Generated independent cohort labels {'training': 60, 'validation': 20, 'calibration': 30, 'test': 40, 'shift_test': 20}
Validation workload_heuristic MAE 37.336 latency_ms 0.005
Validation resource_fluid MAE 23.656 latency_ms 0.012
Validation ridge MAE 12.031 latency_ms 0.178
Validation random_forest MAE 10.382 latency_ms 8.915
Validation hist_gradient_boosting MAE 10.307 latency_ms 2.486
Validation monte_carlo MAE 3.824 latency_ms 806.869
Selection frozen before calibration: monte_carlo
Held-out workload_heuristic MAE 33.567 joint coverage 0.975
Held-out resource_fluid MAE 22.351 joint coverage 0.85
Held-out ridge MAE 11.614 joint coverage 0.925
Held-out random_forest MAE 9.243 joint coverage 0.95
Held-out hist_gradient_boosting MAE 9.356 joint coverage 0.925
Held-out monte_carlo MAE 3.942 joint coverage 0.925
{"selected_model": "monte_carlo", "seconds": 77.5362872229889}
