import sys, time, math, itertools from pathlib import Path import numpy as np import pandas as pd import gc sys.path.insert(0, str(Path(__file__).parent)) from nautilus_dolphin.nautilus.alpha_orchestrator import NDAlphaEngine from nautilus_dolphin.nautilus.adaptive_circuit_breaker import AdaptiveCircuitBreaker from nautilus_dolphin.nautilus.ob_features import OBFeatureEngine from nautilus_dolphin.nautilus.ob_provider import MockOBProvider VBT_DIR = Path(r"C:\Users\Lenovo\Documents\- DOLPHIN NG HD HCM TSF Predict\vbt_cache") META_COLS = {'timestamp', 'scan_number', 'v50_lambda_max_velocity', 'v150_lambda_max_velocity', 'v300_lambda_max_velocity', 'v750_lambda_max_velocity', 'vel_div', 'instability_50', 'instability_150'} parquet_files = sorted(VBT_DIR.glob("*.parquet")) parquet_files = [p for p in parquet_files if 'catalog' not in str(p)] print("Loading data...") acb = AdaptiveCircuitBreaker() date_strings = [pf.stem for pf in parquet_files] acb.preload_w750(date_strings) all_vols = [] for pf in parquet_files[:2]: df = pd.read_parquet(pf) if 'BTCUSDT' not in df.columns: continue pr = df['BTCUSDT'].values for i in range(60, len(pr)): seg = pr[max(0,i-50):i] if len(seg)<10: continue diffs = np.zeros(len(seg)-1) for j in range(len(seg)-1): if seg[j] > 0: diffs[j] = (seg[j+1]-seg[j])/seg[j] v = float(np.std(diffs)) if v > 0: all_vols.append(v) vol_p60 = float(np.percentile(all_vols, 60)) pq_data = {} for pf in parquet_files: df = pd.read_parquet(pf) ac = [c for c in df.columns if c not in META_COLS] bp = df['BTCUSDT'].values if 'BTCUSDT' in df.columns else None dv = np.full(len(df), np.nan) if bp is not None: for i in range(50, len(bp)): seg = bp[max(0,i-50):i] if len(seg)<10: continue diffs = np.zeros(len(seg)-1) for j in range(len(seg)-1): if seg[j] > 0: diffs[j] = (seg[j+1]-seg[j])/seg[j] dv[i] = float(np.std(diffs)) pq_data[pf.stem] = (df, ac, dv) VD_THRESH = -0.02; VD_EXTREME = -0.05; CONVEXITY = 3.0 def strength_cubic(vel_div, threshold=-0.02): if vel_div >= threshold: return 0.0 raw = (threshold - vel_div) / (threshold - VD_EXTREME) return min(1.0, max(0.0, raw)) ** CONVEXITY assets = ["BTCUSDT", "ETHUSDT", "BNBUSDT", "SOLUSDT"] def evaluate_params(tp_bps, dc_lookback, dc_magnitude, min_irp, max_hold): engine_kwargs = dict( initial_capital=25000.0, vel_div_threshold=-0.02, vel_div_extreme=-0.05, min_leverage=0.5, max_leverage=5.0, leverage_convexity=3.0, fraction=0.20, fixed_tp_pct=tp_bps/10000.0, stop_pct=1.0, max_hold_bars=max_hold, use_direction_confirm=True, dc_lookback_bars=dc_lookback, dc_min_magnitude_bps=dc_magnitude, dc_skip_contradicts=True, dc_leverage_boost=1.0, dc_leverage_reduce=0.5, use_asset_selection=True, min_irp_alignment=min_irp, use_sp_fees=True, use_sp_slippage=True, sp_maker_entry_rate=0.62, sp_maker_exit_rate=0.50, use_ob_edge=True, ob_edge_bps=5.0, ob_confirm_rate=0.40, lookback=100, use_alpha_layers=True, use_dynamic_leverage=True, seed=42, ) mock = MockOBProvider( imbalance_bias=-0.09, depth_scale=1.0, assets=assets, imbalance_biases={"BTCUSDT": -0.086, "ETHUSDT": -0.092, "BNBUSDT": +0.20, "SOLUSDT": +0.20}, ) ob_eng = OBFeatureEngine(mock) ob_eng.preload_date("mock", assets) engine = NDAlphaEngine(**engine_kwargs) engine.set_ob_engine(ob_eng) bar_idx = 0; ph = {}; dstats = [] # We evaluate on half the parquet files (about 25 days) to make the sweep fast enough for pf in parquet_files[:25]: ds = pf.stem; cs = engine.capital engine.regime_direction = -1 engine.regime_dd_halt = False acb_info = acb.get_dynamic_boost_for_date(ds, ob_engine=ob_eng) engine.regime_size_mult = acb_info['boost'] df, acols, dvol = pq_data[ds] bid = 0 for ri in range(len(df)): row = df.iloc[ri]; vd = row.get("vel_div") if vd is None or not np.isfinite(vd): bar_idx+=1; bid+=1; continue prices = {} for ac in acols: p = row[ac] if p and p > 0 and np.isfinite(p): prices[ac] = float(p) if ac not in ph: ph[ac] = [] ph[ac].append(float(p)) if len(ph[ac]) > 500: ph[ac] = ph[ac][-200:] if not prices: bar_idx+=1; bid+=1; continue vrok = False if bid < 100 else (np.isfinite(dvol[ri]) and dvol[ri] > vol_p60) if acb_info['beta'] > 0: ss = strength_cubic(float(vd)) engine.regime_size_mult = acb_info['boost'] * (1.0 + acb_info['beta'] * ss) engine.process_bar(bar_idx=bar_idx, vel_div=float(vd), prices=prices, vol_regime_ok=vrok, price_histories=ph) bar_idx+=1; bid+=1 dstats.append({'date': ds, 'pnl': engine.capital - cs, 'cap': engine.capital}) tr = engine.trade_history w = [t for t in tr if t.pnl_absolute > 0]; l = [t for t in tr if t.pnl_absolute <= 0] wr = len(w) / len(tr) * 100 if tr else 0.0 pf = sum(t.pnl_absolute for t in w) / abs(sum(t.pnl_absolute for t in l)) if l else 999 roi = (engine.capital - 25000) / 25000 * 100 del engine gc.collect() return roi, wr, pf, len(tr) print("Starting parameter sweep...") params = { 'tp_bps': [89, 99, 109], 'dc_lookback': [5, 7, 9], 'dc_magnitude': [0.5, 0.75], 'min_irp': [0.40, 0.45], 'max_hold': [120, 150] } keys, values = zip(*params.items()) permutations = [dict(zip(keys, v)) for v in itertools.product(*values)] print(f"Total combinations: {len(permutations)}") best_wr = 0 best_roi = 0 for i, p in enumerate(permutations): try: t0 = time.time() roi, wr, pf, trades = evaluate_params( p['tp_bps'], p['dc_lookback'], p['dc_magnitude'], p['min_irp'], p['max_hold'] ) print(f"[{i+1}/{len(permutations)}] {p} -> ROI: {roi:+.2f}% | WR: {wr:.2f}% | PF: {pf:.2f} | Trades: {trades} [{time.time()-t0:.1f}s]") if wr > best_wr: best_wr = wr print(f" *** NEW BEST WR: {wr:.2f}% ***") if roi > best_roi: best_roi = roi print(f" *** NEW BEST ROI: {roi:+.2f}% ***") except Exception as e: print(f"Error on {p}: {e}")