refactor: 精简仓库为 chanlun 核心与 web 分析，移除威科夫与遗留模块

删除根目录旧 Chan 模块、策略、配置、文档及 wyckoff 相关代码；更新缠论 pipeline 与笔中枢计算；补充 research 研究与 web 测试。 Co-authored-by: Cursor <cursoragent@cursor.com>
2026-08-27 01:05:12 +08:00
parent 5c10e35b76
commit 7f393b93ed
360 changed files with 140008 additions and 41167 deletions
@@ -0,0 +1,239 @@
+"""Step 28：小级别 × 大级别的全网格扫描。
+
+此前的配对（15m配1h、30m配2h）是拍脑袋定的，而 step26 意外发现
+5m 配 1h 明显好过配 15m —— 说明「大级别该拉多远」本身是个未扫过的维度。
+
+关键省算：大级别只充当过滤标记，不改变入场点，所以每个小级别的
+中枢/信号/回测只跑一次，然后把所有候选大级别一次性挂上去。
+"""
+from __future__ import annotations
+
+import argparse
+import os
+import sys
+import warnings
+from concurrent.futures import ProcessPoolExecutor, as_completed
+from pathlib import Path
+
+import numpy as np
+import pandas as pd
+
+warnings.filterwarnings("ignore")
+for v in ("OMP_NUM_THREADS", "OPENBLAS_NUM_THREADS", "MKL_NUM_THREADS"):
+    os.environ.setdefault(v, "1")
+
+HERE = Path(__file__).resolve().parent
+sys.path.insert(0, str(HERE))
+sys.path.insert(0, str(HERE.parent))
+pd.set_option("display.width", 320)
+
+SL, TP, MAXB = 1.5, 3.0, 48
+FEE, SLIP = 0.0004, 0.0001
+
+LTF_HTFS = {
+    "5m": ["15m", "30m", "1h", "2h", "4h"],
+    "15m": ["30m", "1h", "2h", "4h", "1d"],
+    "30m": ["1h", "2h", "4h", "1d"],
+    "1h": ["2h", "4h", "1d", "1w"],
+}
+
+
+def run_one(task: tuple) -> dict | None:
+    import warnings as _w
+    _w.filterwarnings("ignore")
+    sys.path.insert(0, str(HERE))
+    sys.path.insert(0, str(HERE.parent))
+
+    from chanlun import TF_DF
+    from lib.breakout import run_trades
+    from lib.data import fetch_ohlcv
+    from lib.fast_bsp3 import find_fast_bsp3
+    from lib.fx_signal import extract_fx_signals, signals_to_frame
+    from lib.nested_bsp import attach_htf_context, htf_fx_timeline
+    from lib.nested_level import build_htf_zones
+
+    sym, ltf = task
+    htfs = LTF_HTFS[ltf]
+    try:
+        df_l = fetch_ohlcv(f"{sym}/USDT:USDT", ltf, 10**9)
+        if df_l is None or len(df_l) < 3000:
+            return None
+        chan_l = TF_DF(df_l, 1, ltf)
+        cdf = chan_l.dataframe
+        zones = build_htf_zones(cdf, ltf, chan=chan_l).reset_index(drop=True)
+        if zones.empty:
+            return None
+        sig = find_fast_bsp3(cdf, zones)
+        if sig.empty or len(sig) < 10:
+            return None
+
+        z = zones.copy()
+        pg, pdn = z["zg"].shift(), z["zd"].shift()
+        z["z_above"] = z["zd"] > pg
+        z["z_below"] = z["zg"] < pdn
+        z["zone_i"] = np.arange(len(z))
+        sig = sig.merge(z[["zone_i", "z_above", "z_below"]], on="zone_i", how="left")
+
+        # 所有候选大级别一次挂完，前缀就用周期名
+        for tf in htfs:
+            df_h = fetch_ohlcv(f"{sym}/USDT:USDT", tf, 10**9)
+            if df_h is None or len(df_h) < 200:
+                continue
+            chan_h = TF_DF(df_h, 1, tf)
+            s = signals_to_frame(extract_fx_signals(chan_h, chan_h.dataframe))
+            if s.empty:
+                continue
+            sig = attach_htf_context(sig, cdf, htf_fx_timeline(s, chan_h.dataframe),
+                                     f"p{tf}")
+
+        entries = list(zip(sig["entry_idx"].astype(int), sig["direction"].astype(int)))
