more training

DQN_SENSITIVITY_LEARNING_SUMMARY.md

@@ -0,0 +1,234 @@
+# DQN RL-based Sensitivity Learning & 300s Data Preloading Summary
+
+## Overview
+This document summarizes the implementation of DQN RL-based sensitivity learning and 300s data preloading features that make the trading system more adaptive and responsive.
+
+## 🧠 DQN RL-based Sensitivity Learning
+
+### Core Concept
+The system now uses a Deep Q-Network (DQN) to learn optimal sensitivity levels for trading decisions based on market conditions and trade outcomes. After each completed trade, the system evaluates the performance and creates a learning case for the DQN agent.
+
+### Implementation Details
+
+#### 1. Sensitivity Levels (5 levels: 0-4)
+```python
+sensitivity_levels = {
+    0: {'name': 'very_conservative', 'open_threshold_multiplier': 1.5, 'close_threshold_multiplier': 2.0},
+    1: {'name': 'conservative', 'open_threshold_multiplier': 1.2, 'close_threshold_multiplier': 1.5},
+    2: {'name': 'medium', 'open_threshold_multiplier': 1.0, 'close_threshold_multiplier': 1.0},
+    3: {'name': 'aggressive', 'open_threshold_multiplier': 0.8, 'close_threshold_multiplier': 0.7},
+    4: {'name': 'very_aggressive', 'open_threshold_multiplier': 0.6, 'close_threshold_multiplier': 0.5}
+}
+```
+
+#### 2. Trade Tracking System
+- **Active Trades**: Tracks open positions with entry conditions
+- **Completed Trades**: Records full trade lifecycle with outcomes
+- **Learning Queue**: Stores DQN training cases from completed trades
+
+#### 3. DQN State Vector (15 features)
+- Market volatility (normalized)
+- Price momentum (5-period)
+- Volume ratio
+- RSI indicator
+- MACD signal
+- Bollinger Band position
+- Recent price changes (5 periods)
+- Current sensitivity level
+- Recent performance metrics (avg P&L, win rate, avg duration)
+
+#### 4. Reward Calculation
+```python
+def _calculate_sensitivity_reward(self, completed_trade):
+    base_reward = pnl_pct * 10  # Scale P&L percentage
+    
+    # Duration factor
+    if duration < 300: duration_factor = 0.8      # Too quick
+    elif duration < 1800: duration_factor = 1.2   # Good for scalping
+    elif duration < 3600: duration_factor = 1.0   # Acceptable
+    else: duration_factor = 0.7                   # Too slow
+    
+    # Confidence factor
+    conf_factor = (entry_conf + exit_conf) / 2 if profitable else exit_conf
+    
+    final_reward = base_reward * duration_factor * conf_factor
+    return np.clip(final_reward, -2.0, 2.0)
+```
+
+#### 5. Dynamic Threshold Adjustment
+- **Opening Positions**: Higher thresholds (more conservative)
+- **Closing Positions**: Lower thresholds (more sensitive to exit signals)
+- **Real-time Adaptation**: DQN continuously adjusts sensitivity based on market conditions
+
+### Files Modified
+- `core/enhanced_orchestrator.py`: Added sensitivity learning methods
+- `core/config.py`: Added `confidence_threshold_close` parameter
+- `web/scalping_dashboard.py`: Added sensitivity info display
+- `NN/models/dqn_agent.py`: Existing DQN agent used for sensitivity learning
+
+## 📊 300s Data Preloading
+
+### Core Concept
+The system now preloads 300 seconds worth of data for all symbols and timeframes on first load, providing better initial performance and reducing latency for trading decisions.
+
+### Implementation Details
+
+#### 1. Smart Preloading Logic
+```python
+def _should_preload_data(self, symbol: str, timeframe: str, limit: int) -> bool:
+    # Check if we already have cached data
+    if cached_data exists and len(cached_data) > 0:
+        return False
+    
+    # Calculate candles needed for 300s
+    timeframe_seconds = self.timeframe_seconds.get(timeframe, 60)
+    candles_in_300s = 300 // timeframe_seconds
+    
+    # Preload if beneficial
+    return candles_in_300s > limit or timeframe in ['1s', '1m']
+```
+
+#### 2. Timeframe-Specific Limits
+- **1s timeframe**: Max 300 candles (5 minutes)
+- **1m timeframe**: Max 60 candles (1 hour)
+- **Other timeframes**: Max 500 candles
+- **Minimum**: Always at least 100 candles
+
+#### 3. Preloading Process
+1. Check if data already exists (cache or memory)
+2. Calculate optimal number of candles for 300s
+3. Fetch data from Binance API
+4. Add technical indicators
+5. Cache data for future use
+6. Store in memory for immediate access
+
+#### 4. Performance Benefits
+- **Faster Initial Load**: Charts populate immediately
+- **Reduced API Calls**: Bulk loading vs individual requests
+- **Better User Experience**: No waiting for data on first load
+- **Improved Trading Decisions**: More historical context available
+
+### Files Modified
+- `core/data_provider.py`: Added preloading methods
+- `web/scalping_dashboard.py`: Integrated preloading in initialization
+
+## 🎨 Enhanced Dashboard Features
+
+### 1. Color-Coded Position Display
+- **LONG positions**: Green text with `[LONG]` prefix
+- **SHORT positions**: Red text with `[SHORT]` prefix
+- **Format**: `[SIDE] size @ $entry_price | P&L: $unrealized_pnl`
+
+### 2. Enhanced Model Training Status
+Now displays three columns:
+- **RL Training**: Queue size, win rate, actions
+- **CNN Training**: Perfect moves, confidence, retrospective learning
+- **DQN Sensitivity**: Current level, completed trades, learning queue, thresholds
+
+### 3. Sensitivity Learning Info
+```python
+{
+    'level_name': 'MEDIUM',           # Current sensitivity level
+    'completed_trades': 15,           # Number of completed trades
+    'learning_queue_size': 8,         # DQN training queue size
+    'open_threshold': 0.600,          # Current opening threshold
+    'close_threshold': 0.250          # Current closing threshold
+}
+```
+
+## 🧪 Testing & Verification
+
+### Test Script: `test_sensitivity_learning.py`
+Comprehensive test suite covering:
+1. **300s Data Preloading**: Verifies preloading functionality
+2. **Sensitivity Learning Initialization**: Checks system setup
+3. **Trading Scenario Simulation**: Tests learning case creation
+4. **Threshold Adjustment**: Verifies dynamic threshold changes
+5. **Dashboard Integration**: Tests UI components
+6. **DQN Training Simulation**: Verifies neural network training
+
+### Running Tests
+```bash
+python test_sensitivity_learning.py
+```
+
+Expected output:
+```
+🎯 SENSITIVITY LEARNING SYSTEM READY!
+Features verified:
+  ✅ DQN RL-based sensitivity learning from completed trades
+  ✅ 300s data preloading for faster initial performance
+  ✅ Dynamic threshold adjustment (lower for closing positions)
+  ✅ Color-coded position display ([LONG] green, [SHORT] red)
+  ✅ Enhanced model training status with sensitivity info
+```
+
+## 🚀 Usage Instructions
+
+### 1. Start the Enhanced Dashboard
+```bash
+python run_enhanced_scalping_dashboard.py
+```
+
+### 2. Monitor Sensitivity Learning
+- Watch the "DQN Sensitivity" section in the dashboard
+- Observe threshold adjustments as trades complete
+- Monitor learning queue size for training activity
+
+### 3. Verify Data Preloading
+- Check console logs for preloading status
+- Observe faster initial chart population
+- Monitor reduced API call frequency
+
+## 📈 Expected Benefits
+
+### 1. Improved Trading Performance
+- **Adaptive Sensitivity**: System learns optimal aggressiveness levels
+- **Better Exit Timing**: Lower thresholds for closing positions
+- **Market-Aware Decisions**: Sensitivity adjusts to market conditions
+
+### 2. Enhanced User Experience
+- **Faster Startup**: 300s preloading reduces initial wait time
+- **Visual Clarity**: Color-coded positions improve readability
+- **Better Monitoring**: Enhanced status displays provide more insight
+
+### 3. System Intelligence
+- **Continuous Learning**: DQN improves over time
+- **Retrospective Analysis**: Perfect opportunity detection
+- **Performance Optimization**: Automatic threshold tuning
+
+## 🔧 Configuration
+
+### Key Parameters
+```yaml
+orchestrator:
+  confidence_threshold: 0.5        # Base opening threshold
+  confidence_threshold_close: 0.25 # Base closing threshold (much lower)
+  
+sensitivity_learning:
+  enabled: true
+  state_size: 15
+  action_space: 5
+  learning_rate: 0.001
+  gamma: 0.95
+  epsilon: 0.3
+  batch_size: 32
+```
+
+## 📝 Next Steps
+
+1. **Monitor Performance**: Track sensitivity learning effectiveness
+2. **Tune Parameters**: Adjust DQN hyperparameters based on results
+3. **Expand Features**: Add more market indicators to state vector
+4. **Optimize Preloading**: Fine-tune preloading amounts per timeframe
+5. **Add Persistence**: Save/load DQN models between sessions
+
+## 🎯 Success Metrics
+
+- **Sensitivity Adaptation**: DQN successfully adjusts sensitivity levels
+- **Improved Win Rate**: Better trade outcomes through learned sensitivity
+- **Faster Startup**: <5 seconds for full data preloading
+- **Reduced Latency**: Immediate chart updates on dashboard load
+- **User Satisfaction**: Clear visual feedback and status information
+
+The system now provides intelligent, adaptive trading with enhanced user experience and faster performance! 

