 1e-5: # (واقي العملة المستقرة)
                 self.simulation_results = {'error': 'Zero volatility detected (Stablecoin?)'}
-                # العودة إلى المرحلة 1 (التي ستمسك بها أيضاً وترجع None)
                 return await self.generate_1h_price_distribution(ohlcv_data, target_profit_percent)
-            # 🔴 --- END OF CHANGE --- 🔴
 
             # 3. (Phase 2) توقع التقلب باستخدام GARCH(1,1)
             try:
-                # (Rescale by 100, and set rescale=False to stop GARCH from auto-scaling)
                 garch_model = arch_model(log_returns_series * 100, vol='Garch', p=1, q=1, dist='t', rescale=False)
                 res = garch_model.fit(update_freq=0, disp='off') 
                 forecast = res.forecast(horizon=1)
-                # (Divide by 100^2 = 10000)
                 forecasted_var = forecast.variance.iloc[-1, 0] / (100**2)
                 forecasted_std_return = np.sqrt(forecasted_var)
             except Exception as garch_err:
-                forecasted_std_return = std_return_check # (استخدام القيمة التي تم التحقق منها)
+                forecasted_std_return = std_return_check 
                 print(f"⚠️ GARCH failed, using std: {garch_err}")
 
 
@@ -195,7 +245,7 @@ class MonteCarloAnalyzer:
                     delta = df['close'].diff()
                     gain = (delta.where(delta > 0, 0)).rolling(window=14).mean()
                     loss = (-delta.where(delta < 0, 0)).rolling(window=14).mean()
-                    rs = gain / (loss + 1e-9) # (Added 1e-9 to prevent zero division)
+                    rs = gain / (loss + 1e-9)
                     df['rsi'] = 100 - (100 / (1 + rs))
                     df['macd_hist'] = df['close'].ewm(span=12).mean() - df['close'].ewm(span=26).mean()
 
@@ -281,7 +331,6 @@ class MonteCarloAnalyzer:
                 'raw_simulated_prices': simulated_prices[:100]
             }
             
-            # (Sanitize the results before returning)
             return _sanitize_results_for_json(self.simulation_results)
             
         except Exception as e:
@@ -303,4 +352,4 @@ class MonteCarloAnalyzer:
             else: return 1.0
         except Exception: return 1.0
 
-print("✅ ML Module: Advanced Monte Carlo Analyzer loaded (FIXED: NumPy 2.0 & GARCH Scale v2)")
+print("✅ ML Module: Advanced Monte Carlo Analyzer loaded (V10.0 - Simple Sim Added)")