""" Exploration EXHAUSTIVE de toutes les configurations possibles """ import logging import pandas as pd import numpy as np from apollo_client import ApolloDegenClient from signals.short_term_signals import ShortTermSignalGenerator from strategies.inverse_strategy_v10 import InverseStrategyV10 import config logging.basicConfig(level=logging.WARNING) logger = logging.getLogger(__name__) print("\n" + "="*100) print("🔬 EXPLORATION EXHAUSTIVE - RECHERCHE DE CONFIGURATIONS PROFITABLES") print("="*100) # RĂ©cupĂ©rer donnĂ©es client = ApolloDegenClient() df = client.get_klines('BTCUSDT', '1m', limit=1500) signal_gen = ShortTermSignalGenerator() inverse_strategy = InverseStrategyV10() # ParamĂštres Ă  tester test_configs = { 'inversion': [False, True], # Direct ou InversĂ© 'sl_pct': [0.001, 0.002, 0.003, 0.004, 0.005, 0.006, 0.007, 0.008, 0.01], 'tp_ratio': [2, 3, 4, 5, 6, 8, 10, 15, 20], 'leverage': [50, 100, 150, 200, 250] } results = [] total_tests = len(test_configs['inversion']) * len(test_configs['sl_pct']) * len(test_configs['tp_ratio']) * len(test_configs['leverage']) print(f"\n📊 Tests Ă  effectuer: {total_tests}") print(f"⏱ Estimation: ~{total_tests * 2}s\n") current_test = 0 for use_inversion in test_configs['inversion']: for sl_pct in test_configs['sl_pct']: for tp_ratio in test_configs['tp_ratio']: for leverage in test_configs['leverage']: current_test += 1 if current_test % 50 == 0: print(f" Progress: {current_test}/{total_tests} ({current_test/total_tests*100:.1f}%)") try: # Backtest simulation capital = 1000 trades = [] risk_per_trade = 0.1 # 10% du capital par trade for i in range(100, len(df) - 50, 20): # Check every 20 candles if len([t for t in trades if 'exit_idx' not in t]) >= 3: # Max 3 positions parallĂšles continue window_df = df.iloc[:i] # GĂ©nĂ©rer signal if use_inversion: signal = inverse_strategy.generate_inverse_signal(window_df, 'BTCUSDT') else: signals = signal_gen.generate_signals(window_df, 'BTCUSDT') signal = max(signals, key=lambda x: x.score) if signals else None if not signal: continue # Entry entry_price = df.iloc[i]['close'] entry_idx = i # Calculate SL/TP if signal.side == 'LONG': sl = entry_price * (1 - sl_pct) tp = entry_price * (1 + sl_pct * tp_ratio) else: # SHORT sl = entry_price * (1 + sl_pct) tp = entry_price * (1 - sl_pct * tp_ratio) # Track trade trades.append({ 'entry_price': entry_price, 'entry_idx': entry_idx, 'side': signal.side, 'sl': sl, 'tp': tp }) # Process exits for all trades for trade in trades: if 'exit_idx' in trade: continue for j in range(trade['entry_idx'] + 1, len(df)): candle = df.iloc[j] exit_price = None exit_reason = None if trade['side'] == 'LONG': if candle['low'] <= trade['sl']: exit_price = trade['sl'] exit_reason = 'SL' elif candle['high'] >= trade['tp']: exit_price = trade['tp'] exit_reason = 'TP' else: # SHORT if candle['high'] >= trade['sl']: exit_price = trade['sl'] exit_reason = 'SL' elif candle['low'] <= trade['tp']: exit_price = trade['tp'] exit_reason = 'TP' if exit_price: # Calculate PnL if trade['side'] == 'LONG': pnl_pct = (exit_price - trade['entry_price']) / trade['entry_price'] else: # SHORT pnl_pct = (trade['entry_price'] - exit_price) / trade['entry_price'] pnl_usd = capital * risk_per_trade * leverage * pnl_pct capital += pnl_usd trade['exit_idx'] = j trade['exit_price'] = exit_price trade['exit_reason'] = exit_reason trade['pnl_pct'] = pnl_pct * 100 trade['pnl_usd'] = pnl_usd break # Calculate stats completed_trades = [t for t in trades if 'exit_idx' in t] if len(completed_trades) < 5: # Skip if too few trades continue final_capital = capital total_return = ((final_capital - 1000) / 1000) * 100 wins = [t for t in