ENHANCED_IMPROVEMENTS_SUMMARY.md

@@ -0,0 +1,214 @@
+# Enhanced Trading System Improvements Summary
+
+## Overview
+This document summarizes the major improvements made to the trading system to address:
+1. Color-coded position display
+2. Enhanced model training detection and retrospective learning
+3. Lower confidence thresholds for closing positions
+
+## 🎨 Color-Coded Position Display
+
+### Implementation
+- **File**: `web/scalping_dashboard.py`
+- **Location**: Dashboard callback function (lines ~720-750)
+
+### Features
+- **LONG positions**: Display in green (`text-success` class) with `[LONG]` prefix
+- **SHORT positions**: Display in red (`text-danger` class) with `[SHORT]` prefix
+- **Real-time P&L**: Shows unrealized profit/loss for each position
+- **Format**: `[SIDE] size @ $entry_price | P&L: $unrealized_pnl`
+
+### Example Display
+```
+[LONG] 0.100 @ $2558.15 | P&L: +$0.72    (Green text)
+[SHORT] 0.050 @ $45123.45 | P&L: -$3.66  (Red text)
+```
+
+### Layout Changes
+- Increased open-positions column from `col-md-2` to `col-md-3` for better display
+- Adjusted other columns to maintain layout balance
+
+## 🧠 Enhanced Model Training Detection
+
+### CNN Training Status
+- **File**: `web/scalping_dashboard.py` - `_create_model_training_status()`
+- **Features**:
+  - Active/Idle status indicators
+  - Perfect moves count tracking
+  - Retrospective learning status
+  - Color-coded status (green for active, yellow for idle)
+
+### Training Events Log
+- **File**: `web/scalping_dashboard.py` - `_create_training_events_log()`
+- **Features**:
+  - Real-time perfect opportunity detection
+  - Confidence adjustment recommendations
+  - Pattern detection events
+  - Priority-based event sorting
+  - Detailed outcome percentages
+
+### Event Types
+- 🧠 **CNN**: Perfect move detection with outcome percentages
+- 🤖 **RL**: Experience replay and queue activity
+- ⚙️ **TUNE**: Confidence threshold adjustments
+- ⚡ **TICK**: Violent move pattern detection
+
+## 📊 Retrospective Learning System
+
+### Core Implementation
+- **File**: `core/enhanced_orchestrator.py`
+- **Key Methods**:
+  - `trigger_retrospective_learning()`: Main analysis trigger
+  - `_analyze_missed_opportunities()`: Scans for perfect opportunities
+  - `_adjust_confidence_thresholds()`: Dynamic threshold adjustment
+
+### Perfect Opportunity Detection
+- **Criteria**: Price movements >1% in 5 minutes
+- **Learning**: Creates `PerfectMove` objects for training
+- **Frequency**: Analysis every 5 minutes to avoid overload
+- **Adaptive**: Adjusts thresholds based on recent performance
+
+### Violent Move Detection
+- **Raw Ticks**: Detects price changes >0.1% in <50ms
+- **1s Bars**: Identifies significant bar ranges >0.2%
+- **Patterns**: Analyzes rapid_fire, volume_spike, price_acceleration
+- **Immediate Learning**: Creates perfect moves in real-time
+
+## ⚖️ Dual Confidence Thresholds
+
+### Configuration
+- **File**: `core/config.py`
+- **Opening Threshold**: 0.5 (default) - Higher bar for new positions
+- **Closing Threshold**: 0.25 (default) - Much lower for position exits
+
+### Implementation
+- **File**: `core/enhanced_orchestrator.py`
+- **Method**: `_make_coordinated_decision()`
+- **Logic**: 
+  - Determines if action is opening or closing via `_is_closing_action()`
+  - Applies appropriate threshold based on action type
+  - Tracks positions internally for accurate classification
+
+### Position Tracking
+- **Internal State**: `self.open_positions` tracks current positions
+- **Updates**: Automatically updated on each trading action
+- **Logic**: 
+  - BUY closes SHORT, opens LONG
+  - SELL closes LONG, opens SHORT
+
+### Benefits
+- **Faster Exits**: Lower threshold allows quicker position closure
+- **Risk Management**: Easier to exit losing positions
+- **Scalping Optimized**: Better for high-frequency trading
+
+## 🔄 Background Processing
+
+### Orchestrator Loop
+- **File**: `web/scalping_dashboard.py` - `_start_orchestrator_trading()`
+- **Features**:
+  - Automatic retrospective learning triggers
+  - 30-second decision cycles
+  - Error handling and recovery
+  - Background thread execution
+
+### Data Processing
+- **Raw Tick Handler**: `_handle_raw_tick()` - Processes violent moves
+- **OHLCV Bar Handler**: `_handle_ohlcv_bar()` - Analyzes bar patterns
+- **Pattern Weights**: Configurable weights for different pattern types
+
+## 📈 Enhanced Metrics
+
+### Performance Tracking
+- **File**: `core/enhanced_orchestrator.py` - `get_performance_metrics()`
+- **New Metrics**:
+  - Retrospective learning status
+  - Pattern detection counts
+  - Position tracking information
+  - Dual threshold configuration
+  - Average confidence needed
+
+### Dashboard Integration
+- **Real-time Updates**: All metrics update in real-time
+- **Visual Indicators**: Color-coded status for quick assessment
+- **Detailed Logs**: Comprehensive event logging with priorities
+
+## 🧪 Testing
+
+### Test Script
+- **File**: `test_enhanced_improvements.py`
+- **Coverage**:
+  - Color-coded position display
+  - Confidence threshold logic
+  - Retrospective learning
+  - Tick pattern detection
+  - Dashboard integration
+
+### Verification
+Run the test script to verify all improvements:
+```bash
+python test_enhanced_improvements.py
+```
+
+## 🚀 Key Benefits
+
+### For Traders
+1. **Visual Clarity**: Instant position identification with color coding
+2. **Faster Exits**: Lower closing thresholds for better risk management
+3. **Learning System**: Continuous improvement from missed opportunities
+4. **Real-time Feedback**: Live model training status and events
+
+### For System Performance
+1. **Adaptive Thresholds**: Self-adjusting based on market conditions
+2. **Pattern Recognition**: Enhanced detection of violent moves
+3. **Retrospective Analysis**: Learning from historical perfect opportunities
+4. **Optimized Scalping**: Tailored for high-frequency trading
+
+## 📋 Configuration
+
+### Key Settings
+```yaml
+orchestrator:
+  confidence_threshold: 0.5        # Opening positions
+  confidence_threshold_close: 0.25 # Closing positions (much lower)
+  decision_frequency: 60
+```
+
+### Pattern Weights
+```python
+pattern_weights = {
+    'rapid_fire': 1.5,
+    'volume_spike': 1.3,
+    'price_acceleration': 1.4,
+    'high_frequency_bar': 1.2,
+    'volume_concentration': 1.1
+}
+```
+
+## 🔧 Technical Implementation
+
+### Files Modified
+1. `web/scalping_dashboard.py` - Color-coded positions, enhanced training status
+2. `core/enhanced_orchestrator.py` - Dual thresholds, retrospective learning
+3. `core/config.py` - New configuration parameters
+4. `test_enhanced_improvements.py` - Comprehensive testing
+
+### Dependencies
+- No new dependencies required
+- Uses existing Dash, NumPy, and Pandas libraries
+- Maintains backward compatibility
+
+## 🎯 Results
+
+### Expected Improvements
+1. **Better Position Management**: Clear visual feedback on position status
+2. **Improved Model Performance**: Continuous learning from perfect opportunities
+3. **Faster Risk Response**: Lower thresholds for position exits
+4. **Enhanced Monitoring**: Real-time training status and event logging
+
+### Performance Metrics
+- **Opening Threshold**: 0.5 (conservative for new positions)
+- **Closing Threshold**: 0.25 (aggressive for exits)
+- **Learning Frequency**: Every 5 minutes
+- **Pattern Detection**: Real-time on violent moves
+
+This comprehensive enhancement package addresses all requested improvements while maintaining system stability and performance. 

core/data_provider.py

@@ -137,7 +137,14 @@ class DataProvider:
                         logger.info(f"Using cached data for {symbol} {timeframe}")
                         return cached_data.tail(limit)
 
-            # Fetch from API
+            # Check if we need to preload 300s of data for first load
+            should_preload = self._should_preload_data(symbol, timeframe, limit)
+            
+            if should_preload:
+                logger.info(f"Preloading 300s of data for {symbol} {timeframe}")
+                df = self._preload_300s_data(symbol, timeframe)
+            else:
+                # Fetch from API with requested limit
                 logger.info(f"Fetching historical data for {symbol} {timeframe}")
                 df = self._fetch_from_binance(symbol, timeframe, limit)
 
@@ -154,7 +161,8 @@ class DataProvider:
                     self.historical_data[symbol] = {}
                 self.historical_data[symbol][timeframe] = df
 
-                return df
+                # Return requested amount
+                return df.tail(limit)
 
             logger.warning(f"No data received for {symbol} {timeframe}")
             return None
@@ -163,6 +171,124 @@ class DataProvider:
             logger.error(f"Error fetching historical data for {symbol} {timeframe}: {e}")
             return None
     
+    def _should_preload_data(self, symbol: str, timeframe: str, limit: int) -> bool:
+        """Determine if we should preload 300s of data"""
+        try:
+            # Check if we have any cached data
+            if self.cache_enabled:
+                cached_data = self._load_from_cache(symbol, timeframe)
+                if cached_data is not None and len(cached_data) > 0:
+                    return False  # Already have some data
+            
+            # Check if we have data in memory
+            if (symbol in self.historical_data and 
+                timeframe in self.historical_data[symbol] and 
+                len(self.historical_data[symbol][timeframe]) > 0):
+                return False  # Already have data in memory
+            
+            # Calculate if 300s worth of data would be more than requested limit
+            timeframe_seconds = self.timeframe_seconds.get(timeframe, 60)
+            candles_in_300s = 300 // timeframe_seconds
+            
+            # Preload if we need more than the requested limit or if it's a short timeframe
+            if candles_in_300s > limit or timeframe in ['1s', '1m']:
+                return True
+            
+            return False
+            
+        except Exception as e:
+            logger.error(f"Error determining if should preload data: {e}")
+            return False
+    
+    def _preload_300s_data(self, symbol: str, timeframe: str) -> Optional[pd.DataFrame]:
+        """Preload 300 seconds worth of data for better initial performance"""
+        try:
+            # Calculate how many candles we need for 300 seconds
+            timeframe_seconds = self.timeframe_seconds.get(timeframe, 60)
+            candles_needed = max(300 // timeframe_seconds, 100)  # At least 100 candles
+            
+            # For very short timeframes, limit to reasonable amount
+            if timeframe == '1s':
+                candles_needed = min(candles_needed, 300)  # Max 300 1s candles
+            elif timeframe == '1m':
+                candles_needed = min(candles_needed, 60)   # Max 60 1m candles (1 hour)
+            else:
+                candles_needed = min(candles_needed, 500)  # Max 500 candles for other timeframes
+            
+            logger.info(f"Preloading {candles_needed} candles for {symbol} {timeframe} (300s worth)")
+            
+            # Fetch the data
+            df = self._fetch_from_binance(symbol, timeframe, candles_needed)
+            
+            if df is not None and not df.empty:
+                logger.info(f"Successfully preloaded {len(df)} candles for {symbol} {timeframe}")
+                return df
+            else:
+                logger.warning(f"Failed to preload data for {symbol} {timeframe}")
+                return None
+                
+        except Exception as e:
+            logger.error(f"Error preloading 300s data for {symbol} {timeframe}: {e}")
+            return None
+    
+    def preload_all_symbols_data(self, timeframes: List[str] = None) -> Dict[str, Dict[str, bool]]:
+        """Preload 300s of data for all symbols and timeframes"""
+        try:
+            if timeframes is None:
+                timeframes = self.timeframes
+            
+            preload_results = {}
+            
+            for symbol in self.symbols:
+                preload_results[symbol] = {}
+                
+                for timeframe in timeframes:
+                    try:
+                        logger.info(f"Preloading data for {symbol} {timeframe}")
+                        
+                        # Check if we should preload
+                        if self._should_preload_data(symbol, timeframe, 100):
+                            df = self._preload_300s_data(symbol, timeframe)
+                            
+                            if df is not None and not df.empty:
+                                # Add technical indicators
+                                df = self._add_technical_indicators(df)
+                                
+                                # Cache the data
+                                if self.cache_enabled:
+                                    self._save_to_cache(df, symbol, timeframe)
+                                
+                                # Store in memory
+                                if symbol not in self.historical_data:
+                                    self.historical_data[symbol] = {}
+                                self.historical_data[symbol][timeframe] = df
+                                
+                                preload_results[symbol][timeframe] = True
+                                logger.info(f"✅ Preloaded {len(df)} candles for {symbol} {timeframe}")
+                            else:
+                                preload_results[symbol][timeframe] = False
+                                logger.warning(f"❌ Failed to preload {symbol} {timeframe}")
+                        else:
+                            preload_results[symbol][timeframe] = True  # Already have data
+                            logger.info(f"⏭️ Skipped preloading {symbol} {timeframe} (already have data)")
+                            
+                    except Exception as e:
+                        logger.error(f"Error preloading {symbol} {timeframe}: {e}")
+                        preload_results[symbol][timeframe] = False
+            
+            # Log summary
+            total_pairs = len(self.symbols) * len(timeframes)
+            successful_pairs = sum(1 for symbol_results in preload_results.values() 
+                                 for success in symbol_results.values() if success)
+            
+            logger.info(f"Preloading completed: {successful_pairs}/{total_pairs} symbol-timeframe pairs loaded")
+            
+            return preload_results
+            
+        except Exception as e:
+            logger.error(f"Error in preload_all_symbols_data: {e}")
+            return {}
+    
     def _fetch_from_binance(self, symbol: str, timeframe: str, limit: int) -> Optional[pd.DataFrame]:
         """Fetch data from Binance API"""
         try:

core/enhanced_orchestrator.py

@@ -117,6 +117,25 @@ class EnhancedTradingOrchestrator:
         self.confidence_threshold_close = self.config.orchestrator.get('confidence_threshold_close', 0.25)  # Much lower for closing
         self.decision_frequency = self.config.orchestrator.get('decision_frequency', 30)
         
