newman-p2
Intersection F1
0.897
Fiducial F1
0.712
Label accuracy
1.000
Leader accuracy
1.000
Trace IoU
0.980
End-to-end F1
0.685
Seconds / PDF
28.3s
Metric comparison (normalized)
Each metric scaled to [0,1] so you can see correlation and co-movement
Evolution across milestones
Same pdf evaluated at each milestone
| Milestone | Intersection F1 | Fiducial F1 | Label | Leader | Trace IoU | e2e F1 | Seconds / PDF |
|---|---|---|---|---|---|---|---|
| m01-v2-weights | 0.831 | 0.712 | 1.000 | 0.962 | 0.809 | 0.548 | 21.4s |
| m02-v3-weights | 0.888 | 0.712 | 1.000 | 0.885 | 0.976 | 0.685 | 24.2s |
| m03-v4-weights | 0.897 | 0.712 | 1.000 | 1.000 | 0.980 | 0.685 | 28.3s |
| m04-classifier | 0.910 | 0.712 | 1.000 | 1.000 | 0.980 | 0.685 | 22.9s |
| m05-llm-initial | 0.995 | 1.000 | 1.000 | 1.000 | 0.982 | 0.964 | 24.1s |
| m06-grid-bbox | 0.995 | 1.000 | 1.000 | 1.000 | 0.982 | 0.964 | 24.0s |
| m07-full-detector | 0.995 | 1.000 | 1.000 | 1.000 | 0.982 | 0.964 | 23.7s |
| m08-spatial-index | 0.995 | 1.000 | 1.000 | 1.000 | 0.982 | 0.964 | 15.7s |
| m09-current-logic | 0.995 | 1.000 | 1.000 | 1.000 | 0.982 | 0.964 | 15.9s |
| m10-v5-weights | 0.958 | 1.000 | 1.000 | 1.000 | 0.948 | 0.929 | 15.1s |
| m11-latest-pipeline | 0.958 | 1.000 | 1.000 | 1.000 | 0.948 | 0.929 | 15.6s |
| auto-fbce4876 | 0.958 | 1.000 | 1.000 | 1.000 | 0.948 | 0.929 | 9.6s |
| auto-b62c0378 | 0.958 | 1.000 | 1.000 | 1.000 | 0.948 | 0.929 | 9.5s |