跳到主要內容

Feature Iteration Run #918: zz_sylvain

Status: completed | Started: 2026-07-25 14:34:20 | Ended: 2026-07-25 14:40:20

⭐ Best Accepted Iteration

Iteration 1: k_momentum_pct_change * obv_ensemble_3 / (1 + atr_norm * 10…

baseline PF: 1.947 → modified PF: 2.449 | Cohen's d: 0.242 | p-value: 0.252

📊 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 C8 gradient_boosting — — #7 k_loop_iter_1_d44640 k_momentum_pct_change * obv_ensemble_3 … ✅ 1.947 2.449 0.8286 0.000 0.252 0.242 ✅ ACCEPT accepted
2 C8 gradient_boosting — — #1241 k_loop_iter_2_a48a17 i_macd(k_close, 12, 26, 9) * obv_ensemb… ✅ 2.449 1.181 -0.6402 0.000 0.113 0.358 ❌ REJECT normalized_score -0.2362 <= 0.0, sharpe_delta -19.7631 < -0.1, hypothesis test not confirmed
3 C8 gradient_boosting — — #1241 k_iter3_k_momentum_pct_chang_price_position_atr_norm k_momentum_pct_change * (1 - price_posi… ❌ 0.000 0.000 0.0000 0.000 1.000 0.000 ❌ REJECT quality_gate: canonical duplicate: k_momentum_pct_change * (1 - price_position) * obv_ensemble_3 / (1 + atr_norm)
4 C8 gradient_boosting — — #1241 k_loop_iter_4_5d32b4 i_macd(k_close, 12, 26, 9) * (1 - price… ✅ 2.449 1.301 -0.4915 0.000 0.002 0.265 ❌ REJECT normalized_score -0.2566 <= 0.0, sharpe_delta -17.2574 < -0.1, hypothesis test not confirmed
5 C8 gradient_boosting — — #1241 k_iter5_i_macd_k_close_atr_norm i_macd(k_close, 12, 26, 9) * obv_ensemb… ❌ 0.000 0.000 0.0000 0.000 1.000 0.000 ❌ REJECT quality_gate: canonical duplicate: i_macd(k_close) * obv_ensemble_3 / (1 + atr_norm * 10)
6 C8 gradient_boosting — — #1241 k_loop_iter_6_2a70d2 (i_macd(k_close, 12, 26, 9) - i_macd(k_… ✅ 2.449 1.181 -4.0222 0.000 0.000 0.839 ❌ REJECT normalized_score -0.2362 <= 0.0, sharpe_delta -19.7631 < -0.1, hypothesis test not confirmed
7 C8 gradient_boosting — — #1241 k_iter7_k_high_pct_change_vol_ratio_atr_norm (k_high.pct_change(12) * vol_ratio) / (… ❌ 0.000 0.000 0.0000 0.000 1.000 0.000 ❌ REJECT quality_gate: NaN ratio 1.000 exceeds 0.3
8 C8 gradient_boosting — — #1241 k_loop_iter_8_88fd92 (k_high.pct_change(12) - i_macd(k_close… ✅ 2.449 1.181 -4.0222 0.000 0.000 0.839 ❌ REJECT normalized_score -0.2362 <= 0.0, sharpe_delta -19.7631 < -0.1, hypothesis test not confirmed

📄 Report Markdown

# AUTOLOOP: zz_sylvain (8 iterations)

> **Phase**: Y4.7 Option B (DSL → Python code → FeatureSet candidate)
> **Target indicator**: zz_sylvain
> **Base feature_set_id**: 1241
> **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**: 360.08s
> **Total accepted**: 1
> **Total rejected**: 7

## Iterations 詳情

### Iteration 1
- DSL: `k_momentum_pct_change * obv_ensemble_3 / (1 + atr_norm * 10) - i_macd(k_close, 12, 26, 9)` (feature=k_loop_iter_1_d44640)
- DSL valid: True
- FeatureSet candidate id: 1241
- baseline PF: 1.947 → modified PF: 2.449
- t_stat: 0.000, p_value: 0.252, cohens_d: 0.242
- scalar_score: 0.8286
- accepted: True, reason: accepted