+        # DQN RL-based sensitivity learning parameters
+        self.sensitivity_learning_enabled = True
+        self.sensitivity_dqn_agent = None  # Will be initialized when first DQN model is available
+        self.sensitivity_state_size = 20  # Features for sensitivity learning
+        self.sensitivity_action_space = 5  # 5 sensitivity levels: very_low, low, medium, high, very_high
+        self.current_sensitivity_level = 2  # Start with medium (index 2)
+        self.sensitivity_levels = {
+            0: {'name': 'very_low', 'close_threshold_multiplier': 0.5, 'open_threshold_multiplier': 1.2},
+            1: {'name': 'low', 'close_threshold_multiplier': 0.7, 'open_threshold_multiplier': 1.1},
+            2: {'name': 'medium', 'close_threshold_multiplier': 1.0, 'open_threshold_multiplier': 1.0},
+            3: {'name': 'high', 'close_threshold_multiplier': 1.3, 'open_threshold_multiplier': 0.9},
+            4: {'name': 'very_high', 'close_threshold_multiplier': 1.5, 'open_threshold_multiplier': 0.8}
+        }
+        
+        # Trade tracking for sensitivity learning
+        self.active_trades = {}  # symbol -> trade_info with entry details
+        self.completed_trades = deque(maxlen=1000)  # Store last 1000 completed trades for learning
+        self.sensitivity_learning_queue = deque(maxlen=500)  # Queue for DQN training
+        
         # Enhanced weighting system
         self.timeframe_weights = self._initialize_timeframe_weights()
         self.symbol_correlation_matrix = self._initialize_correlation_matrix()
@@ -172,6 +191,7 @@ class EnhancedTradingOrchestrator:
         logger.info("Real-time tick processor integrated for ultra-low latency processing")
         logger.info("Raw tick and OHLCV bar processing enabled for pattern detection")
         logger.info("Enhanced retrospective learning enabled for perfect opportunity detection")
+        logger.info("DQN RL-based sensitivity learning enabled for adaptive thresholds")
     
     def _initialize_timeframe_weights(self) -> Dict[str, float]:
         """Initialize weights for different timeframes"""
@@ -640,8 +660,10 @@ class EnhancedTradingOrchestrator:
         if action.action == 'BUY':
             # Close any short position, open long position
             if symbol in self.open_positions and self.open_positions[symbol]['side'] == 'SHORT':
+                self._close_trade_for_sensitivity_learning(symbol, action)
                 del self.open_positions[symbol]
             else:
+                self._open_trade_for_sensitivity_learning(symbol, action)
                 self.open_positions[symbol] = {
                     'side': 'LONG',
                     'entry_price': action.price,
@@ -650,474 +672,464 @@ class EnhancedTradingOrchestrator:
         elif action.action == 'SELL':
             # Close any long position, open short position
             if symbol in self.open_positions and self.open_positions[symbol]['side'] == 'LONG':
+                self._close_trade_for_sensitivity_learning(symbol, action)
                 del self.open_positions[symbol]
             else:
+                self._open_trade_for_sensitivity_learning(symbol, action)
                 self.open_positions[symbol] = {
                     'side': 'SHORT',
                     'entry_price': action.price,
                     'timestamp': action.timestamp
                 }
     
-    async def trigger_retrospective_learning(self):
+    def _open_trade_for_sensitivity_learning(self, symbol: str, action: TradingAction):
-        """Trigger retrospective learning analysis on recent perfect opportunities"""
+        """Track trade opening for sensitivity learning"""
         try:
-            current_time = datetime.now()
+            # Get current market state for learning context
+            market_state = self._get_current_market_state_for_sensitivity(symbol)
             
-            # Only run retrospective analysis every 5 minutes to avoid overload
+            trade_info = {
-            if (current_time - self.last_retrospective_analysis).total_seconds() < 300:
+                'symbol': symbol,
-                return
+                'side': 'LONG' if action.action == 'BUY' else 'SHORT',
+                'entry_price': action.price,
+                'entry_time': action.timestamp,
+                'entry_confidence': action.confidence,
+                'entry_market_state': market_state,
+                'sensitivity_level_at_entry': self.current_sensitivity_level,
+                'thresholds_used': {
+                    'open': self._get_current_open_threshold(),
+                    'close': self._get_current_close_threshold()
+                }
+            }
             
-            self.last_retrospective_analysis = current_time
+            self.active_trades[symbol] = trade_info
-            
+            logger.info(f"Opened trade for sensitivity learning: {symbol} {trade_info['side']} @ ${action.price:.2f}")
-            # Analyze recent market moves for missed opportunities
-            await self._analyze_missed_opportunities()
-            
-            # Update model confidence thresholds based on recent performance
-            self._adjust_confidence_thresholds()
-            
-            # Mark retrospective learning as active
-            self.retrospective_learning_active = True
-            
-            logger.info("Retrospective learning analysis completed")
             
         except Exception as e:
-            logger.error(f"Error in retrospective learning: {e}")
+            logger.error(f"Error tracking trade opening for sensitivity learning: {e}")
     
-    async def _analyze_missed_opportunities(self):
+    def _close_trade_for_sensitivity_learning(self, symbol: str, action: TradingAction):
-        """Analyze recent price movements to identify missed perfect opportunities"""
+        """Track trade closing and create learning case for DQN"""
+        try:
+            if symbol not in self.active_trades:
+                return
+            
+            trade_info = self.active_trades[symbol]
+            
+            # Calculate trade outcome
+            entry_price = trade_info['entry_price']
+            exit_price = action.price
+            side = trade_info['side']
+            
+            if side == 'LONG':
+                pnl_pct = (exit_price - entry_price) / entry_price
+            else:  # SHORT
+                pnl_pct = (entry_price - exit_price) / entry_price
+            
+            # Calculate trade duration
+            duration = (action.timestamp - trade_info['entry_time']).total_seconds()
+            
+            # Get current market state for exit context
+            exit_market_state = self._get_current_market_state_for_sensitivity(symbol)
+            
+            # Create completed trade record
+            completed_trade = {
+                'symbol': symbol,
+                'side': side,
+                'entry_price': entry_price,
+                'exit_price': exit_price,
+                'entry_time': trade_info['entry_time'],
+                'exit_time': action.timestamp,
+                'duration': duration,
+                'pnl_pct': pnl_pct,
+                'entry_confidence': trade_info['entry_confidence'],
+                'exit_confidence': action.confidence,
+                'entry_market_state': trade_info['entry_market_state'],
+                'exit_market_state': exit_market_state,
+                'sensitivity_level_used': trade_info['sensitivity_level_at_entry'],
+                'thresholds_used': trade_info['thresholds_used']
+            }
+            
+            self.completed_trades.append(completed_trade)
+            
+            # Create sensitivity learning case for DQN
+            self._create_sensitivity_learning_case(completed_trade)
+            
+            # Remove from active trades
+            del self.active_trades[symbol]
+            
+            logger.info(f"Closed trade for sensitivity learning: {symbol} {side} P&L: {pnl_pct*100:+.2f}% Duration: {duration:.0f}s")
+            
+        except Exception as e:
+            logger.error(f"Error tracking trade closing for sensitivity learning: {e}")
+    
+    def _get_current_market_state_for_sensitivity(self, symbol: str) -> Dict[str, float]:
+        """Get current market state features for sensitivity learning"""
         try:
-            for symbol in self.symbols:
             # Get recent price data
-                recent_data = self.data_provider.get_latest_candles(symbol, '1m', limit=60)
+            recent_data = self.data_provider.get_historical_data(symbol, '1m', limit=20)
             
             if recent_data is None or len(recent_data) < 10:
-                    continue
+                return self._get_default_market_state()
             
-                # Look for significant price movements (>1% in 5 minutes)
+            # Calculate market features
-                for i in range(5, len(recent_data)):
+            current_price = recent_data['close'].iloc[-1]
-                    price_change = (recent_data.iloc[i]['close'] - recent_data.iloc[i-5]['close']) / recent_data.iloc[i-5]['close']
             
-                    if abs(price_change) > 0.01:  # 1% move
+            # Volatility (20-period)
-                        # This was a perfect opportunity
+            volatility = recent_data['close'].pct_change().std() * 100
-                        optimal_action = 'BUY' if price_change > 0 else 'SELL'
             
-                        # Create perfect move for retrospective learning
+            # Price momentum (5-period)
-                        perfect_move = PerfectMove(
+            momentum_5 = (current_price - recent_data['close'].iloc[-6]) / recent_data['close'].iloc[-6] * 100
-                            symbol=symbol,
-                            timeframe='1m',
-                            timestamp=recent_data.iloc[i-5]['timestamp'],
-                            optimal_action=optimal_action,
-                            actual_outcome=price_change,
-                            market_state_before=None,  # Would need to reconstruct
-                            market_state_after=None,   # Would need to reconstruct
-                            confidence_should_have_been=min(0.95, abs(price_change) * 20)  # Higher confidence for bigger moves
-                        )
             
-                        self.perfect_moves.append(perfect_move)
+            # Volume ratio
+            avg_volume = recent_data['volume'].mean()
+            current_volume = recent_data['volume'].iloc[-1]
+            volume_ratio = current_volume / avg_volume if avg_volume > 0 else 1.0
             
-                        logger.info(f"Retrospective perfect opportunity identified: {optimal_action} {symbol} ({price_change*100:+.2f}%)")
+            # RSI
+            rsi = recent_data['rsi'].iloc[-1] if 'rsi' in recent_data.columns else 50.0
             
-        except Exception as e:
+            # MACD signal
-            logger.error(f"Error analyzing missed opportunities: {e}")
+            macd_signal = 0.0
+            if 'macd' in recent_data.columns and 'macd_signal' in recent_data.columns:
+                macd_signal = recent_data['macd'].iloc[-1] - recent_data['macd_signal'].iloc[-1]
             
-    def _adjust_confidence_thresholds(self):
+            # Bollinger Band position
-        """Dynamically adjust confidence thresholds based on recent performance"""
+            bb_position = 0.5  # Default middle
-        try:
+            if 'bb_upper' in recent_data.columns and 'bb_lower' in recent_data.columns:
-            if len(self.perfect_moves) < 10:
+                bb_upper = recent_data['bb_upper'].iloc[-1]
-                return
+                bb_lower = recent_data['bb_lower'].iloc[-1]
+                if bb_upper > bb_lower:
+                    bb_position = (current_price - bb_lower) / (bb_upper - bb_lower)
             