+        tr = run_trades(cdf, entries, SL, TP, MAXB, fee=0.0, entry_delay=1)
+        if tr.empty:
+            return None
+        m = sig.drop_duplicates("entry_idx").set_index("entry_idx")
+        tr["symbol"], tr["ltf"] = sym, ltf
+        tr["date"] = cdf["date"].to_numpy()[tr["entry_idx"].to_numpy()]
+        for c in ["z_above", "z_below"] + [f"p{tf}_agree" for tf in htfs]:
+            tr[c] = tr["entry_idx"].map(m[c]) if c in m.columns else np.nan
+        return {"task": f"{sym} {ltf}", "trades": tr}
+    except Exception as e:
+        return {"task": f"{sym} {ltf}", "error": repr(e)[:200]}
+
+
+def stat(r: np.ndarray) -> dict | None:
+    if len(r) < 40:
+        return None
+    w, o = r[r > 0], r[r <= 0]
+    sd = r.std(ddof=1)
+    return {
+        "笔数": len(r), "胜率": (r > 0).mean() * 100,
+        "均收益": r.mean() * 100, "中位": np.median(r) * 100,
+        "PF": w.sum() / abs(o.sum()) if len(o) else np.inf,
+        "偏度": pd.Series(r).skew(),
+        "t值": r.mean() / (sd / np.sqrt(len(r))),
+    }
+
+
+def main() -> None:
+    ap = argparse.ArgumentParser()
+    ap.add_argument("--symbols", default="BTC,ETH,SOL")
+    ap.add_argument("--ltfs", default="5m,15m,30m,1h")
+    ap.add_argument("--workers", type=int, default=6)
+    ap.add_argument("--reuse", action="store_true")
+    args = ap.parse_args()
+
+    cache = HERE / "out" / "step28_grid.csv"
+    if args.reuse and cache.exists():
+        allt = pd.read_csv(cache, parse_dates=["date"])
+        print(f"[复用] {len(allt)} 笔\n")
+    else:
+        tasks = [(s, l) for l in args.ltfs.split(",") for s in args.symbols.split(",")]
+        print(f"[网格] {len(tasks)} 个任务，每个内部挂多个大级别\n", flush=True)
+        res = []
+        with ProcessPoolExecutor(max_workers=args.workers) as ex:
+            futs = {ex.submit(run_one, t): t for t in tasks}
+            for i, f in enumerate(as_completed(futs), 1):
+                r = f.result()
+                if r is None or "error" in (r or {}):
+                    print(f"  [{i}] 跳过 {(r or {}).get('error', '')}", flush=True)
+                    continue
+                res.append(r)
+                print(f"  [{i}/{len(tasks)}] {r['task']} — {len(r['trades'])} 笔",
+                      flush=True)
+        if not res:
+            return
+        allt = pd.concat([r["trades"] for r in res], ignore_index=True)
+        allt.to_csv(cache, index=False)
+
+    allt["date"] = pd.to_datetime(allt["date"])
+    allt["ret_net"] = allt["gross"] - FEE - SLIP
+    allt["push"] = np.where(allt["direction"] == 1,
+                            allt["z_above"], allt["z_below"]).astype(bool)
+
+    print("=" * 120)
+    print("########## 1. 全网格：小级别 × 大级别（仅大级别同向）##########")
+    rows = []
+    for ltf, g in allt.groupby("ltf"):
+        for tf in LTF_HTFS.get(ltf, []):
+            col = f"p{tf}_agree"
+            if col not in g.columns:
+                continue
+            s = stat(g[g[col] == 1]["ret_net"].to_numpy())
+            if s:
+                rows.append({"小级别": ltf, "大级别": tf, **s})
+    df = pd.DataFrame(rows)
+    if df.empty:
+        print("样本不足")
+        return