**Metrics summary:**
  - profit_factor: baseline=1.947, modified=2.449, diff=0.501
  - sharpe_ratio: baseline=19.162, modified=24.736, diff=5.574
  - win_rate: baseline=0.600, modified=0.576, diff=-0.024
  - max_drawdown: baseline=0.056, modified=0.061, diff=0.005
  - test_accuracy: baseline=0.992, modified=0.992, diff=-0.000

### Iteration 2
- DSL: `i_macd(k_close, 12, 26, 9) * obv_ensemble_3 / (1 + atr_norm * 10)` (feature=k_loop_iter_2_a48a17)
- DSL valid: True
- FeatureSet candidate id: 1242
- baseline PF: 2.449 → modified PF: 1.181
- t_stat: 0.000, p_value: 0.113, cohens_d: 0.358
- scalar_score: -0.6402
- accepted: False, reason: normalized_score -0.2362 <= 0.0, sharpe_delta -19.7631 < -0.1, hypothesis test not confirmed

**Metrics summary:**
  - profit_factor: baseline=2.449, modified=1.181, diff=-1.268
  - sharpe_ratio: baseline=24.736, modified=4.973, diff=-19.763
  - win_rate: baseline=0.576, modified=0.586, diff=0.010
  - max_drawdown: baseline=0.061, modified=0.187, diff=0.125
  - test_accuracy: baseline=0.992, modified=0.992, diff=0.000

### Iteration 3
- DSL: `k_momentum_pct_change * (1 - price_position) * obv_ensemble_3 / (1 + atr_norm)` (feature=k_iter3_k_momentum_pct_chang_price_position_atr_norm)
- 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: quality_gate: canonical duplicate: k_momentum_pct_change * (1 - price_position) * obv_ensemble_3 / (1 + atr_norm)

### Iteration 4
- DSL: `i_macd(k_close, 12, 26, 9) * (1 - price_position) / (1 + atr_norm)` (feature=k_loop_iter_4_5d32b4)
- DSL valid: True
- FeatureSet candidate id: 1243
- baseline PF: 2.449 → modified PF: 1.301
- t_stat: 0.000, p_value: 0.002, cohens_d: 0.265
- scalar_score: -0.4915
- accepted: False, reason: normalized_score -0.2566 <= 0.0, sharpe_delta -17.2574 < -0.1, hypothesis test not confirmed

**Metrics summary:**
  - profit_factor: baseline=2.449, modified=1.301, diff=-1.148
  - sharpe_ratio: baseline=24.736, modified=7.479, diff=-17.257
  - win_rate: baseline=0.576, modified=0.446, diff=-0.130
  - max_drawdown: baseline=0.061, modified=0.097, diff=0.036
  - test_accuracy: baseline=0.992, modified=0.969, diff=-0.023

### Iteration 5
- DSL: `i_macd(k_close, 12, 26, 9) * obv_ensemble_3 / (1 + atr_norm * 10)` (feature=k_iter5_i_macd_k_close_atr_norm)
- 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: quality_gate: canonical duplicate: i_macd(k_close) * obv_ensemble_3 / (1 + atr_norm * 10)

### Iteration 6
- DSL: `(i_macd(k_close, 12, 26, 9) - i_macd(k_close, 6, 13, 4)) / (1 + i_atr(k_high, k_low, k_close, 14) / k_close * 10) * vol_ratio` (feature=k_loop_iter_6_2a70d2)
- DSL valid: True
- FeatureSet candidate id: 1244
- baseline PF: 2.449 → modified PF: 1.181
- t_stat: 0.000, p_value: 0.000, cohens_d: 0.839
- scalar_score: -4.0222
- accepted: False, reason: normalized_score -0.2362 <= 0.0, sharpe_delta -19.7631 < -0.1, hypothesis test not confirmed

**Metrics summary:**
  - profit_factor: baseline=2.449, modified=1.181, diff=-1.268
  - sharpe_ratio: baseline=24.736, modified=4.973, diff=-19.763
  - win_rate: baseline=0.576, modified=0.586, diff=0.010
  - max_drawdown: baseline=0.061, modified=0.187, diff=0.125
  - test_accuracy: baseline=0.992, modified=0.992, diff=0.000