-            # Analyze recent perfect moves
+            # Recent price change patterns
-            recent_moves = list(self.perfect_moves)[-20:]
+            price_changes = recent_data['close'].pct_change().tail(5).tolist()
-            avg_confidence_needed = np.mean([move.confidence_should_have_been for move in recent_moves])
-            avg_outcome = np.mean([abs(move.actual_outcome) for move in recent_moves])
-            
-            # Adjust opening threshold based on missed opportunities
-            if avg_confidence_needed > self.confidence_threshold_open:
-                adjustment = min(0.1, (avg_confidence_needed - self.confidence_threshold_open) * 0.1)
-                self.confidence_threshold_open = max(0.3, self.confidence_threshold_open - adjustment)
-                logger.info(f"Adjusted opening confidence threshold to {self.confidence_threshold_open:.3f}")
-            
-            # Keep closing threshold very low for sensitivity
-            if avg_outcome > 0.02:  # If we're seeing big moves
-                self.confidence_threshold_close = max(0.15, self.confidence_threshold_close * 0.9)
-                logger.info(f"Lowered closing confidence threshold to {self.confidence_threshold_close:.3f}")
-                
-        except Exception as e:
-            logger.error(f"Error adjusting confidence thresholds: {e}")
-    
-    def _get_correlated_sentiment(self, symbol: str, 
-                                all_predictions: Dict[str, List[EnhancedPrediction]]) -> Dict[str, Any]:
-        """Get sentiment from correlated symbols"""
-        correlated_actions = []
-        correlated_confidences = []
-        
-        for other_symbol, predictions in all_predictions.items():
-            if other_symbol != symbol and predictions:
-                correlation = self.symbol_correlation_matrix.get((symbol, other_symbol), 0.0)
-                
-                if correlation > 0.5:  # Only consider significantly correlated symbols
-                    best_pred = max(predictions, key=lambda p: p.overall_confidence)
-                    correlated_actions.append(best_pred.overall_action)
-                    correlated_confidences.append(best_pred.overall_confidence * correlation)
-        
-        if not correlated_actions:
-            return {'agreement': 1.0, 'sentiment': 'NEUTRAL'}
-        
-        # Calculate agreement
-        primary_pred = all_predictions[symbol][0] if all_predictions.get(symbol) else None
-        if primary_pred:
-            agreement_count = sum(1 for action in correlated_actions 
-                                if action == primary_pred.overall_action)
-            agreement = agreement_count / len(correlated_actions)
-        else:
-            agreement = 0.5
-        
-        # Calculate overall sentiment
-        action_weights = {'BUY': 0.0, 'SELL': 0.0, 'HOLD': 0.0}
-        for action, confidence in zip(correlated_actions, correlated_confidences):
-            action_weights[action] += confidence
-        
-        dominant_sentiment = max(action_weights, key=action_weights.get)
             
             return {
-            'agreement': agreement,
+                'volatility': volatility,
-            'sentiment': dominant_sentiment,
+                'momentum_5': momentum_5,
-            'correlated_symbols': len(correlated_actions)
+                'volume_ratio': volume_ratio,
+                'rsi': rsi,
+                'macd_signal': macd_signal,
+                'bb_position': bb_position,
+                'price_change_1': price_changes[-1] if len(price_changes) > 0 else 0.0,
+                'price_change_2': price_changes[-2] if len(price_changes) > 1 else 0.0,
+                'price_change_3': price_changes[-3] if len(price_changes) > 2 else 0.0,
+                'price_change_4': price_changes[-4] if len(price_changes) > 3 else 0.0,
+                'price_change_5': price_changes[-5] if len(price_changes) > 4 else 0.0
             }
             
-    def _queue_for_rl_evaluation(self, action: TradingAction, market_state: MarketState):
-        """Queue trading action for RL evaluation"""
-        evaluation_item = {
-            'action': action,
-            'market_state_before': market_state,
-            'timestamp': datetime.now(),
-            'evaluation_pending': True
-        }
-        self.rl_evaluation_queue.append(evaluation_item)
-    
-    async def evaluate_actions_with_rl(self):
-        """Evaluate recent actions using RL agents for continuous learning"""
-        if not self.rl_evaluation_queue:
-            return
-        
-        current_time = datetime.now()
-        
-        # Process actions that are ready for evaluation (e.g., 1 hour old)
-        for item in list(self.rl_evaluation_queue):
-            if item['evaluation_pending']:
-                time_since_action = (current_time - item['timestamp']).total_seconds()
-                
-                # Evaluate after sufficient time has passed
-                if time_since_action >= 3600:  # 1 hour
-                    await self._evaluate_single_action(item)
-                    item['evaluation_pending'] = False
-    
-    async def _evaluate_single_action(self, evaluation_item: Dict[str, Any]):
-        """Evaluate a single action using RL"""
-        try:
-            action = evaluation_item['action']
-            initial_state = evaluation_item['market_state_before']
-            
-            # Get current market state for comparison
-            current_market_states = await self._get_all_market_states_universal(self.universal_adapter.get_universal_data_stream())
-            current_state = current_market_states.get(action.symbol)
-            
-            if current_state:
-                # Calculate reward based on price movement
-                initial_price = initial_state.prices.get(self.timeframes[0], 0)
-                current_price = current_state.prices.get(self.timeframes[0], 0)
-                
-                if initial_price > 0:
-                    price_change = (current_price - initial_price) / initial_price
-                    
-                    # Calculate reward based on action and price movement
-                    reward = self._calculate_reward(action.action, price_change, action.confidence)
-                    
-                    # Update RL agents
-                    await self._update_rl_agents(action, initial_state, current_state, reward)
-                    
-                    # Check if this was a perfect move for CNN training
-                    if abs(reward) > 0.02:  # Significant outcome
-                        self._mark_perfect_move(action, initial_state, current_state, reward)
-                        
         except Exception as e:
-            logger.error(f"Error evaluating action: {e}")
+            logger.error(f"Error getting market state for sensitivity learning: {e}")
+            return self._get_default_market_state()
     
-    def _calculate_reward(self, action: str, price_change: float, confidence: float) -> float:
+    def _get_default_market_state(self) -> Dict[str, float]:
-        """Calculate reward for RL training"""
+        """Get default market state when data is unavailable"""
-        base_reward = 0.0
+        return {
+            'volatility': 2.0,
+            'momentum_5': 0.0,
+            'volume_ratio': 1.0,
+            'rsi': 50.0,
+            'macd_signal': 0.0,
+            'bb_position': 0.5,
+            'price_change_1': 0.0,
+            'price_change_2': 0.0,
+            'price_change_3': 0.0,
+            'price_change_4': 0.0,
+            'price_change_5': 0.0
+        }
     
-        if action == 'BUY' and price_change > 0:
+    def _create_sensitivity_learning_case(self, completed_trade: Dict[str, Any]):
-            base_reward = price_change * 10  # Reward proportional to gain
+        """Create a learning case for the DQN sensitivity agent"""
-        elif action == 'SELL' and price_change < 0:
-            base_reward = abs(price_change) * 10  # Reward for avoiding loss
-        elif action == 'HOLD':
-            base_reward = 0.01 if abs(price_change) < 0.005 else -0.01  # Small reward for correct holds
-        else:
-            base_reward = -abs(price_change) * 5  # Penalty for wrong actions
-        
-        # Adjust reward based on confidence
-        confidence_multiplier = 0.5 + confidence  # 0.5 to 1.5 range
-        
-        return base_reward * confidence_multiplier
-    
-    async def _update_rl_agents(self, action: TradingAction, initial_state: MarketState,
-                              current_state: MarketState, reward: float):
-        """Update RL agents with action evaluation"""
-        for model_name, model in self.model_registry.models.items():
-            if isinstance(model, RLAgentInterface):
         try:
-                    # Convert market states to RL state format
+            # Create state vector from market conditions at entry
-                    initial_rl_state = self._market_state_to_rl_state(initial_state)
+            entry_state = self._market_state_to_sensitivity_state(
-                    current_rl_state = self._market_state_to_rl_state(current_state)
+                completed_trade['entry_market_state'],
-                    
+                completed_trade['sensitivity_level_used']
-                    # Convert action to RL action index
-                    action_idx = {'SELL': 0, 'HOLD': 1, 'BUY': 2}.get(action.action, 1)
-                    
-                    # Store experience
-                    model.remember(
-                        state=initial_rl_state,
-                        action=action_idx,
-                        reward=reward,
-                        next_state=current_rl_state,
-                        done=False
             )
             
-                    # Trigger replay learning
+            # Create state vector from market conditions at exit
-                    loss = model.replay()
+            exit_state = self._market_state_to_sensitivity_state(
-                    if loss is not None:
+                completed_trade['exit_market_state'],
-                        logger.info(f"RL agent {model_name} updated with loss: {loss:.4f}")
+                completed_trade['sensitivity_level_used']
-                        
-                except Exception as e:
-                    logger.error(f"Error updating RL agent {model_name}: {e}")
-    
-    def _mark_perfect_move(self, action: TradingAction, initial_state: MarketState,
-                          final_state: MarketState, reward: float):
-        """Mark a perfect move for CNN training"""
-        try:
-            # Determine what the optimal action should have been
-            optimal_action = action.action if reward > 0 else ('HOLD' if action.action == 'HOLD' else 
-                           ('SELL' if action.action == 'BUY' else 'BUY'))
-            
-            # Calculate what confidence should have been
-            optimal_confidence = min(0.95, abs(reward) * 10)  # Higher reward = higher confidence should have been
-            
-            for tf_pred in action.timeframe_analysis:
-                perfect_move = PerfectMove(
-                    symbol=action.symbol,
-                    timeframe=tf_pred.timeframe,
-                    timestamp=action.timestamp,
-                    optimal_action=optimal_action,
-                    actual_outcome=reward,
-                    market_state_before=initial_state,
-                    market_state_after=final_state,
-                    confidence_should_have_been=optimal_confidence
             )
-                self.perfect_moves.append(perfect_move)
             
-            logger.info(f"Marked perfect move for {action.symbol}: {optimal_action} with confidence {optimal_confidence:.3f}")
+            # Calculate reward based on trade outcome
+            reward = self._calculate_sensitivity_reward(completed_trade)
             
-        except Exception as e:
+            # Determine optimal sensitivity action based on outcome
-            logger.error(f"Error marking perfect move: {e}")
+            optimal_sensitivity = self._determine_optimal_sensitivity(completed_trade)
             
-    def get_recent_perfect_moves(self, limit: int = 10) -> List[PerfectMove]:
+            # Create learning experience
-        """Get recent perfect moves for display/monitoring"""
+            learning_case = {
-        return list(self.perfect_moves)[-limit:]
+                'state': entry_state,
-    
+                'action': completed_trade['sensitivity_level_used'],
-    async def queue_action_for_evaluation(self, action: TradingAction):
+                'reward': reward,
-        """Queue a trading action for future RL evaluation"""
+                'next_state': exit_state,
-        try:
+                'done': True,  # Trade is completed
-            # Get current market state
+                'optimal_action': optimal_sensitivity,
-            market_states = await self._get_all_market_states_universal(self.universal_adapter.get_universal_data_stream())
+                'trade_outcome': completed_trade['pnl_pct'],
-            if action.symbol in market_states:
+                'trade_duration': completed_trade['duration'],
-                evaluation_item = {
+                'symbol': completed_trade['symbol']
-                    'action': action,
-                    'market_state_before': market_states[action.symbol],
-                    'timestamp': datetime.now()
             }
-                self.rl_evaluation_queue.append(evaluation_item)
-                logger.debug(f"Queued action for RL evaluation: {action.action} {action.symbol}")
-        except Exception as e:
-            logger.error(f"Error queuing action for evaluation: {e}")
             
-    def get_perfect_moves_for_training(self, symbol: str = None, timeframe: str = None, 
+            self.sensitivity_learning_queue.append(learning_case)
-                                     limit: int = 1000) -> List[PerfectMove]:
-        """Get perfect moves for CNN training"""
-        moves = list(self.perfect_moves)
             
-        # Filter by symbol if specified
+            # Train DQN if we have enough cases
-        if symbol:
+            if len(self.sensitivity_learning_queue) >= 32:  # Batch size
-            moves = [move for move in moves if move.symbol == symbol]
+                self._train_sensitivity_dqn()
             
