report validation results

README.md

@@ -26,14 +26,17 @@ model-index:
           split: validation
         metrics:
           - type: mean_average_precision
-            value: 0.00
+            value: 0.2842
             name: COCO mAP @ [0.5:0.95]
           - type: coco_ap_50
-            value: 0.00
+            value: 0.4189
             name: COCO AP@IoU0.5
-          - type: metric
-            value: 0.00
-            name: per-class mAP
+          - type: coco_ap_75
+            value: 0.3196
+            name: COCO AP@IoU0.75
+          - type: cascade_average_recall
+            value: 0.4302
+            name: AR@maxDets=100
 ---
 
 # opencode-r2
@@ -64,15 +67,28 @@ dataset (BigLam / Library of Congress). It detects the 7 page-layout content cla
 
 ### Validation results
 
+Evaluated on the 712-example `biglam/loc_beyond_words` validation split (COCO metrics via pycocotools).
+
 | Metric | Value |
 |---|---|
-| COCO mAP @ IoU [0.5:0.95] | _TBD_ |
-| COCO AP @ IoU 0.5 | _TBD_ |
-| COCO AP @ IoU 0.75 | _TBD_ |
-| AR@maxDets=100 | _TBD_ |
-| Per-class mean AP | _TBD_ |
+| COCO mAP @ IoU [0.5:0.95] | **0.2842** |
+| COCO AP @ IoU 0.5 | **0.4189** |
+| COCO AP @ IoU 0.75 | **0.3196** |
+| AR@maxDets=100 | **0.4302** |
+
+Per-class COCO mAP @ [0.5:0.95]:
 
-**Validation results will be reported here after training completes.**
+| Class | mAP |
+|---|---|
+| Photograph | 0.4027 |
+| Illustration | 0.0485 |
+| Map | 0.0431 |
+| Comics/Cartoon | 0.2761 |
+| Editorial Cartoon | 0.1702 |
+| Headline | 0.5226 |
+| Advertisement | 0.5259 |
+
+Training loss: 21.06 (epoch 0) → 7.44 (epoch 4).
 
 ## Usage