bw4sz commited on
Commit
a9b1a95
·
verified ·
1 Parent(s): f790f78

Add comparison figures to the model card

Browse files
.gitattributes CHANGED
@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ figures/E014_overlays_wd.png filter=lfs diff=lfs merge=lfs -text
37
+ figures/E014_per_source_f1_wd.png filter=lfs diff=lfs merge=lfs -text
README.md CHANGED
@@ -55,6 +55,26 @@ Predictions are matched to ground truth at IoU 0.4. AP40 is scored with up to 10
55
  held out. It shows how the recipe does on sources it never saw. These weights were trained on the
56
  within-distribution split.
57
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
58
  For a stronger but heavier model, see `weecology/deepforest-tree-dinov3` (RetinaNet with a DINOv3 ViT-L
59
  backbone, which needs access to Meta's DINOv3 weights).
60
 
 
55
  held out. It shows how the recipe does on sources it never saw. These weights were trained on the
56
  within-distribution split.
57
 
58
+ ### Figures
59
+
60
+ ![F1 and AP40 for the previous release, the ResNet-50 fine-tune and the DINOv3 fine-tune](figures/E014_macro.png)
61
+
62
+ Macro-averaged over sources on the MillionTrees TreeBoxes v0.25 test splits. The out-of-distribution bars
63
+ come from checkpoints trained with whole sources held out.
64
+
65
+ ![Per-source F1 on the within-distribution test split](figures/E014_per_source_f1_wd.png)
66
+
67
+ Per source on the within-distribution test split. The new models improve on the NEON release everywhere
68
+ except NEON MultiTemporal, where all three are close. The largest gains are on sources unlike NEON:
69
+ drone imagery, palms, tropical forest and urban trees.
70
+
71
+ ![Predictions on held-out test tiles](figures/E014_overlays_wd.png)
72
+
73
+ The same held-out test tile for each model; purple is ground truth and orange is prediction. Rows are the
74
+ six largest test sources, each showing the image with that source's median number of trees, so the tiles
75
+ were not picked for gain. Some sources label only some of the trees (for example OAM-TCD and the palm
76
+ counting set), so an unmatched orange box is not always a false positive.
77
+
78
  For a stronger but heavier model, see `weecology/deepforest-tree-dinov3` (RetinaNet with a DINOv3 ViT-L
79
  backbone, which needs access to Meta's DINOv3 weights).
80
 
figures/E014_macro.png ADDED
figures/E014_overlays_wd.png ADDED

Git LFS Details

  • SHA256: 62ba7d96a278d81ca860aa27c10277192e6fa4871b4a093690062ff7628e0714
  • Pointer size: 132 Bytes
  • Size of remote file: 3.6 MB
figures/E014_per_source_f1_wd.png ADDED

Git LFS Details

  • SHA256: 181464ecfb40f0ed30feed552afd54c5ef3aae7507e2738ba2cb6fdcdbc8efab
  • Pointer size: 131 Bytes
  • Size of remote file: 188 kB