-        # Filter by timeframe if specified
+            logger.info(f"Created sensitivity learning case: reward={reward:.3f}, optimal_sensitivity={optimal_sensitivity}")
-        if timeframe:
-            moves = [move for move in moves if move.timeframe == timeframe]
-        
-        return moves[-limit:]  # Return most recent moves
-    
-    # Helper methods for market analysis using universal data
-    def _calculate_volatility_from_universal(self, symbol: str, universal_stream: UniversalDataStream) -> float:
-        """Calculate current volatility for symbol using universal data"""
-        try:
-            if symbol == 'ETH/USDT' and len(universal_stream.eth_ticks) > 10:
-                # Calculate volatility from tick data
-                prices = universal_stream.eth_ticks[-10:, 4]  # Last 10 close prices
-                returns = np.diff(prices) / prices[:-1]
-                volatility = np.std(returns) * np.sqrt(86400)  # Annualized volatility
-                return float(volatility)
-            elif symbol == 'BTC/USDT' and len(universal_stream.btc_ticks) > 10:
-                # Calculate volatility from BTC tick data
-                prices = universal_stream.btc_ticks[-10:, 4]  # Last 10 close prices
-                returns = np.diff(prices) / prices[:-1]
-                volatility = np.std(returns) * np.sqrt(86400)  # Annualized volatility
-                return float(volatility)
-        except Exception as e:
-            logger.error(f"Error calculating volatility from universal data: {e}")
-        
-        return 0.02  # Default 2% volatility
-    
-    def _get_current_volume_from_universal(self, symbol: str, universal_stream: UniversalDataStream) -> float:
-        """Get current volume ratio compared to average using universal data"""
-        try:
-            if symbol == 'ETH/USDT':
-                # Use 1m data for volume analysis
-                if len(universal_stream.eth_1m) > 10:
-                    volumes = universal_stream.eth_1m[-10:, 5]  # Last 10 volume values
-                    current_volume = universal_stream.eth_1m[-1, 5]
-                    avg_volume = np.mean(volumes[:-1])
-                    if avg_volume > 0:
-                        return float(current_volume / avg_volume)
-            elif symbol == 'BTC/USDT':
-                # Use BTC tick data for volume analysis
-                if len(universal_stream.btc_ticks) > 10:
-                    volumes = universal_stream.btc_ticks[-10:, 5]  # Last 10 volume values
-                    current_volume = universal_stream.btc_ticks[-1, 5]
-                    avg_volume = np.mean(volumes[:-1])
-                    if avg_volume > 0:
-                        return float(current_volume / avg_volume)
-        except Exception as e:
-            logger.error(f"Error calculating volume from universal data: {e}")
-        
-        return 1.0  # Normal volume
-    
-    def _calculate_trend_strength_from_universal(self, symbol: str, universal_stream: UniversalDataStream) -> float:
-        """Calculate trend strength using universal data"""
-        try:
-            if symbol == 'ETH/USDT':
-                # Use multiple timeframes to determine trend strength
-                trend_scores = []
-                
-                # Check 1m trend
-                if len(universal_stream.eth_1m) > 20:
-                    prices = universal_stream.eth_1m[-20:, 4]  # Last 20 close prices
-                    slope = np.polyfit(range(len(prices)), prices, 1)[0]
-                    trend_scores.append(abs(slope) / np.mean(prices))
-                
-                # Check 1h trend
-                if len(universal_stream.eth_1h) > 10:
-                    prices = universal_stream.eth_1h[-10:, 4]  # Last 10 close prices
-                    slope = np.polyfit(range(len(prices)), prices, 1)[0]
-                    trend_scores.append(abs(slope) / np.mean(prices))
-                
-                if trend_scores:
-                    return float(np.mean(trend_scores))
-                    
-            elif symbol == 'BTC/USDT':
-                # Use BTC tick data for trend analysis
-                if len(universal_stream.btc_ticks) > 20:
-                    prices = universal_stream.btc_ticks[-20:, 4]  # Last 20 close prices
-                    slope = np.polyfit(range(len(prices)), prices, 1)[0]
-                    return float(abs(slope) / np.mean(prices))
             
         except Exception as e:
-            logger.error(f"Error calculating trend strength from universal data: {e}")
+            logger.error(f"Error creating sensitivity learning case: {e}")
     
-        return 0.5  # Moderate trend
+    def _market_state_to_sensitivity_state(self, market_state: Dict[str, float], current_sensitivity: int) -> np.ndarray:
-    
+        """Convert market state to DQN state vector for sensitivity learning"""
-    def _determine_market_regime_from_universal(self, symbol: str, universal_stream: UniversalDataStream) -> str:
-        """Determine current market regime using universal data"""
         try:
-            if symbol == 'ETH/USDT':
+            # Create state vector with market features + current sensitivity
-                # Analyze volatility and trend from multiple timeframes
+            state_features = [
-                volatility = self._calculate_volatility_from_universal(symbol, universal_stream)
+                market_state.get('volatility', 2.0) / 10.0,  # Normalize volatility
-                trend_strength = self._calculate_trend_strength_from_universal(symbol, universal_stream)
+                market_state.get('momentum_5', 0.0) / 5.0,   # Normalize momentum
+                market_state.get('volume_ratio', 1.0),       # Volume ratio
+                market_state.get('rsi', 50.0) / 100.0,       # Normalize RSI
+                market_state.get('macd_signal', 0.0) / 2.0,  # Normalize MACD
+                market_state.get('bb_position', 0.5),        # BB position (already 0-1)
+                market_state.get('price_change_1', 0.0) * 100,  # Recent price changes
+                market_state.get('price_change_2', 0.0) * 100,
+                market_state.get('price_change_3', 0.0) * 100,
+                market_state.get('price_change_4', 0.0) * 100,
+                market_state.get('price_change_5', 0.0) * 100,
+                current_sensitivity / 4.0,  # Normalize current sensitivity (0-4 -> 0-1)
+            ]
             
-                # Determine regime based on volatility and trend
+            # Add recent performance metrics
-                if volatility > 0.05:  # High volatility
+            if len(self.completed_trades) > 0:
-                    return 'volatile'
+                recent_trades = list(self.completed_trades)[-10:]  # Last 10 trades
-                elif trend_strength > 0.002:  # Strong trend
+                avg_pnl = np.mean([t['pnl_pct'] for t in recent_trades])
-                    return 'trending'
+                win_rate = len([t for t in recent_trades if t['pnl_pct'] > 0]) / len(recent_trades)
+                avg_duration = np.mean([t['duration'] for t in recent_trades]) / 3600  # Normalize to hours
             else:
-                    return 'ranging'
+                avg_pnl = 0.0
+                win_rate = 0.5
+                avg_duration = 0.5
             
-            elif symbol == 'BTC/USDT':
+            state_features.extend([
-                # Analyze BTC regime
+                avg_pnl * 10,  # Recent average P&L
-                volatility = self._calculate_volatility_from_universal(symbol, universal_stream)
+                win_rate,      # Recent win rate
-                
+                avg_duration,  # Recent average duration
-                if volatility > 0.04:  # High volatility for BTC
-                    return 'volatile'
-                else:
-                    return 'trending'  # Default for BTC
-                    
-        except Exception as e:
-            logger.error(f"Error determining market regime from universal data: {e}")
-        
-        return 'trending'  # Default regime
-    
-    # Legacy helper methods (kept for compatibility)
-    def _calculate_volatility(self, symbol: str) -> float:
-        """Calculate current volatility for symbol (legacy method)"""
-        return 0.02  # 2% default volatility
-    
-    def _get_current_volume(self, symbol: str) -> float:
-        """Get current volume ratio compared to average (legacy method)"""
-        return 1.0  # Normal volume
-    
-    def _calculate_trend_strength(self, symbol: str) -> float:
-        """Calculate trend strength (legacy method)"""
-        return 0.5  # Moderate trend
-    
-    def _determine_market_regime(self, symbol: str) -> str:
-        """Determine current market regime (legacy method)"""
-        return 'trending'  # Default to trending
-    
-    def _get_symbol_correlation(self, symbol: str) -> Dict[str, float]:
-        """Get correlations with other symbols"""
-        correlations = {}
-        for other_symbol in self.symbols:
-            if other_symbol != symbol:
-                correlations[other_symbol] = self.symbol_correlation_matrix.get((symbol, other_symbol), 0.0)
-        return correlations
-    
-    def _calculate_position_size(self, symbol: str, action: str, confidence: float) -> float:
-        """Calculate position size based on confidence and risk management"""
-        base_size = 0.02  # 2% of portfolio
-        confidence_multiplier = confidence  # Scale by confidence
-        max_size = 0.05  # 5% maximum
-        
-        return min(base_size * confidence_multiplier, max_size)
-    
-    def _market_state_to_rl_state(self, market_state: MarketState) -> np.ndarray:
-        """Convert market state to RL state vector"""
-        # Combine features from all timeframes into a single state vector
-        state_components = []
-        
-        # Add price features
-        state_components.extend([
-            market_state.volatility,
-            market_state.volume,
-            market_state.trend_strength
             ])
             
-        # Add flattened features from each timeframe
+            # Pad or truncate to exact state size
-        for timeframe in sorted(market_state.features.keys()):
+            while len(state_features) < self.sensitivity_state_size:
-            features = market_state.features[timeframe]
+                state_features.append(0.0)
-            if features is not None:
-                # Take the last row (most recent) and flatten
-                latest_features = features[-1] if len(features.shape) > 1 else features
-                state_components.extend(latest_features.flatten())
             