### Iteration 7
- DSL: `(k_high.pct_change(12) * vol_ratio) / (1 + atr_norm * i_rsi(k_close, 14))` (feature=k_iter7_k_high_pct_change_vol_ratio_atr_norm)
- 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: quality_gate: NaN ratio 1.000 exceeds 0.3

### Iteration 8
- DSL: `(k_high.pct_change(12) - i_macd(k_close, 12, 26, 9)) * obv_ensemble_3 / (1 + atr_norm * 10)` (feature=k_loop_iter_8_88fd92)
- DSL valid: True
- FeatureSet candidate id: 1245
- baseline PF: 2.449 → modified PF: 1.181
- t_stat: 0.000, p_value: 0.000, cohens_d: 0.839
- scalar_score: -4.0222
- accepted: False, reason: normalized_score -0.2362 <= 0.0, sharpe_delta -19.7631 < -0.1, hypothesis test not confirmed

**Metrics summary:**
  - profit_factor: baseline=2.449, modified=1.181, diff=-1.268
  - sharpe_ratio: baseline=24.736, modified=4.973, diff=-19.763
  - win_rate: baseline=0.576, modified=0.586, diff=0.010
  - max_drawdown: baseline=0.061, modified=0.187, diff=0.125
  - test_accuracy: baseline=0.992, modified=0.992, diff=0.000

📊 Multi-Objective Metrics

Iteration 1: k_loop_iter_1_d44640

Y Label: C8 | Parent FS: #7

Objective mode: scalarize

Scalar score: 0.8286

MetricBaselineModifiedDiff
win_rate 0.600 0.576 -0.024
max_drawdown 0.056 0.061 0.005
sharpe_ratio 19.162 24.736 5.574
profit_factor 1.947 2.449 0.501
test_accuracy 0.992 0.992 0.000

Iteration 2: k_loop_iter_2_a48a17

Y Label: C8 | Parent FS: #1241

Objective mode: scalarize

Scalar score: -0.6402

MetricBaselineModifiedDiff
win_rate 0.576 0.586 0.010
max_drawdown 0.061 0.187 0.125
sharpe_ratio 24.736 4.973 -19.763
profit_factor 2.449 1.181 -1.268
test_accuracy 0.992 0.992 0.000

Iteration 3: k_iter3_k_momentum_pct_chang_price_position_atr_norm

Y Label: C8 | Parent FS: #1241

Objective mode: single

Scalar score: 0.0000

Iteration 4: k_loop_iter_4_5d32b4

Y Label: C8 | Parent FS: #1241

Objective mode: scalarize

Scalar score: -0.4915

MetricBaselineModifiedDiff
win_rate 0.576 0.446 -0.130
max_drawdown 0.061 0.097 0.036
sharpe_ratio 24.736 7.479 -17.257
profit_factor 2.449 1.301 -1.148
test_accuracy 0.992 0.969 -0.023

Iteration 5: k_iter5_i_macd_k_close_atr_norm

Y Label: C8 | Parent FS: #1241

Objective mode: single

Scalar score: 0.0000

Iteration 6: k_loop_iter_6_2a70d2

Y Label: C8 | Parent FS: #1241

Objective mode: scalarize

Scalar score: -4.0222

MetricBaselineModifiedDiff
win_rate 0.576 0.586 0.010
max_drawdown 0.061 0.187 0.125
sharpe_ratio 24.736 4.973 -19.763
profit_factor 2.449 1.181 -1.268
test_accuracy 0.992 0.992 0.000

Iteration 7: k_iter7_k_high_pct_change_vol_ratio_atr_norm

Y Label: C8 | Parent FS: #1241

Objective mode: single

Scalar score: 0.0000

Iteration 8: k_loop_iter_8_88fd92

Y Label: C8 | Parent FS: #1241

Objective mode: scalarize

Scalar score: -4.0222

MetricBaselineModifiedDiff
win_rate 0.576 0.586 0.010
max_drawdown 0.061 0.187 0.125
sharpe_ratio 24.736 4.973 -19.763
profit_factor 2.449 1.181 -1.268
test_accuracy 0.992 0.992 0.000

📁 Offline Report

Markdown backup: reports/autoloop/zz_sylvain/20260725_224020.md