Update README.md
Browse files
README.md
CHANGED
|
@@ -1,4 +1,24 @@
|
|
| 1 |
---
|
| 2 |
license: etalab-2.0
|
| 3 |
pipeline_tag: object-detection
|
| 4 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: etalab-2.0
|
| 3 |
pipeline_tag: object-detection
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
This model has be trained for the Panoramax project in order to detect:
|
| 7 |
+
- people face to blur them
|
| 8 |
+
- licence plates to blur them
|
| 9 |
+
- road signs to classify them with other models
|
| 10 |
+
|
| 11 |
+
The last model has been trained on yolo11l with imgsz of 2048 and 300 epochs, the older one on yolo8s.
|
| 12 |
+
|
| 13 |
+
Here is the last run validation :
|
| 14 |
+
```
|
| 15 |
+
Validating runs/detect/train5/weights/best.pt...
|
| 16 |
+
Ultralytics 8.3.29 🚀 Python-3.12.3 torch-2.5.1+cu124 CUDA:0 (NVIDIA GeForce RTX 4090, 24207MiB)
|
| 17 |
+
YOLO11l summary (fused): 464 layers, 25,281,625 parameters, 0 gradients, 86.6 GFLOPs
|
| 18 |
+
Class Images Instances Box(P R mAP50 mAP50-95): 100%|██████████| 83/83 [00:08<00:00, 9.78it/s]
|
| 19 |
+
all 329 1209 0.812 0.768 0.815 0.412
|
| 20 |
+
sign 231 507 0.879 0.836 0.898 0.561
|
| 21 |
+
plate 202 410 0.833 0.849 0.889 0.438
|
| 22 |
+
face 118 292 0.724 0.619 0.657 0.237
|
| 23 |
+
Speed: 1.2ms preprocess, 18.4ms inference, 0.0ms loss, 1.6ms postprocess per image
|
| 24 |
+
```
|