跳到主要內容

Feature Iteration Run #2033: zz_sylvain

Status: completed | Started: 2026-10-09 06:07:05 | Ended: 2026-10-09 06:08:02

📊 Iterations

Iter Y Label Model Warm Proxy Parent FS Feature DSL DSL Valid Baseline PF Modified PF Scalar Score t-stat p-value Cohen's d Result Reason
1 zz_sylvain gradient_boosting — — #1748 k_bias_atr_norm_mom result = (close - ta.sma(close, 12)) / … ✅ 1.129 1.161 0.0339 0.000 0.019 0.274 ❌ REJECT DSR gate failed: champion SR=2.0142 ≤ expected_max=1.8908 (N=6 effective trials, PSR=0.5337)

📄 Report Markdown

# AUTOLOOP: zz_sylvain (1 iterations)

> **Phase**: Y4.7 Option B (DSL → Python code → FeatureSet candidate)
> **Target indicator**: zz_sylvain
> **Base feature_set_id**: 1748
> **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**: 56.83s
> **Total accepted**: 0
> **Total rejected**: 1

## Iterations 詳情

### Iteration 1
- DSL: `result = (close - ta.sma(close, 12)) / (ta.atr(high, low, close, 14) + 1e-8)` (feature=k_bias_atr_norm_mom)
- DSL valid: True
- FeatureSet candidate id: -2
- baseline PF: 1.129 → modified PF: 1.161
- t_stat: 0.000, p_value: 0.019, cohens_d: 0.274
- scalar_score: 0.0339
- accepted: False, reason: DSR gate failed: champion SR=2.0142 ≤ expected_max=1.8908 (N=6 effective trials, PSR=0.5337)

**Metrics summary:**
  - profit_factor: baseline=1.129, modified=1.161, diff=0.031
  - sharpe_ratio: baseline=4.170, modified=5.511, diff=1.341
  - win_rate: baseline=0.434, modified=0.415, diff=-0.019
  - max_drawdown: baseline=0.032, modified=0.039, diff=0.007
  - test_accuracy: baseline=0.944, modified=0.941, diff=-0.003

📊 Multi-Objective Metrics

Iteration 1: k_bias_atr_norm_mom

Y Label: zz_sylvain | Parent FS: #1748

Objective mode: scalarize

Scalar score: 0.0339

MetricBaselineModifiedDiff
win_rate 0.434 0.415 -0.019
max_drawdown 0.032 0.039 0.007
sharpe_ratio 4.170 5.511 1.341
profit_factor 1.129 1.161 0.031
test_accuracy 0.944 0.941 -0.003

📁 Offline Report

Markdown backup: reports/autoloop/zz_sylvain/20261009_140802.md