跳到主要內容

Feature Iteration Run #454: zz_sylvain

Status: completed | Started: 2026-07-16 22:16:58 | Ended: 2026-07-16 22:19:10

📊 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_2_9b7fe8 return_5 - return_20 ✅ 1.286 1.019 -1.9376 0.000 1.000 0.000 ❌ REJECT p=1.000 ≥ 0.05, cohens_d=0.000 ≤ 0.3, constraints: ['sharpe_ratio >= 1.0: got 0.5207']
2 — gradient_boosting — — — k_loop_iter_2_c45ea3 dist_to_high - dist_to_low ✅ 1.286 1.187 -0.5523 0.000 1.000 0.000 ❌ REJECT p=1.000 ≥ 0.05, cohens_d=0.000 ≤ 0.3
3 — gradient_boosting — — — k_loop_iter_3_713d47 k_close.pct_change(10) * k_volume / k_v… ✅ 1.286 1.159 -0.9355 0.000 1.000 0.000 ❌ REJECT p=1.000 ≥ 0.05, cohens_d=0.000 ≤ 0.3

📄 Report Markdown

# 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**: 131.78s
> **Total accepted**: 0
> **Total rejected**: 3

## Iterations 詳情

### Iteration 1
- DSL: `return_5 - return_20` (feature=k_loop_iter_2_9b7fe8)
- DSL valid: True
- FeatureSet candidate id: 468
- baseline PF: 1.286 → modified PF: 1.019
- t_stat: 0.000, p_value: 1.000, cohens_d: 0.000
- scalar_score: -1.9376
- accepted: False, reason: p=1.000 ≥ 0.05, cohens_d=0.000 ≤ 0.3, constraints: ['sharpe_ratio >= 1.0: got 0.5207']

**Metrics summary:**
  - profit_factor: baseline=1.286, modified=1.019, diff=-0.266
  - sharpe_ratio: baseline=7.879, modified=0.521, diff=-7.358
  - win_rate: baseline=0.553, modified=0.425, diff=-0.128
  - max_drawdown: baseline=0.075, modified=0.133, diff=0.059
  - test_accuracy: baseline=0.986, modified=0.972, diff=-0.014

### Iteration 2
- DSL: `dist_to_high - dist_to_low` (feature=k_loop_iter_2_c45ea3)
- DSL valid: True
- FeatureSet candidate id: 469
- baseline PF: 1.286 → modified PF: 1.187
- t_stat: 0.000, p_value: 1.000, cohens_d: 0.000
- scalar_score: -0.5523
- accepted: False, reason: p=1.000 ≥ 0.05, cohens_d=0.000 ≤ 0.3

**Metrics summary:**
  - profit_factor: baseline=1.286, modified=1.187, diff=-0.099
  - sharpe_ratio: baseline=7.879, modified=5.779, diff=-2.100
  - win_rate: baseline=0.553, modified=0.540, diff=-0.013
  - max_drawdown: baseline=0.075, modified=0.133, diff=0.058
  - test_accuracy: baseline=0.986, modified=0.949, diff=-0.037

### Iteration 3
- DSL: `k_close.pct_change(10) * k_volume / k_volume.rolling(15).mean()` (feature=k_loop_iter_3_713d47)
- DSL valid: True
- FeatureSet candidate id: 470
- baseline PF: 1.286 → modified PF: 1.159
- t_stat: 0.000, p_value: 1.000, cohens_d: 0.000
- scalar_score: -0.9355
- accepted: False, reason: p=1.000 ≥ 0.05, cohens_d=0.000 ≤ 0.3

**Metrics summary:**
  - profit_factor: baseline=1.286, modified=1.159, diff=-0.126
  - sharpe_ratio: baseline=7.879, modified=4.309, diff=-3.570
  - win_rate: baseline=0.553, modified=0.449, diff=-0.105
  - max_drawdown: baseline=0.075, modified=0.191, diff=0.116
  - test_accuracy: baseline=0.986, modified=0.970, diff=-0.016

📊 Multi-Objective Metrics

Iteration 1: k_loop_iter_2_9b7fe8

Y Label: — | Parent FS: #—

Objective mode: scalarize

Scalar score: -1.9376

Failed constraints: sharpe_ratio >= 1.0: got 0.5207

MetricBaselineModifiedDiff
win_rate 0.553 0.425 -0.128
max_drawdown 0.075 0.133 0.059
sharpe_ratio 7.879 0.521 -7.358
profit_factor 1.286 1.019 -0.266
test_accuracy 0.986 0.972 -0.014

Iteration 2: k_loop_iter_2_c45ea3

Y Label: — | Parent FS: #—

Objective mode: scalarize

Scalar score: -0.5523

MetricBaselineModifiedDiff
win_rate 0.553 0.540 -0.013
max_drawdown 0.075 0.133 0.058
sharpe_ratio 7.879 5.779 -2.100
profit_factor 1.286 1.187 -0.099
test_accuracy 0.986 0.949 -0.037

Iteration 3: k_loop_iter_3_713d47

Y Label: — | Parent FS: #—

Objective mode: scalarize

Scalar score: -0.9355

MetricBaselineModifiedDiff
win_rate 0.553 0.449 -0.105
max_drawdown 0.075 0.191 0.116
sharpe_ratio 7.879 4.309 -3.570
profit_factor 1.286 1.159 -0.126
test_accuracy 0.986 0.970 -0.016

📁 Offline Report

Markdown backup: reports/autoloop/zz_sylvain/20260717_061910.md