Add trained YOLO model checkpoints

README.md

@@ -1,3 +1,41 @@
 ---
 license: apache-2.0
+tags:
+  - object-detection
+  - yolo
+  - water-meter
+  - computer-vision
 ---
+
+# Rwanda Water Meter Reading Model
+
+This repository contains a YOLO model trained to detect:
+
+- water meter
+- reading window
+- digits `0` through `9`
+- unreadable digits (`unknown`)
+
+## Files
+
+- `best.pt`: best checkpoint from training
+- `last.pt`: most recent checkpoint
+
+## Training source
+
+The model was trained on the prepared dataset from the Rwanda water meter
+reading project.
+
+## Suggested usage
+
+```python
+from ultralytics import YOLO
+
+model = YOLO("best.pt")
+results = model.predict("image.jpg", imgsz=640)
+```
+
+## Notes
+
+- `best.pt` is the recommended checkpoint for inference
+- `last.pt` can be used to continue training

best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7215bf64c237d7b376c4f0ac85141ace5b75acdbf5f8be805e2b8ac573269167
+size 6219050

last.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ddc392cff38efbf83d8352f8ecfbc7461a37eca38553fef4db6564bc157aad3
+size 6219050