-        return np.array(state_components, dtype=np.float32)
+            state_features = state_features[:self.sensitivity_state_size]
+            
+            return np.array(state_features, dtype=np.float32)
+            
+        except Exception as e:
+            logger.error(f"Error converting market state to sensitivity state: {e}")
+            return np.zeros(self.sensitivity_state_size, dtype=np.float32)
+    
+    def _calculate_sensitivity_reward(self, completed_trade: Dict[str, Any]) -> float:
+        """Calculate reward for sensitivity learning based on trade outcome"""
+        try:
+            pnl_pct = completed_trade['pnl_pct']
+            duration = completed_trade['duration']
+            
+            # Base reward from P&L
+            base_reward = pnl_pct * 10  # Scale P&L percentage
+            
+            # Duration penalty/bonus
+            if duration < 300:  # Less than 5 minutes - too quick
+                duration_factor = 0.8
+            elif duration < 1800:  # Less than 30 minutes - good for scalping
+                duration_factor = 1.2
+            elif duration < 3600:  # Less than 1 hour - acceptable
+                duration_factor = 1.0
+            else:  # More than 1 hour - too slow for scalping
+                duration_factor = 0.7
+            
+            # Confidence factor - reward appropriate confidence levels
+            entry_conf = completed_trade['entry_confidence']
+            exit_conf = completed_trade['exit_confidence']
+            
+            if pnl_pct > 0:  # Winning trade
+                # Reward high entry confidence and appropriate exit confidence
+                conf_factor = (entry_conf + exit_conf) / 2
+            else:  # Losing trade
+                # Reward quick exit (high exit confidence for losses)
+                conf_factor = exit_conf
+            
+            # Calculate final reward
+            final_reward = base_reward * duration_factor * conf_factor
+            
+            # Clip reward to reasonable range
+            final_reward = np.clip(final_reward, -2.0, 2.0)
+            
+            return float(final_reward)
+            
+        except Exception as e:
+            logger.error(f"Error calculating sensitivity reward: {e}")
+            return 0.0
+    
+    def _determine_optimal_sensitivity(self, completed_trade: Dict[str, Any]) -> int:
+        """Determine optimal sensitivity level based on trade outcome"""
+        try:
+            pnl_pct = completed_trade['pnl_pct']
+            duration = completed_trade['duration']
+            current_sensitivity = completed_trade['sensitivity_level_used']
+            
+            # If trade was profitable and quick, current sensitivity was good
+            if pnl_pct > 0.01 and duration < 1800:  # >1% profit in <30 min
+                return current_sensitivity
+            
+            # If trade was very profitable, could have been more aggressive
+            if pnl_pct > 0.02:  # >2% profit
+                return min(4, current_sensitivity + 1)  # Increase sensitivity
+            
+            # If trade was a small loss, might need more sensitivity
+            if -0.01 < pnl_pct < 0:  # Small loss
+                return min(4, current_sensitivity + 1)  # Increase sensitivity
+            
+            # If trade was a big loss, need less sensitivity
+            if pnl_pct < -0.02:  # >2% loss
+                return max(0, current_sensitivity - 1)  # Decrease sensitivity
+            
+            # If trade took too long, need more sensitivity
+            if duration > 3600:  # >1 hour
+                return min(4, current_sensitivity + 1)  # Increase sensitivity
+            
+            # Default: keep current sensitivity
+            return current_sensitivity
+            
+        except Exception as e:
+            logger.error(f"Error determining optimal sensitivity: {e}")
+            return 2  # Default to medium
+    
+    def _train_sensitivity_dqn(self):
+        """Train the DQN agent for sensitivity learning"""
+        try:
+            # Initialize DQN agent if not already done
+            if self.sensitivity_dqn_agent is None:
+                self._initialize_sensitivity_dqn()
+            
+            if self.sensitivity_dqn_agent is None:
+                return
+            
+            # Get batch of learning cases
+            batch_size = min(32, len(self.sensitivity_learning_queue))
+            if batch_size < 8:  # Need minimum batch size
+                return
+            
+            # Sample random batch
+            batch_indices = np.random.choice(len(self.sensitivity_learning_queue), batch_size, replace=False)
+            batch = [self.sensitivity_learning_queue[i] for i in batch_indices]
+            
+            # Train the DQN agent
+            for case in batch:
+                self.sensitivity_dqn_agent.remember(
+                    state=case['state'],
+                    action=case['action'],
+                    reward=case['reward'],
+                    next_state=case['next_state'],
+                    done=case['done']
+                )
+            
+            # Perform replay training
+            loss = self.sensitivity_dqn_agent.replay()
+            
+            if loss is not None:
+                logger.info(f"Sensitivity DQN training completed. Loss: {loss:.4f}")
+                
+                # Update current sensitivity level based on recent performance
+                self._update_current_sensitivity_level()
+            
+        except Exception as e:
+            logger.error(f"Error training sensitivity DQN: {e}")
+    
+    def _initialize_sensitivity_dqn(self):
+        """Initialize the DQN agent for sensitivity learning"""
+        try:
+            # Try to import DQN agent
+            from NN.models.dqn_agent import DQNAgent
+            
+            # Create DQN agent for sensitivity learning
+            self.sensitivity_dqn_agent = DQNAgent(
+                state_shape=(self.sensitivity_state_size,),
+                n_actions=self.sensitivity_action_space,
+                learning_rate=0.001,
+                gamma=0.95,
+                epsilon=0.3,  # Lower epsilon for more exploitation
+                epsilon_min=0.05,
+                epsilon_decay=0.995,
+                buffer_size=1000,
+                batch_size=32,
+                target_update=10
+            )
+            
+            logger.info("Sensitivity DQN agent initialized successfully")
+            
+        except Exception as e:
+            logger.error(f"Error initializing sensitivity DQN agent: {e}")
+            self.sensitivity_dqn_agent = None
+    
+    def _update_current_sensitivity_level(self):
+        """Update current sensitivity level using trained DQN"""
+        try:
+            if self.sensitivity_dqn_agent is None:
+                return
+            
+            # Get current market state
+            current_market_state = self._get_current_market_state_for_sensitivity('ETH/USDT')  # Use ETH as primary
+            current_state = self._market_state_to_sensitivity_state(current_market_state, self.current_sensitivity_level)
+            
+            # Get action from DQN (without exploration for production use)
+            action = self.sensitivity_dqn_agent.act(current_state, explore=False)
+            
+            # Update sensitivity level if it changed
+            if action != self.current_sensitivity_level:
+                old_level = self.current_sensitivity_level
+                self.current_sensitivity_level = action
+                
+                # Update thresholds based on new sensitivity level
+                self._update_thresholds_from_sensitivity()
+                
+                logger.info(f"Sensitivity level updated: {self.sensitivity_levels[old_level]['name']} -> {self.sensitivity_levels[action]['name']}")
+            
+        except Exception as e:
+            logger.error(f"Error updating current sensitivity level: {e}")
+    
+    def _update_thresholds_from_sensitivity(self):
+        """Update confidence thresholds based on current sensitivity level"""
+        try:
+            sensitivity_config = self.sensitivity_levels[self.current_sensitivity_level]
+            
+            # Get base thresholds from config
+            base_open_threshold = self.config.orchestrator.get('confidence_threshold', 0.6)
+            base_close_threshold = self.config.orchestrator.get('confidence_threshold_close', 0.25)
+            
+            # Apply sensitivity multipliers
+            self.confidence_threshold_open = base_open_threshold * sensitivity_config['open_threshold_multiplier']
+            self.confidence_threshold_close = base_close_threshold * sensitivity_config['close_threshold_multiplier']
+            
+            # Ensure thresholds stay within reasonable bounds
+            self.confidence_threshold_open = np.clip(self.confidence_threshold_open, 0.3, 0.9)
+            self.confidence_threshold_close = np.clip(self.confidence_threshold_close, 0.1, 0.6)
+            
+            logger.info(f"Updated thresholds - Open: {self.confidence_threshold_open:.3f}, Close: {self.confidence_threshold_close:.3f}")
+            
+        except Exception as e:
+            logger.error(f"Error updating thresholds from sensitivity: {e}")
+    
+    def _get_current_open_threshold(self) -> float:
+        """Get current opening threshold"""
+        return self.confidence_threshold_open
+    
+    def _get_current_close_threshold(self) -> float:
+        """Get current closing threshold"""
+        return self.confidence_threshold_close
 
     def process_realtime_features(self, feature_dict: Dict[str, Any]):
         """Process real-time tick features from the tick processor"""

test_enhanced_improvements.py

@@ -0,0 +1,243 @@
+#!/usr/bin/env python3
+"""
+Test Enhanced Trading System Improvements
+
+This script tests:
+1. Color-coded position display ([LONG] green, [SHORT] red)
+2. Enhanced model training detection and retrospective learning
+3. Lower confidence thresholds for closing positions (0.25 vs 0.6 for opening)
+4. Perfect opportunity detection and learning
+"""
+
+import asyncio
+import logging
+import time
+from datetime import datetime, timedelta
+from core.data_provider import DataProvider
+from core.enhanced_orchestrator import EnhancedTradingOrchestrator, TradingAction
+from web.scalping_dashboard import RealTimeScalpingDashboard, TradingSession
+
+# Setup logging
+logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(name)s - %(levelname)s - %(message)s')
+logger = logging.getLogger(__name__)
+
+def test_color_coded_positions():
+    """Test color-coded position display functionality"""
+    logger.info("=== Testing Color-Coded Position Display ===")
+    
+    # Create trading session
+    session = TradingSession()
+    
+    # Simulate some positions
+    session.positions = {
+        'ETH/USDT': {
+            'side': 'LONG',
+            'size': 0.1,
+            'entry_price': 2558.15
+        },
+        'BTC/USDT': {
+            'side': 'SHORT',
+            'size': 0.05,
+            'entry_price': 45123.45
+        }
+    }
+    
+    logger.info("Created test positions:")
+    logger.info(f"ETH/USDT: LONG 0.1 @ $2558.15")
+    logger.info(f"BTC/USDT: SHORT 0.05 @ $45123.45")
+    
+    # Test position display logic (simulating dashboard logic)
+    live_prices = {'ETH/USDT': 2565.30, 'BTC/USDT': 45050.20}
+    
+    for symbol, pos in session.positions.items():
+        side = pos['side']
+        size = pos['size']
+        entry_price = pos['entry_price']
+        current_price = live_prices.get(symbol, entry_price)
+        
+        # Calculate unrealized P&L
+        if side == 'LONG':
+            unrealized_pnl = (current_price - entry_price) * size
+            color_class = "text-success"  # Green for LONG
+            side_display = "[LONG]"
+        else:  # SHORT
+            unrealized_pnl = (entry_price - current_price) * size
+            color_class = "text-danger"   # Red for SHORT
+            side_display = "[SHORT]"
+        
+        position_text = f"{side_display} {size:.3f} @ ${entry_price:.2f} | P&L: ${unrealized_pnl:+.2f}"
+        logger.info(f"Position Display: {position_text} (Color: {color_class})")
+    
+    logger.info("✅ Color-coded position display test completed")
+
+def test_confidence_thresholds():
+    """Test different confidence thresholds for opening vs closing"""
+    logger.info("=== Testing Confidence Thresholds ===")
+    
+    # Create orchestrator
+    data_provider = DataProvider()
+    orchestrator = EnhancedTradingOrchestrator(data_provider)
+    
+    logger.info(f"Opening threshold: {orchestrator.confidence_threshold_open}")
+    logger.info(f"Closing threshold: {orchestrator.confidence_threshold_close}")
+    
+    # Test opening action with medium confidence
+    test_confidence = 0.45
+    logger.info(f"\nTesting opening action with confidence {test_confidence}")
+    
+    if test_confidence >= orchestrator.confidence_threshold_open:
+        logger.info("✅ Would OPEN position (confidence above opening threshold)")
+    else:
+        logger.info("❌ Would NOT open position (confidence below opening threshold)")
+    
+    # Test closing action with same confidence
+    logger.info(f"Testing closing action with confidence {test_confidence}")
+    
+    if test_confidence >= orchestrator.confidence_threshold_close:
+        logger.info("✅ Would CLOSE position (confidence above closing threshold)")
+    else:
+        logger.info("❌ Would NOT close position (confidence below closing threshold)")
+    
+    logger.info("✅ Confidence threshold test completed")
+
+def test_retrospective_learning():
+    """Test retrospective learning and perfect opportunity detection"""
+    logger.info("=== Testing Retrospective Learning ===")
+    
+    # Create orchestrator
+    data_provider = DataProvider()
+    orchestrator = EnhancedTradingOrchestrator(data_provider)
+    
+    # Simulate perfect moves
+    from core.enhanced_orchestrator import PerfectMove
+    
+    perfect_move = PerfectMove(
+        symbol='ETH/USDT',
+        timeframe='1m',
+        timestamp=datetime.now(),
+        optimal_action='BUY',
+        actual_outcome=0.025,  # 2.5% price increase
+        market_state_before=None,
+        market_state_after=None,
+        confidence_should_have_been=0.85
+    )
+    
+    orchestrator.perfect_moves.append(perfect_move)
+    orchestrator.retrospective_learning_active = True
+    
+    logger.info(f"Added perfect move: {perfect_move.optimal_action} {perfect_move.symbol}")
+    logger.info(f"Outcome: {perfect_move.actual_outcome*100:+.2f}%")
+    logger.info(f"Confidence should have been: {perfect_move.confidence_should_have_been:.3f}")
+    
+    # Test performance metrics
+    metrics = orchestrator.get_performance_metrics()
+    retro_metrics = metrics['retrospective_learning']
+    
+    logger.info(f"Retrospective learning active: {retro_metrics['active']}")
+    logger.info(f"Recent perfect moves: {retro_metrics['perfect_moves_recent']}")
+    logger.info(f"Average confidence needed: {retro_metrics['avg_confidence_needed']:.3f}")
+    
+    logger.info("✅ Retrospective learning test completed")
+
+async def test_tick_pattern_detection():
+    """Test tick pattern detection for violent moves"""
+    logger.info("=== Testing Tick Pattern Detection ===")
+    
+    # Create orchestrator
+    data_provider = DataProvider()
+    orchestrator = EnhancedTradingOrchestrator(data_provider)
+    
+    # Simulate violent tick
+    from core.tick_aggregator import RawTick
+    
+    violent_tick = RawTick(
+        timestamp=datetime.now(),
+        price=2560.0,
+        volume=1000.0,
+        quantity=0.5,
+        side='buy',
+        trade_id='test123',
+        time_since_last=25.0,  # Very fast tick (25ms)
+        price_change=5.0,      # $5 price jump
+        volume_intensity=3.5   # High volume
+    )
+    
+    # Add symbol attribute for testing
+    violent_tick.symbol = 'ETH/USDT'
+    
+    logger.info(f"Simulating violent tick:")
+    logger.info(f"Price change: ${violent_tick.price_change:+.2f}")
+    logger.info(f"Time since last: {violent_tick.time_since_last:.0f}ms")
+    logger.info(f"Volume intensity: {violent_tick.volume_intensity:.1f}x")
+    
+    # Process the tick
+    orchestrator._handle_raw_tick(violent_tick)
+    
+    # Check if perfect move was created
+    if orchestrator.perfect_moves:
+        latest_move = orchestrator.perfect_moves[-1]
+        logger.info(f"✅ Perfect move detected: {latest_move.optimal_action}")
+        logger.info(f"Confidence: {latest_move.confidence_should_have_been:.3f}")
+    else:
+        logger.info("❌ No perfect move detected")
+    
+    logger.info("✅ Tick pattern detection test completed")
+
+def test_dashboard_integration():
+    """Test dashboard integration with new features"""
+    logger.info("=== Testing Dashboard Integration ===")
+    
+    # Create components
+    data_provider = DataProvider()
+    orchestrator = EnhancedTradingOrchestrator(data_provider)
+    
+    # Test model training status
+    metrics = orchestrator.get_performance_metrics()
+    
+    logger.info("Model Training Metrics:")
+    logger.info(f"Perfect moves: {metrics['perfect_moves']}")
+    logger.info(f"RL queue size: {metrics['rl_queue_size']}")
+    logger.info(f"Retrospective learning: {metrics['retrospective_learning']}")
+    logger.info(f"Position tracking: {metrics['position_tracking']}")
+    logger.info(f"Thresholds: {metrics['thresholds']}")
+    
+    logger.info("✅ Dashboard integration test completed")
+
+async def main():
+    """Run all tests"""
+    logger.info("🚀 Starting Enhanced Trading System Tests")
+    logger.info("=" * 60)
+    
+    try:
+        # Run tests
+        test_color_coded_positions()
+        print()
+        
+        test_confidence_thresholds()
+        print()
+        
+        test_retrospective_learning()
+        print()
+        
+        await test_tick_pattern_detection()
+        print()
+        
+        test_dashboard_integration()
+        print()
+        
+        logger.info("=" * 60)
+        logger.info("🎉 All tests completed successfully!")
+        logger.info("Key improvements verified:")
+        logger.info("✅ Color-coded positions ([LONG] green, [SHORT] red)")
+        logger.info("✅ Lower closing thresholds (0.25 vs 0.6)")
+        logger.info("✅ Retrospective learning on perfect opportunities")
+        logger.info("✅ Enhanced model training detection")
+        logger.info("✅ Violent move pattern detection")
+        
+    except Exception as e:
+        logger.error(f"❌ Test failed: {e}")
+        import traceback
+        logger.error(traceback.format_exc())
+
+if __name__ == "__main__":
+    asyncio.run(main()) 

