跳到主要內容

Feature Iteration Run #337: zz_sylvain

Status: completed | Started: 2026-07-16 13:14:09 | Ended: 2026-07-16 13:19:31

⭐ Best Accepted Iteration

Iteration 1: k_volume / k_volume.rolling(20).mean() * (k_close - k_low) …

baseline PF: 1.036 → modified PF: 1.100 | Cohen's d: 0.000 | p-value: 1.000

📊 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 — gradient_boosting — — — k_loop_iter_1_69d01c k_volume / k_volume.rolling(20).mean() … ✅ 1.036 1.100 0.4563 0.000 1.000 0.000 ✅ ACCEPT accepted

📄 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**: 7
> **Claim**: Adding a feature improves Sharpe by 10%
> **Objective mode**: scalarize
> **Metrics**: ['profit_factor', 'sharpe_ratio', 'win_rate', 'max_drawdown', 'test_accuracy']
> **Total duration**: 321.84s
> **Total accepted**: 1
> **Total rejected**: 0

## Iterations 詳情

### Iteration 1
- DSL: `k_volume / k_volume.rolling(20).mean() * (k_close - k_low) / (k_high - k_low)` (feature=k_loop_iter_1_69d01c)
- DSL valid: True
- FeatureSet candidate id: 302
- baseline PF: 1.036 → modified PF: 1.100
- t_stat: 0.000, p_value: 1.000, cohens_d: 0.000
- scalar_score: 0.4563
- accepted: True, reason: accepted

**Metrics summary:**
  - profit_factor: baseline=1.036, modified=1.100, diff=0.064
  - sharpe_ratio: baseline=1.010, modified=2.756, diff=1.746
  - win_rate: baseline=0.460, modified=0.455, diff=-0.005
  - max_drawdown: baseline=0.100, modified=0.110, diff=0.010
  - test_accuracy: baseline=0.968, modified=0.969, diff=0.001

📊 Multi-Objective Metrics

Iteration 1: k_loop_iter_1_69d01c

Y Label: — | Parent FS: #—

Objective mode: scalarize

Scalar score: 0.4563

MetricBaselineModifiedDiff
win_rate 0.460 0.455 -0.005
max_drawdown 0.100 0.110 0.010
sharpe_ratio 1.010 2.756 1.746
profit_factor 1.036 1.100 0.064
test_accuracy 0.968 0.969 0.001

📁 Offline Report

Markdown backup: reports/autoloop/zz_sylvain/20260716_211931.md