milford-p2
Intersection F1
0.921
Fiducial F1
0.916
Label accuracy
1.000
Leader accuracy
0.950
Trace IoU
0.822
End-to-end F1
0.672
Seconds / PDF
40.6s
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 | 0.859 | 0.868 | 1.000 | 0.305 | 0.542 | 0.147 | 35.2s |
| m02-v3-weights | 0.884 | 0.868 | 1.000 | 0.593 | 0.784 | 0.588 | 43.2s |
| m03-v4-weights | 0.950 | 0.868 | 1.000 | 0.576 | 0.841 | 0.647 | 53.4s |
| m04-classifier | 0.942 | 0.868 | 1.000 | 0.797 | 0.818 | 0.603 | 47.5s |
| m05-llm-initial | 0.948 | 0.857 | 1.000 | 0.800 | 0.812 | 0.586 | 52.8s |
| m06-grid-bbox | 0.948 | 0.857 | 1.000 | 0.800 | 0.812 | 0.586 | 49.9s |
| m07-full-detector | 0.941 | 0.816 | 1.000 | 0.800 | 0.830 | 0.585 | 49.4s |
| m08-spatial-index | 0.953 | 0.943 | 1.000 | 0.793 | 0.818 | 0.667 | 41.3s |
| m09-current-logic | 0.948 | 0.902 | 1.000 | 0.800 | 0.831 | 0.647 | 43.0s |
| m10-v5-weights | 0.934 | 0.916 | 1.000 | 0.800 | 0.821 | 0.656 | 43.6s |
| m11-latest-pipeline | 0.921 | 0.916 | 1.000 | 0.950 | 0.822 | 0.672 | 40.6s |
| auto-fbce4876 | 0.941 | 0.960 | 1.000 | 0.950 | 0.831 | 0.704 | 23.6s |
| auto-b62c0378 | 0.941 | 0.960 | 1.000 | 0.950 | 0.831 | 0.704 | 23.2s |