+    for c, f in (("胜率", "{:.1f}%"), ("均收益", "{:+.3f}%"), ("中位", "{:+.3f}%"),
+                 ("PF", "{:.2f}"), ("偏度", "{:.2f}"), ("t值", "{:+.2f}")):
+        df[c] = df[c].map(f.format)
+    print(df.to_string(index=False))
+
+    print("\n########## 2. 加上顺向推进过滤后的全网格 ##########")
+    rows = []
+    for ltf, g in allt[allt.push].groupby("ltf"):
+        for tf in LTF_HTFS.get(ltf, []):
+            col = f"p{tf}_agree"
+            if col not in g.columns:
+                continue
+            s = stat(g[g[col] == 1]["ret_net"].to_numpy())
+            if s:
+                rows.append({"小级别": ltf, "大级别": tf, **s})
+    df2 = pd.DataFrame(rows)
+    if not df2.empty:
+        best = df2.sort_values("PF", ascending=False).head(3)
+        for c, f in (("胜率", "{:.1f}%"), ("均收益", "{:+.3f}%"), ("中位", "{:+.3f}%"),
+                     ("PF", "{:.2f}"), ("偏度", "{:.2f}"), ("t值", "{:+.2f}")):
+            df2[c] = df2[c].map(f.format)
+        print(df2.to_string(index=False))
+        print(f"\n  PF 前三：{', '.join(f'{r.小级别}/{r.大级别}' for r in best.itertuples())}")
+
+    print("\n########## 3. 每个小级别的最优大级别（按 t 值，中位需为正）##########")
+    rows = []
+    for ltf, g in allt[allt.push].groupby("ltf"):
+        cand = []
+        for tf in LTF_HTFS.get(ltf, []):
+            col = f"p{tf}_agree"
+            if col not in g.columns:
+                continue
+            s = stat(g[g[col] == 1]["ret_net"].to_numpy())
+            if s and s["中位"] > 0:
+                cand.append((tf, s))
+        if cand:
+            tf, s = max(cand, key=lambda x: x[1]["t值"])
+            rows.append({"小级别": ltf, "最优大级别": tf,
+                         "笔数": s["笔数"], "胜率": f"{s['胜率']:.1f}%",
+                         "中位": f"{s['中位']:+.3f}%", "PF": f"{s['PF']:.2f}",
+                         "t值": f"{s['t值']:+.2f}"})
+    if rows:
+        print(pd.DataFrame(rows).to_string(index=False))
+
+    print("\n########## 4. 最优配对的组合资金曲线 ##########")
+    picks = {r["小级别"]: r["最优大级别"] for r in rows} if rows else {}
+    if picks:
+        parts = []
+        for ltf, tf in picks.items():
+            g = allt[(allt.ltf == ltf) & allt.push & (allt[f"p{tf}_agree"] == 1)]
+            parts.append(g)
+        g = pd.concat(parts).sort_values("date")
+        r = g["ret_net"].to_numpy()
+        lev = np.clip(0.01 / np.clip(g["risk_pct"].to_numpy(), 0.002, None), 0, 20)
+        pnl = r * lev
+        eq = np.cumprod(1 + pnl)
+        yrs = (g["date"].max() - g["date"].min()).days / 365.25
+        dd = (1 - eq / np.maximum.accumulate(eq)).max()
+        print(f"  配对 {picks}")
+        print(f"  n={len(r)} 年{len(r) / yrs:.0f}笔  "
+              f"年化 {(eq[-1] ** (1 / yrs) - 1) * 100:+.1f}%  回撤 {dd * 100:.1f}%  "
+              f"Sharpe {pnl.mean() / pnl.std(ddof=1) * np.sqrt(len(pnl) / yrs):.2f}  "
+              f"中位 {np.median(r) * 100:+.3f}%")
+        gg = g.copy()
+        gg["y"] = gg["date"].dt.year
+        yr = [(y, (x["ret_net"] > 0).mean() * 100, x["ret_net"].mean() * 100, len(x))
+              for y, x in gg.groupby("y") if len(x) >= 20]
+        print("  分年：" + "  ".join(f"{y}:{m:+.2f}%({n})" for y, _, m, n in yr))
+
+
+if __name__ == "__main__":
+    main()