test_sensitivity_learning.py

@@ -0,0 +1,372 @@
+#!/usr/bin/env python3
+"""
+Test DQN RL-based Sensitivity Learning and 300s Data Preloading
+
+This script tests:
+1. DQN RL-based sensitivity learning from completed trades
+2. 300s data preloading on first load
+3. Dynamic threshold adjustment based on sensitivity levels
+4. Color-coded position display integration
+5. Enhanced model training status with sensitivity info
+
+Usage:
+    python test_sensitivity_learning.py
+"""
+
+import asyncio
+import logging
+import time
+import numpy as np
+from datetime import datetime, timedelta
+from core.data_provider import DataProvider
+from core.enhanced_orchestrator import EnhancedTradingOrchestrator, TradingAction
+from web.scalping_dashboard import RealTimeScalpingDashboard
+from NN.models.dqn_agent import DQNAgent
+
+# Setup logging
+logging.basicConfig(
+    level=logging.INFO,
+    format='%(asctime)s - %(name)s - %(levelname)s - %(message)s'
+)
+logger = logging.getLogger(__name__)
+
+class SensitivityLearningTester:
+    """Test class for sensitivity learning features"""
+    
+    def __init__(self):
+        self.data_provider = DataProvider()
+        self.orchestrator = EnhancedTradingOrchestrator(self.data_provider)
+        self.dashboard = None
+        
+    async def test_300s_data_preloading(self):
+        """Test 300s data preloading functionality"""
+        logger.info("=== Testing 300s Data Preloading ===")
+        
+        # Test preloading for all symbols and timeframes
+        start_time = time.time()
+        preload_results = self.data_provider.preload_all_symbols_data(['1s', '1m', '5m', '15m', '1h'])
+        end_time = time.time()
+        
+        logger.info(f"Preloading completed in {end_time - start_time:.2f} seconds")
+        
+        # Verify results
+        total_pairs = 0
+        successful_pairs = 0
+        
+        for symbol, timeframe_results in preload_results.items():
+            for timeframe, success in timeframe_results.items():
+                total_pairs += 1
+                if success:
+                    successful_pairs += 1
+                    
+                    # Verify data was actually loaded
+                    data = self.data_provider.get_historical_data(symbol, timeframe, limit=50)
+                    if data is not None and len(data) > 0:
+                        logger.info(f"✅ {symbol} {timeframe}: {len(data)} candles loaded")
+                    else:
+                        logger.warning(f"❌ {symbol} {timeframe}: No data despite success flag")
+                else:
+                    logger.warning(f"❌ {symbol} {timeframe}: Failed to preload")
+        
+        success_rate = (successful_pairs / total_pairs) * 100 if total_pairs > 0 else 0
+        logger.info(f"Preloading success rate: {success_rate:.1f}% ({successful_pairs}/{total_pairs})")
+        
+        return success_rate > 80  # Consider test passed if >80% success rate
+    
+    def test_sensitivity_learning_initialization(self):
+        """Test sensitivity learning system initialization"""
+        logger.info("=== Testing Sensitivity Learning Initialization ===")
+        
+        # Check if sensitivity learning is enabled
+        if hasattr(self.orchestrator, 'sensitivity_learning_enabled'):
+            logger.info(f"✅ Sensitivity learning enabled: {self.orchestrator.sensitivity_learning_enabled}")
+        else:
+            logger.warning("❌ Sensitivity learning not found in orchestrator")
+            return False
+        
+        # Check sensitivity levels configuration
+        if hasattr(self.orchestrator, 'sensitivity_levels'):
+            levels = self.orchestrator.sensitivity_levels
+            logger.info(f"✅ Sensitivity levels configured: {len(levels)} levels")
+            for level, config in levels.items():
+                logger.info(f"  Level {level}: {config['name']} - Open: {config['open_threshold_multiplier']:.2f}, Close: {config['close_threshold_multiplier']:.2f}")
+        else:
+            logger.warning("❌ Sensitivity levels not configured")
+            return False
+        
+        # Check DQN agent initialization
+        if hasattr(self.orchestrator, 'sensitivity_dqn_agent'):
+            if self.orchestrator.sensitivity_dqn_agent is not None:
+                logger.info("✅ DQN agent initialized")
+                stats = self.orchestrator.sensitivity_dqn_agent.get_stats()
+                logger.info(f"  Device: {stats['device']}")
+                logger.info(f"  Memory size: {stats['memory_size']}")
+                logger.info(f"  Epsilon: {stats['epsilon']:.3f}")
+            else:
+                logger.info("⏳ DQN agent not yet initialized (will be created on first use)")
+        
+        # Check learning queues
+        if hasattr(self.orchestrator, 'sensitivity_learning_queue'):
+            logger.info(f"✅ Sensitivity learning queue initialized: {len(self.orchestrator.sensitivity_learning_queue)} items")
+        
+        if hasattr(self.orchestrator, 'completed_trades'):
+            logger.info(f"✅ Completed trades tracking initialized: {len(self.orchestrator.completed_trades)} trades")
+        
+        if hasattr(self.orchestrator, 'active_trades'):
+            logger.info(f"✅ Active trades tracking initialized: {len(self.orchestrator.active_trades)} active")
+        
+        return True
+    
+    def simulate_trading_scenario(self):
+        """Simulate a trading scenario to test sensitivity learning"""
+        logger.info("=== Simulating Trading Scenario ===")
+        
+        # Simulate some trades to test the learning system
+        test_trades = [
+            {
+                'symbol': 'ETH/USDT',
+                'action': 'BUY',
+                'price': 2500.0,
+                'confidence': 0.7,
+                'timestamp': datetime.now() - timedelta(minutes=10)
+            },
+            {
+                'symbol': 'ETH/USDT', 
+                'action': 'SELL',
+                'price': 2510.0,
+                'confidence': 0.6,
+                'timestamp': datetime.now() - timedelta(minutes=5)
+            },
+            {
+                'symbol': 'ETH/USDT',
+                'action': 'BUY',
+                'price': 2505.0,
+                'confidence': 0.8,
+                'timestamp': datetime.now() - timedelta(minutes=3)
+            },
+            {
+                'symbol': 'ETH/USDT',
+                'action': 'SELL',
+                'price': 2495.0,
+                'confidence': 0.9,
+                'timestamp': datetime.now()
+            }
+        ]
+        
+        # Process each trade through the orchestrator
+        for i, trade_data in enumerate(test_trades):
+            action = TradingAction(
+                symbol=trade_data['symbol'],
+                action=trade_data['action'],
+                quantity=0.1,
+                confidence=trade_data['confidence'],
+                price=trade_data['price'],
+                timestamp=trade_data['timestamp'],
+                reasoning={'test': f'simulated_trade_{i}'},
+                timeframe_analysis=[]
+            )
+            
+            # Update position tracking (this should trigger sensitivity learning)
+            self.orchestrator._update_position_tracking(trade_data['symbol'], action)
+            
+            logger.info(f"Processed trade {i+1}: {trade_data['action']} @ ${trade_data['price']:.2f}")
+        
+        # Check if learning cases were created
+        if hasattr(self.orchestrator, 'sensitivity_learning_queue'):
+            queue_size = len(self.orchestrator.sensitivity_learning_queue)
+            logger.info(f"✅ Learning queue now has {queue_size} cases")
+        
+        if hasattr(self.orchestrator, 'completed_trades'):
+            completed_count = len(self.orchestrator.completed_trades)
+            logger.info(f"✅ Completed trades: {completed_count}")
+        
+        return True
+    
+    def test_threshold_adjustment(self):
+        """Test dynamic threshold adjustment based on sensitivity"""
+        logger.info("=== Testing Threshold Adjustment ===")
+        
+        # Test different sensitivity levels
+        for level in range(5):  # 0-4 sensitivity levels
+            if hasattr(self.orchestrator, 'current_sensitivity_level'):
+                self.orchestrator.current_sensitivity_level = level
+                
+                if hasattr(self.orchestrator, '_update_thresholds_from_sensitivity'):
+                    self.orchestrator._update_thresholds_from_sensitivity()
+                    
+                    open_threshold = getattr(self.orchestrator, 'confidence_threshold_open', 0.6)
+                    close_threshold = getattr(self.orchestrator, 'confidence_threshold_close', 0.25)
+                    
+                    logger.info(f"Level {level}: Open={open_threshold:.3f}, Close={close_threshold:.3f}")
+        
+        return True
+    
+    def test_dashboard_integration(self):
+        """Test dashboard integration with sensitivity learning"""
+        logger.info("=== Testing Dashboard Integration ===")
+        
+        try:
+            # Create dashboard instance
+            self.dashboard = RealTimeScalpingDashboard(
+                data_provider=self.data_provider,
+                orchestrator=self.orchestrator
+            )
+            
+            # Test sensitivity learning info retrieval
+            sensitivity_info = self.dashboard._get_sensitivity_learning_info()
+            
+            logger.info("✅ Dashboard sensitivity info:")
+            logger.info(f"  Level: {sensitivity_info['level_name']}")
+            logger.info(f"  Completed trades: {sensitivity_info['completed_trades']}")
+            logger.info(f"  Learning queue: {sensitivity_info['learning_queue_size']}")
+            logger.info(f"  Open threshold: {sensitivity_info['open_threshold']:.3f}")
+            logger.info(f"  Close threshold: {sensitivity_info['close_threshold']:.3f}")
+            
+            return True
+            
+        except Exception as e:
+            logger.error(f"❌ Dashboard integration test failed: {e}")
+            return False
+    
+    def test_dqn_training_simulation(self):
+        """Test DQN training with simulated data"""
+        logger.info("=== Testing DQN Training Simulation ===")
+        
+        try:
+            # Initialize DQN agent if not already done
+            if not hasattr(self.orchestrator, 'sensitivity_dqn_agent') or self.orchestrator.sensitivity_dqn_agent is None:
+                self.orchestrator._initialize_sensitivity_dqn()
+            
+            if self.orchestrator.sensitivity_dqn_agent is None:
+                logger.warning("❌ Could not initialize DQN agent")
+                return False
+            
+            # Create some mock learning cases
+            for i in range(10):
+                # Create mock market state
+                mock_state = np.random.random(self.orchestrator.sensitivity_state_size)
+                action = np.random.randint(0, self.orchestrator.sensitivity_action_space)
+                reward = np.random.random() - 0.5  # Random reward between -0.5 and 0.5
+                next_state = np.random.random(self.orchestrator.sensitivity_state_size)
+                done = True
+                
+                # Add to learning queue
+                learning_case = {
+                    'state': mock_state,
+                    'action': action,
+                    'reward': reward,
+                    'next_state': next_state,
+                    'done': done,
+                    'optimal_action': action,
+                    'trade_outcome': reward * 0.02,  # Convert to percentage
+                    'trade_duration': 300 + np.random.randint(-100, 100),
+                    'symbol': 'ETH/USDT'
+                }
+                
+                self.orchestrator.sensitivity_learning_queue.append(learning_case)
+            
+            # Trigger training
+            initial_queue_size = len(self.orchestrator.sensitivity_learning_queue)
+            self.orchestrator._train_sensitivity_dqn()
+            
+            logger.info(f"✅ DQN training completed")
+            logger.info(f"  Initial queue size: {initial_queue_size}")
+            logger.info(f"  Final queue size: {len(self.orchestrator.sensitivity_learning_queue)}")
+            
+            # Check agent stats
+            if self.orchestrator.sensitivity_dqn_agent:
+                stats = self.orchestrator.sensitivity_dqn_agent.get_stats()
+                logger.info(f"  Training steps: {stats['training_step']}")
+                logger.info(f"  Memory size: {stats['memory_size']}")
+                logger.info(f"  Epsilon: {stats['epsilon']:.3f}")
+            
+            return True
+            
+        except Exception as e:
+            logger.error(f"❌ DQN training simulation failed: {e}")
+            return False
+    
+    async def run_all_tests(self):
+        """Run all sensitivity learning tests"""
+        logger.info("🚀 Starting Sensitivity Learning Test Suite")
+        logger.info("=" * 60)
+        
+        test_results = {}
+        
+        # Test 1: 300s Data Preloading
+        test_results['preloading'] = await self.test_300s_data_preloading()
+        
+        # Test 2: Sensitivity Learning Initialization
+        test_results['initialization'] = self.test_sensitivity_learning_initialization()
+        
+        # Test 3: Trading Scenario Simulation
+        test_results['trading_simulation'] = self.simulate_trading_scenario()
+        
+        # Test 4: Threshold Adjustment
+        test_results['threshold_adjustment'] = self.test_threshold_adjustment()
+        
+        # Test 5: Dashboard Integration
+        test_results['dashboard_integration'] = self.test_dashboard_integration()
+        
+        # Test 6: DQN Training Simulation
+        test_results['dqn_training'] = self.test_dqn_training_simulation()
+        
+        # Summary
+        logger.info("=" * 60)
+        logger.info("🏁 Test Suite Results:")
+        
+        passed_tests = 0
+        total_tests = len(test_results)
+        
+        for test_name, result in test_results.items():
+            status = "✅ PASSED" if result else "❌ FAILED"
+            logger.info(f"  {test_name}: {status}")
+            if result:
+                passed_tests += 1
+        
+        success_rate = (passed_tests / total_tests) * 100
+        logger.info(f"Overall success rate: {success_rate:.1f}% ({passed_tests}/{total_tests})")
+        
+        if success_rate >= 80:
+            logger.info("🎉 Test suite PASSED! Sensitivity learning system is working correctly.")
+        else:
+            logger.warning("⚠️ Test suite FAILED! Some issues need to be addressed.")
+        
+        return success_rate >= 80
+
+async def main():
+    """Main test function"""
+    tester = SensitivityLearningTester()
+    
+    try:
+        success = await tester.run_all_tests()
+        
+        if success:
+            logger.info("✅ All tests passed! The sensitivity learning system is ready for production.")
+        else:
+            logger.error("❌ Some tests failed. Please review the issues above.")
+        
+        return success
+        
+    except Exception as e:
+        logger.error(f"Test suite failed with exception: {e}")
+        return False
+
+if __name__ == "__main__":
+    # Run the test suite
+    result = asyncio.run(main())
+    
+    if result:
+        print("\n🎯 SENSITIVITY LEARNING SYSTEM READY!")
+        print("Features verified:")
+        print("  ✅ DQN RL-based sensitivity learning from completed trades")
+        print("  ✅ 300s data preloading for faster initial performance")
+        print("  ✅ Dynamic threshold adjustment (lower for closing positions)")
+        print("  ✅ Color-coded position display ([LONG] green, [SHORT] red)")
+        print("  ✅ Enhanced model training status with sensitivity info")
+        print("\nYou can now run the dashboard with these enhanced features!")
+    else:
+        print("\n❌ SOME TESTS FAILED")
+        print("Please review the test output above and fix any issues.")
+    
+    exit(0 if result else 1) 

