Feature Iteration Run #676: zz_sylvain

# AUTOLOOP: zz_sylvain (3 iterations)

> **Phase**: Y4.7 Option B (DSL → Python code → FeatureSet candidate)
> **Target indicator**: zz_sylvain
> **Base feature_set_id**: 7
> **Claim**: (default) Propose a new k_* feature that improves risk-adjusted returns
> **Objective mode**: scalarize
> **Metrics**: ['profit_factor', 'sharpe_ratio', 'win_rate', 'max_drawdown', 'test_accuracy']
> **Total duration**: 70.56s
> **Total accepted**: 0
> **Total rejected**: 3

## Iterations 詳情

### Iteration 1
- DSL: `k_high.pct_change(12) * i_rsi(k_close, 14) / (1 + atr_norm) * (1 - dist_to_high) * (1 - price_position) / (1 + k_body_size_pct) * (1 + return_5) * (1 + return_20) / (1 + abs(return_1)) / (1 + vol_ratio)` (feature=k_iter_1_k_high)
- DSL valid: False
- FeatureSet candidate id: None
- baseline PF: 0.000 → modified PF: 0.000
- t_stat: 0.000, p_value: 1.000, cohens_d: 0.000
- scalar_score: 0.0000
- accepted: False, reason: safe_eval DSL: 變數未定義: abs

### Iteration 2
- DSL: `k_high.pct_change(12) * i_rsi(k_close, 14) * (1 - price_position) * i_ema(return_5, 10) / (1 + hl_range) / (1 + k_consecutive_shrink) / (1 + abs(return_20))` (feature=k_iter_2_k_high)
- DSL valid: False
- FeatureSet candidate id: None
- baseline PF: 0.000 → modified PF: 0.000
- t_stat: 0.000, p_value: 1.000, cohens_d: 0.000
- scalar_score: 0.0000
- accepted: False, reason: safe_eval DSL: 變數未定義: abs

### Iteration 3
- DSL: `(k_high.pct_change(12) * i_rsi(k_close, 14) * (1 - price_position) * (1 + i_macd(k_close, 12, 26, 9))) / ((1 + atr_norm) * (1 + hl_range) * (1 + k_consecutive_shrink) * (1 + vol_ratio) * (1 + return_20))` (feature=k_loop_iter_3_419797)
- DSL valid: True
- FeatureSet candidate id: 852
- baseline PF: 1.286 → modified PF: 1.160
- t_stat: 0.000, p_value: 1.000, cohens_d: 0.000
- scalar_score: -0.8854
- accepted: False, reason: p=1.000 ≥ 0.05, cohens_d=0.000 ≤ 0.3

**Metrics summary:**
  - profit_factor: baseline=1.286, modified=1.160, diff=-0.125
  - sharpe_ratio: baseline=7.879, modified=4.532, diff=-3.347
  - win_rate: baseline=0.553, modified=0.510, diff=-0.043
  - max_drawdown: baseline=0.075, modified=0.059, diff=-0.016
  - test_accuracy: baseline=0.986, modified=0.986, diff=0.000