feat(ml): parameterize SL/TP multipliers in dataset_builder

Add atr_sl_mult and atr_tp_mult parameters to _calc_labels_vectorized
and generate_dataset_vectorized, defaulting to existing constants (1.5,
2.0) for full backward compatibility. Callers (train scripts, backtester)
can now pass symbol-specific multipliers without modifying module-level
constants.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

src/dataset_builder.py

@@ -323,6 +323,8 @@ def _calc_labels_vectorized(
     d: pd.DataFrame,
     feat: pd.DataFrame,
     sig_idx: np.ndarray,
+    atr_sl_mult: float = ATR_SL_MULT,
+    atr_tp_mult: float = ATR_TP_MULT,
 ) -> tuple[np.ndarray, np.ndarray]:
     """
     label_builder.py build_labels() 로직을 numpy 2D 배열로 벡터화한다.
@@ -348,11 +350,11 @@ def _calc_labels_vectorized(
             continue
 
         if signal == "LONG":
-            sl = entry - atr * ATR_SL_MULT
+            sl = entry - atr * atr_sl_mult
-            tp = entry + atr * ATR_TP_MULT
+            tp = entry + atr * atr_tp_mult
         else:
-            sl = entry + atr * ATR_SL_MULT
+            sl = entry + atr * atr_sl_mult
-            tp = entry - atr * ATR_TP_MULT
+            tp = entry - atr * atr_tp_mult
 
         end = min(idx + 1 + LOOKAHEAD, n_total)
         fut_high = highs[idx + 1 : end]
@@ -391,6 +393,8 @@ def generate_dataset_vectorized(
     signal_threshold: int = 3,
     adx_threshold: float = 25,
     volume_multiplier: float = 2.5,
+    atr_sl_mult: float = ATR_SL_MULT,
+    atr_tp_mult: float = ATR_TP_MULT,
 ) -> pd.DataFrame:
     """
     전체 시계열을 1회 계산해 학습 데이터셋을 생성한다.
@@ -435,7 +439,10 @@ def generate_dataset_vectorized(
     print(f"  신호 발생 인덱스: {len(sig_idx):,}개")
 
     print("  [3/3] 레이블 계산...")
-    labels, valid_mask = _calc_labels_vectorized(d, feat_all, sig_idx)
+    labels, valid_mask = _calc_labels_vectorized(
+        d, feat_all, sig_idx,
+        atr_sl_mult=atr_sl_mult, atr_tp_mult=atr_tp_mult,
+    )
 
     final_sig_idx = sig_idx[valid_mask]
     available_feature_cols = [c for c in FEATURE_COLS if c in feat_all.columns]

tests/test_ml_pipeline_fixes.py

@@ -0,0 +1,56 @@
+import numpy as np
+import pandas as pd
+import pytest
+from src.dataset_builder import generate_dataset_vectorized, _calc_labels_vectorized
+
+
+@pytest.fixture
+def signal_df():
+    """시그널이 발생하는 데이터."""
+    rng = np.random.default_rng(7)
+    n = 800
+    trend = np.linspace(1.5, 3.0, n)
+    noise = np.cumsum(rng.normal(0, 0.04, n))
+    close = np.clip(trend + noise, 0.01, None)
+    high = close * (1 + rng.uniform(0, 0.015, n))
+    low = close * (1 - rng.uniform(0, 0.015, n))
+    volume = rng.uniform(1e6, 3e6, n)
+    volume[::30] *= 3.0
+    return pd.DataFrame({
+        "open": close, "high": high, "low": low,
+        "close": close, "volume": volume,
+    })
+
+
+def test_sltp_params_are_passed_through(signal_df):
+    """SL/TP 배수가 generate_dataset_vectorized에 전달되어야 한다."""
+    # 파라미터가 수용되는지(TypeError 없이) 확인하는 것이 핵심
+    r1 = generate_dataset_vectorized(
+        signal_df, atr_sl_mult=1.5, atr_tp_mult=2.0,
+        adx_threshold=0, volume_multiplier=1.5,
+    )
+    r2 = generate_dataset_vectorized(
+        signal_df, atr_sl_mult=2.0, atr_tp_mult=2.0,
+        adx_threshold=0, volume_multiplier=1.5,
+    )
+    # 두 결과 모두 DataFrame이어야 한다
+    assert isinstance(r1, pd.DataFrame)
+    assert isinstance(r2, pd.DataFrame)
+    # 신호가 충분히 많을 경우, 다른 SL 배수는 레이블 분포에 영향을 줄 수 있다
+    if len(r1) > 10 and len(r2) > 10:
+        assert not (r1["label"].values == r2["label"].values).all() or len(r1) != len(r2), \
+            "SL 배수가 다르면 레이블이 달라져야 한다"
+
+
+def test_default_sltp_backward_compatible(signal_df):
+    """SL/TP 파라미터 미지정 시 기존 기본값(1.5, 2.0)으로 동작해야 한다."""
+    r_default = generate_dataset_vectorized(
+        signal_df, adx_threshold=0, volume_multiplier=1.5,
+    )
+    r_explicit = generate_dataset_vectorized(
+        signal_df, atr_sl_mult=1.5, atr_tp_mult=2.0,
+        adx_threshold=0, volume_multiplier=1.5,
+    )
+    if len(r_default) > 0:
+        assert len(r_default) == len(r_explicit)
+        assert (r_default["label"].values == r_explicit["label"].values).all()