web/scalping_dashboard.py

@@ -205,6 +205,16 @@ class RealTimeScalpingDashboard:
         logger.info("  4. ETH/USDT 1d")
         logger.info("  5. BTC/USDT ticks (reference)")
         
+        # Preload 300s of data for better initial performance
+        logger.info("PRELOADING 300s OF DATA FOR INITIAL PERFORMANCE:")
+        preload_results = self.data_provider.preload_all_symbols_data(['1s', '1m', '5m', '15m', '1h', '1d'])
+        
+        # Log preload results
+        for symbol, timeframe_results in preload_results.items():
+            for timeframe, success in timeframe_results.items():
+                status = "✅" if success else "❌"
+                logger.info(f"  {status} {symbol} {timeframe}")
+        
         # Test universal data adapter
         try:
             universal_stream = self.orchestrator.universal_adapter.get_universal_data_stream()
@@ -503,6 +513,7 @@ class RealTimeScalpingDashboard:
         logger.info("WebSocket price streaming enabled")
         logger.info(f"Timezone: {self.timezone}")
         logger.info(f"Session Balance: ${self.trading_session.starting_balance:.2f}")
+        logger.info("300s data preloading completed for faster initial performance")
     
     def _setup_layout(self):
         """Setup the ultra-fast real-time dashboard layout"""
@@ -1794,7 +1805,7 @@ class RealTimeScalpingDashboard:
             return fig
     
     def _create_model_training_status(self):
-        """Create enhanced model training progress display with perfect opportunity detection"""
+        """Create enhanced model training progress display with perfect opportunity detection and sensitivity learning"""
         try:
             # Get model training metrics from orchestrator
             if hasattr(self.orchestrator, 'get_performance_metrics'):
@@ -1811,6 +1822,9 @@ class RealTimeScalpingDashboard:
                 is_rl_training = rl_queue_size > 0
                 is_cnn_training = perfect_moves_count > 0
                 
+                # Get sensitivity learning information
+                sensitivity_info = self._get_sensitivity_learning_info()
+                
                 return html.Div([
                     html.Div([
                         html.H6("RL Training", className="text-success" if is_rl_training else "text-warning"),
@@ -1819,7 +1833,7 @@ class RealTimeScalpingDashboard:
                         html.P(f"Queue Size: {rl_queue_size}", className="text-white"),
                         html.P(f"Win Rate: {metrics.get('win_rate', 0)*100:.1f}%", className="text-white"),
                         html.P(f"Actions: {metrics.get('total_actions', 0)}", className="text-white")
-                    ], className="col-md-6"),
+                    ], className="col-md-4"),
                     
                     html.Div([
                         html.H6("CNN Training", className="text-success" if is_cnn_training else "text-warning"),
@@ -1829,7 +1843,17 @@ class RealTimeScalpingDashboard:
                         html.P(f"Confidence: {metrics.get('confidence_threshold', 0.6):.2f}", className="text-white"),
                         html.P(f"Retrospective: {'ON' if recent_perfect_moves else 'OFF'}", 
                                className="text-success" if recent_perfect_moves else "text-muted")
-                    ], className="col-md-6")
+                    ], className="col-md-4"),
+                    
+                    html.Div([
+                        html.H6("DQN Sensitivity", className="text-info"),
+                        html.P(f"Level: {sensitivity_info['level_name']}", 
+                               className="text-info"),
+                        html.P(f"Completed Trades: {sensitivity_info['completed_trades']}", className="text-white"),
+                        html.P(f"Learning Queue: {sensitivity_info['learning_queue_size']}", className="text-white"),
+                        html.P(f"Open: {sensitivity_info['open_threshold']:.3f} | Close: {sensitivity_info['close_threshold']:.3f}", 
+                               className="text-white")
+                    ], className="col-md-4")
                 ], className="row")
             else:
                 return html.Div([
@@ -1842,6 +1866,45 @@ class RealTimeScalpingDashboard:
                 html.P("Error loading model status", className="text-danger")
             ])
     
+    def _get_sensitivity_learning_info(self) -> Dict[str, Any]:
+        """Get sensitivity learning information from orchestrator"""
+        try:
+            if hasattr(self.orchestrator, 'sensitivity_learning_enabled') and self.orchestrator.sensitivity_learning_enabled:
+                current_level = getattr(self.orchestrator, 'current_sensitivity_level', 2)
+                sensitivity_levels = getattr(self.orchestrator, 'sensitivity_levels', {})
+                level_name = sensitivity_levels.get(current_level, {}).get('name', 'medium')
+                
+                completed_trades = len(getattr(self.orchestrator, 'completed_trades', []))
+                learning_queue_size = len(getattr(self.orchestrator, 'sensitivity_learning_queue', []))
+                
+                open_threshold = getattr(self.orchestrator, 'confidence_threshold_open', 0.6)
+                close_threshold = getattr(self.orchestrator, 'confidence_threshold_close', 0.25)
+                
+                return {
+                    'level_name': level_name.upper(),
+                    'completed_trades': completed_trades,
+                    'learning_queue_size': learning_queue_size,
+                    'open_threshold': open_threshold,
+                    'close_threshold': close_threshold
+                }
+            else:
+                return {
+                    'level_name': 'DISABLED',
+                    'completed_trades': 0,
+                    'learning_queue_size': 0,
+                    'open_threshold': 0.6,
+                    'close_threshold': 0.25
+                }
+        except Exception as e:
+            logger.error(f"Error getting sensitivity learning info: {e}")
+            return {
+                'level_name': 'ERROR',
+                'completed_trades': 0,
+                'learning_queue_size': 0,
+                'open_threshold': 0.6,
+                'close_threshold': 0.25
+            }
+    
     def _create_orchestrator_status(self):
         """Create orchestrator data flow status"""
         try: