chore(reset): v2.0.0 — storico certificato Deribit mainnet, ripartenza pulita

Reset del progetto su fondamenta verificate dopo la scoperta che l'intera libreria "validata OOS" era artefatto di feed contaminato (print fantasma del feed Cerbero TESTNET + storico Binance/USDT). - Storico ricostruito da Deribit MAINNET (ccxt pubblico, tokenless) e CERTIFICATO (certify_feed.py): BTC/ETH puliti su TUTTA la storia (mediana 2-6 bps vs Coinbase USD), integrita' OHLC + coerenza resample (maxΔ 0.00) + cross-venue OK. Alt esclusi (illiquidi/divergenti: LTC/DOGE 50-82% barre flat; XRP/BNB non certificabili). - Verdetto sul feed pulito: FADE / PAIRS / XS01 / TSM01 morti (ogni portafoglio Sharpe -2.3..-3.0, DD ~40%); solo SH01 e frammenti HONEST con segnale residuo, da ri-validare in isolamento. - Cleanup "restart pulito": strategie, stack live (src/live, src/portfolio, runner/executor, yml, docker), ~100 script ricerca/gate, waste/games/ portfolios, dati non certificati + cache e 60+ diari -> archiviati in Old/ (preservati, non cancellati). Diario consolidato in un unico documento. - Skeleton ricerca tenuto: Strategy ABC + indicatori + src/fractal + src/backtest/engine + load_data; tool dati certificati (rebuild_history, certify_feed, audit_feed, multi_source_check). - Universo dati ATTIVO: solo BTC/ETH (5m/15m/1h); guardrail fisico (load_data su alt -> FileNotFoundError). Esecuzione DISABILITATA, conto flat. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-19 15:16:03 +00:00
parent 8401a280b9
commit 14522262e6
383 changed files with 1971 additions and 779 deletions
@@ -0,0 +1,81 @@
+"""Filtro edge minimo (`min_tp_frac`): MR01/DIP01 NON devono emettere segnali il cui TP
+(la media) cade entro `min_tp_frac` dall'entry — sarebbero perdenti garantiti netto fee.
+Proprietà testate su dati reali BTC 1h:
+  1. monotonia: alzando min_tp_frac il numero di segnali non aumenta;
+  2. ogni segnale superstite ha gap TP > min_tp_frac;
+  3. con min_tp_frac=0 il comportamento è invariato (default off = backtest validato intatto).
+"""
+import numpy as np
+import pytest
+from src.data.downloader import load_data
+from scripts.strategies.MR01_bollinger_fade import BollingerFade
+import importlib
+
+_dip_mod = importlib.import_module("scripts.strategies.DIP01_dip_buy")
+DipCls = next(v for k, v in vars(_dip_mod).items()
+              if isinstance(v, type) and k.lower().startswith("dip"))
+
+
+@pytest.fixture(scope="module")
+def btc():
+    df = load_data("BTC", "1h")
+    return df, df.index  # ts non usato dalle fade, basta un placeholder
+
+
+def _gaps(signals, df):
+    c = df["close"].values
+    return [abs(s.metadata["tp"] - c[s.idx]) / c[s.idx] for s in signals]
+
+
+def test_mr01_filter_monotone_and_gap(btc):
+    df, ts = btc
+    s = BollingerFade()
+    base = dict(bb_window=50, k=2.5, sl_atr=2.0, max_bars=24)
+    n0 = len(s.generate_signals(df, ts, **base, min_tp_frac=0.0))
+    for f in (0.0010, 0.0015, 0.0020, 0.005):
+        sig = s.generate_signals(df, ts, **base, min_tp_frac=f)
+        assert len(sig) <= n0                               # monotonia
+        gaps = _gaps(sig, df)
+        assert all(g > f for g in gaps)                     # nessun superstite sotto soglia
+
+
+def test_mr01_default_off_unchanged(btc):
+    df, ts = btc
+    s = BollingerFade()
+    base = dict(bb_window=50, k=2.5, sl_atr=2.0, max_bars=24)
+    a = s.generate_signals(df, ts, **base)                  # default (no kw)
+    b = s.generate_signals(df, ts, **base, min_tp_frac=0.0)
+    assert len(a) == len(b)
+
+
+def test_dip01_filter_gap(btc):
+    df, ts = btc
+    s = DipCls()
+    base = dict(n=50, z_in=2.0, sl_atr=2.5, max_bars=24)
+    n0 = len(s.generate_signals(df, ts, **base, min_tp_frac=0.0))
+    sig = s.generate_signals(df, ts, **base, min_tp_frac=0.0020)
+    assert len(sig) <= n0
+    assert all(g > 0.0020 for g in _gaps(sig, df))
+
+
+def _load(mod_name):
+    import importlib
+    m = importlib.import_module(mod_name)
+    return next(v() for k, v in vars(m).items()
+                if isinstance(v, type) and getattr(v, "__module__", "") == m.__name__
+                and hasattr(v, "generate_signals"))
+
+
+def test_mr02_mr07_filter_gap(btc):
+    """Anche MR02 (midpoint canale) e MR07 (ATR-scaled) onorano min_tp_frac."""
+    df, ts = btc
+    for mod, base in (
+        ("scripts.strategies.MR02_donchian_fade", dict(n=20, sl_atr=2.0, max_bars=24)),
+        ("scripts.strategies.MR07_return_reversal",
+         dict(n=50, k=3.5, tp_atr=2.0, sl_atr=1.5, max_bars=24)),
+    ):
+        s = _load(mod)
+        n0 = len(s.generate_signals(df, ts, **base, min_tp_frac=0.0))
+        sig = s.generate_signals(df, ts, **base, min_tp_frac=0.0015)
+        assert len(sig) <= n0
+        assert all(g > 0.0015 for g in _gaps(sig, df))