targeted sync runs/summary_ctt.md
Browse files
workspace/runs/summary_ctt.md
CHANGED
|
@@ -3,14 +3,28 @@
|
|
| 3 |
| Method | Status | Success | Proposal oracle | Support gap | Selector gap | OutcomePTR | PPTC@0.20 | PPTC@0.40 | NegativeNear@0.20 | Calibration ECE | Run |
|
| 4 |
| --- | --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | --- |
|
| 5 |
| V0 residual transport diagnostic | measured diagnostic baseline | 0.3890 | 0.4435 | 0.1264 | 0.0545 | n/a | 0.2366 | 0.5269 | 0.0533 | n/a | `runs/reproduce_v0` |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
| CTT gated residual | train | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | `runs/ctt_gated_residual_smoke` |
|
| 7 |
| CTT gated residual | proxy | n/a | n/a | n/a | n/a | n/a | 0.5000 | 0.7500 | 0.0573 | n/a | `runs/ctt_gated_residual_smoke_proxy` |
|
| 8 |
| CTT gated residual | proxy | n/a | n/a | n/a | n/a | n/a | 0.1739 | 0.5362 | 0.0281 | n/a | `runs/ctt_gated_residual_val_proxy` |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
| CTT residual train | train | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | `runs/ctt_residual_smoke` |
|
| 10 |
| CTT residual proxy | proxy | n/a | n/a | n/a | n/a | n/a | 1.0000 | 1.0000 | 0.1010 | n/a | `runs/ctt_residual_smoke_proxy` |
|
| 11 |
| CTT residual proxy | proxy | n/a | n/a | n/a | n/a | n/a | 0.3478 | 0.7101 | 0.0397 | n/a | `runs/ctt_residual_val_proxy` |
|
| 12 |
| task memory | proxy baseline | n/a | n/a | n/a | n/a | n/a | 0.3188 | 0.5362 | 0.0175 | n/a | `runs/task_memory_val_proxy` |
|
| 13 |
| local atlas | proxy baseline | n/a | n/a | n/a | n/a | n/a | 0.4058 | 0.6812 | 0.0368 | n/a | `runs/local_atlas_val_proxy` |
|
|
|
|
|
|
|
| 14 |
| Utility energy smoke | calibration diagnostic | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | 0.1316 | `runs/utility_energy_smoke` |
|
| 15 |
|
| 16 |
`n/a` means the required measured artifact does not exist yet.
|
|
|
|
| 3 |
| Method | Status | Success | Proposal oracle | Support gap | Selector gap | OutcomePTR | PPTC@0.20 | PPTC@0.40 | NegativeNear@0.20 | Calibration ECE | Run |
|
| 4 |
| --- | --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: | --- |
|
| 5 |
| V0 residual transport diagnostic | measured diagnostic baseline | 0.3890 | 0.4435 | 0.1264 | 0.0545 | n/a | 0.2366 | 0.5269 | 0.0533 | n/a | `runs/reproduce_v0` |
|
| 6 |
+
| CTT gated residual | train | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | `runs/ctt_gated_residual_full_seed0` |
|
| 7 |
+
| CTT gated residual | proxy | n/a | n/a | n/a | n/a | n/a | 0.2319 | 0.5797 | 0.0512 | n/a | `runs/ctt_gated_residual_full_seed0_val_proxy` |
|
| 8 |
+
| CTT gated residual | train | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | `runs/ctt_gated_residual_full_seed1` |
|
| 9 |
+
| CTT gated residual | proxy | n/a | n/a | n/a | n/a | n/a | 0.2174 | 0.6232 | 0.0644 | n/a | `runs/ctt_gated_residual_full_seed1_val_proxy` |
|
| 10 |
+
| CTT gated residual | train | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | `runs/ctt_gated_residual_full_seed2` |
|
| 11 |
+
| CTT gated residual | proxy | n/a | n/a | n/a | n/a | n/a | 0.2464 | 0.6377 | 0.0424 | n/a | `runs/ctt_gated_residual_full_seed2_val_proxy` |
|
| 12 |
| CTT gated residual | train | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | `runs/ctt_gated_residual_smoke` |
|
| 13 |
| CTT gated residual | proxy | n/a | n/a | n/a | n/a | n/a | 0.5000 | 0.7500 | 0.0573 | n/a | `runs/ctt_gated_residual_smoke_proxy` |
|
| 14 |
| CTT gated residual | proxy | n/a | n/a | n/a | n/a | n/a | 0.1739 | 0.5362 | 0.0281 | n/a | `runs/ctt_gated_residual_val_proxy` |
|
| 15 |
+
| CTT residual train | train | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | `runs/ctt_residual_full_seed0` |
|
| 16 |
+
| CTT residual proxy | proxy | n/a | n/a | n/a | n/a | n/a | 0.2029 | 0.5797 | 0.0289 | n/a | `runs/ctt_residual_full_seed0_val_proxy` |
|
| 17 |
+
| CTT residual train | train | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | `runs/ctt_residual_full_seed1` |
|
| 18 |
+
| CTT residual proxy | proxy | n/a | n/a | n/a | n/a | n/a | 0.2319 | 0.5942 | 0.0410 | n/a | `runs/ctt_residual_full_seed1_val_proxy` |
|
| 19 |
+
| CTT residual train | train | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | `runs/ctt_residual_full_seed2` |
|
| 20 |
+
| CTT residual proxy | proxy | n/a | n/a | n/a | n/a | n/a | 0.1594 | 0.6522 | 0.0189 | n/a | `runs/ctt_residual_full_seed2_val_proxy` |
|
| 21 |
| CTT residual train | train | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | `runs/ctt_residual_smoke` |
|
| 22 |
| CTT residual proxy | proxy | n/a | n/a | n/a | n/a | n/a | 1.0000 | 1.0000 | 0.1010 | n/a | `runs/ctt_residual_smoke_proxy` |
|
| 23 |
| CTT residual proxy | proxy | n/a | n/a | n/a | n/a | n/a | 0.3478 | 0.7101 | 0.0397 | n/a | `runs/ctt_residual_val_proxy` |
|
| 24 |
| task memory | proxy baseline | n/a | n/a | n/a | n/a | n/a | 0.3188 | 0.5362 | 0.0175 | n/a | `runs/task_memory_val_proxy` |
|
| 25 |
| local atlas | proxy baseline | n/a | n/a | n/a | n/a | n/a | 0.4058 | 0.6812 | 0.0368 | n/a | `runs/local_atlas_val_proxy` |
|
| 26 |
+
| ctt residual full | proxy aggregate gate=True | n/a | n/a | n/a | n/a | n/a | 0.1981 | 0.6087 | 0.0296 | n/a | `runs/ctt_val_proxy_comparison` |
|
| 27 |
+
| ctt gated residual full | proxy aggregate gate=False | n/a | n/a | n/a | n/a | n/a | 0.2319 | 0.6135 | 0.0527 | n/a | `runs/ctt_val_proxy_comparison` |
|
| 28 |
| Utility energy smoke | calibration diagnostic | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | 0.1316 | `runs/utility_energy_smoke` |
|
| 29 |
|
| 30 |
`n/a` means the required measured artifact does not exist yet.
|