Flagship1 means
- Best-fold PR-AUC: 0.004416
- Best-fold PR lift: 4.129
- Best-fold ROC AUC: 0.637707
- Overall PR-AUC: 0.003338
Seeded External Cocci Evaluation
Seeded external evaluation compares flagship1 and flagship2 across sets 1-10. Values below are manually curated from the current benchmark snapshot and paired statistics tables.
Set range: 1-10 | Seed rows: 20 | Fold rows: 100
Flagship2 means
- Best-fold PR-AUC: 0.007278
- Best-fold PR lift: 6.806
- Best-fold ROC AUC: 0.639068
- Overall PR-AUC: 0.004251
Paired Statistics (flagship2 - flagship1)
| Metric | n pairs | Mean delta | Median delta | 95% bootstrap CI | Sign-test p |
|---|---|---|---|---|---|
| Best-fold ROC AUC | 10 | +0.001361 | +0.000928 | -0.001267 to +0.004166 | 3.44e-1 |
| Best-fold PR-AUC | 10 | +0.002862 | +0.002008 | +0.001045 to +0.004890 | 2.15e-2 |
| Best-fold PR lift | 10 | +2.676529 | +1.877902 | +0.991165 to +4.619555 | 2.15e-2 |
| Overall ROC AUC | 10 | +0.002112 | +0.002052 | +0.000141 to +0.004152 | 3.44e-1 |
| Overall PR-AUC | 10 | +0.000914 | +0.000655 | +0.000443 to +0.001422 | 1.95e-3 |
| Peptide precision | 10 | -0.000003 | +0.000003 | -0.000032 to +0.000026 | 7.54e-1 |
| Peptide recall | 10 | -0.002151 | +0.005376 | -0.025269 to +0.020430 | 1.00e+0 |
| Peptide F1 | 10 | -0.000005 | +0.000005 | -0.000061 to +0.000052 | 7.54e-1 |