completed_trades if t['pnl_usd'] > 0] losses = [t for t in completed_trades if t['pnl_usd'] <= 0] win_rate = len(wins) / len(completed_trades) if completed_trades else 0 avg_win = np.mean([t['pnl_pct'] for t in wins]) if wins else 0 avg_loss = np.mean([t['pnl_pct'] for t in losses]) if losses else 0 total_wins = sum(t['pnl_usd'] for t in wins) if wins else 0 total_losses = abs(sum(t['pnl_usd'] for t in losses)) if losses else 1 profit_factor = total_wins / total_losses if total_losses > 0 else 0 # Store result results.append({ 'inversion': use_inversion, 'sl_pct': sl_pct, 'tp_ratio': tp_ratio, 'leverage': leverage, 'return': total_return, 'win_rate': win_rate, 'profit_factor': profit_factor, 'trades': len(completed_trades), 'avg_win': avg_win, 'avg_loss': avg_loss, 'final_capital': final_capital }) except Exception as e: pass # Analyse des rĂ©sultats print("\n" + "="*100) print("📊 ANALYSE DES RÉSULTATS") print("="*100) if not results: print("\n❌ Aucun rĂ©sultat trouvĂ©") else: # Top 20 configurations sorted_results = sorted(results, key=lambda x: x['return'], reverse=True)[:20] print(f"\n🏆 TOP 20 CONFIGURATIONS PROFITABLES\n") print(f"{'Rank':<5} {'Inv':<4} {'SL%':<6} {'TP:1':<6} {'Lev':<6} {'Return%':<10} {'WR%':<7} {'PF':<7} {'Trades':<8} {'AvgW/L':<12}") print("-"*100) for i, r in enumerate(sorted_results, 1): inv_str = "YES" if r['inversion'] else "NO" ratio_str = f"{r['avg_win']/abs(r['avg_loss']) if r['avg_loss'] != 0 else 0:.2f}:1" print( f"{i:<5} {inv_str:<4} {r['sl_pct']*100:<6.2f} {r['tp_ratio']:<6} {r['leverage']:<6} " f"{r['return']:<10.2f} {r['win_rate']*100:<7.1f} {r['profit_factor']:<7.2f} " f"{r['trades']:<8} {ratio_str:<12}" ) # Meilleure config best = sorted_results[0] print("\n" + "="*100) print("đŸ„‡ MEILLEURE CONFIGURATION") print("="*100) print(f"StratĂ©gie: {'INVERSÉE (V10)' if best['inversion'] else 'DIRECTE (signaux normaux)'}") print(f"Stop-Loss: {best['sl_pct']*100:.2f}%") print(f"Take-Profit: {best['sl_pct']*100*best['tp_ratio']:.2f}% (ratio {best['tp_ratio']}:1)") print(f"Leverage: {best['leverage']}x") print(f"") print(f"Return: {best['return']:+.2f}%") print(f"Win Rate: {best['win_rate']*100:.1f}%") print(f"Profit Factor: {best['profit_factor']:.2f}") print(f"Total Trades: {best['trades']}") print(f"Avg Win: {best['avg_win']:+.2f}%") print(f"Avg Loss: {best['avg_loss']:+.2f}%") print(f"Reward:Risk: {best['avg_win']/abs(best['avg_loss']) if best['avg_loss'] != 0 else 0:.2f}:1") print(f"Final Capital: ${best['final_capital']:.2f}") # Statistiques par type print("\n" + "="*100) print("📈 STATISTIQUES PAR TYPE") print("="*100) profitable = [r for r in results if r['return'] > 0] with_inversion = [r for r in profitable if r['inversion']] without_inversion = [r for r in profitable if not r['inversion']] print(f"\nTotal configs testĂ©es: {len(results)}") print(f"Configs profitables: {len(profitable)} ({len(profitable)/len(results)*100:.1f}%)") print(f" - AVEC inversion: {len(with_inversion)}") print(f" - SANS inversion (direct): {len(without_inversion)}") if with_inversion: avg_return_inv = np.mean([r['return'] for r in with_inversion]) print(f"\nAvg return AVEC inversion: {avg_return_inv:+.2f}%") if without_inversion: avg_return_direct = np.mean([r['return'] for r in without_inversion]) print(f"Avg return SANS inversion: {avg_return_direct:+.2f}%") print("\n" + "="*100) # Recommandation if best['return'] > 10: print("✅ Configuration profitable trouvĂ©e ! PrĂȘte Ă  ĂȘtre appliquĂ©e.") elif best['return'] > 0: print("⚠ Configuration lĂ©gĂšrement profitable. Besoin d'optimisation supplĂ©mentaire.") else: print("❌ Aucune configuration profitable trouvĂ©e sur cette pĂ©riode.") print("="*100)