Upload 3 files

README.md

@@ -109,6 +109,24 @@ Every model DetectionBench has trained and evaluated on Global Wheat Head Datase
 | wheat_head       | 53.82 | 19.65 |
 ---
 
+## Per-Country Performance
+
+Domain shift can matter more than the aggregate score above for field deployment, so this evaluates the same test split broken down by the contributing country/institution, using per-image domain metadata (`domain_metadata.json` in this repository) compiled by this project for this stratified evaluation -- not a file shipped with the original GWHD release. Each row below is computed by re-running this exact model's evaluation restricted to that country's images only -- the same mAP definition as the aggregate number above (Ultralytics' `model.val()` for YOLO, Supervision's `MeanAveragePrecision` for RF-DETR), just on a filtered subset, not a separate metric implementation. One test image with no resolvable country in the source metadata (a documented upstream duplicate-filename quirk) is excluded from every row below.
+
+| Country                    | mAP@50          | mAP@50-95         | Test Images        |
+| --------------------------- | --------------- | ----------------- | ------------------ |
+| Australia       | 35.4 | 10.86 | 281 |
+| China       | 79.78 | 33.58 | 200 |
+| Japan       | 61.07 | 30.42 | 60 |
+| Mexico       | 54.51 | 19.27 | 205 |
+| Sudan       | 61.42 | 24.15 | 30 |
+| US       | 58.7 | 20.9 | 605 |
+
+
+See `country_breakdown.json` (results) and `domain_metadata.json` (the country/growth-stage mapping used to compute them) in this repository for the raw data behind this table.
+
+---
+
 ## Evaluation Visualizations
 
 This model was evaluated with [Supervision](https://github.com/roboflow/supervision)'s detection metrics, which report mAP/Precision/Recall directly but don't produce PR-curve, F1-curve, or confusion-matrix plot images the way Ultralytics' validator does. See the Performance table above for Precision/Recall/F1 and the per-class table above for the full per-class mAP breakdown.
@@ -178,6 +196,8 @@ detections = model.predict("image.jpg", threshold=0.25)
 checkpoint_best_total.pth
 metrics.csv
 config.json
+country_breakdown.json
+domain_metadata.json
 gwhd_rfdetr-nano_showcase.jpg
 README.md
 ```

country_breakdown.json

@@ -0,0 +1,32 @@
+{
+  "Australia": {
+    "mAP50": 0.3540254533290863,
+    "mAP50_95": 0.10858235508203506,
+    "num_images": 281
+  },
+  "China": {
+    "mAP50": 0.7978321313858032,
+    "mAP50_95": 0.3357720375061035,
+    "num_images": 200
+  },
+  "Japan": {
+    "mAP50": 0.6107019186019897,
+    "mAP50_95": 0.30420443415641785,
+    "num_images": 60
+  },
+  "Mexico": {
+    "mAP50": 0.5450758934020996,
+    "mAP50_95": 0.1927119940519333,
+    "num_images": 205
+  },
+  "Sudan": {
+    "mAP50": 0.6141822934150696,
+    "mAP50_95": 0.24153614044189453,
+    "num_images": 30
+  },
+  "US": {
+    "mAP50": 0.5870068073272705,
+    "mAP50_95": 0.2090480625629425,
+    "num_images": 605
+  }
+}