Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,62 @@
|
|
1 |
---
|
2 |
-
license:
|
|
|
|
|
|
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
license: mit
|
3 |
+
library_name: ultralytics
|
4 |
+
tags:
|
5 |
+
- yolov8
|
6 |
+
- object-detection
|
7 |
+
- pytorch
|
8 |
---
|
9 |
+
|
10 |
+
# TabDetect
|
11 |
+
|
12 |
+
<div align="center">
|
13 |
+
<img width="640" alt="https://huggingface.co/camiloa2m/TabDetect-YOLOv8s/blob/main/val_batch1_pred.jpg>
|
14 |
+
</div>
|
15 |
+
|
16 |
+
### Supported Labels
|
17 |
+
|
18 |
+
```
|
19 |
+
['full_lined', 'not_full_lined']
|
20 |
+
```
|
21 |
+
|
22 |
+
### How to use
|
23 |
+
|
24 |
+
- Install ultralytics:
|
25 |
+
|
26 |
+
```bash
|
27 |
+
pip install -U ultralytics==8.0.227
|
28 |
+
```
|
29 |
+
|
30 |
+
- Load model and perform prediction:
|
31 |
+
|
32 |
+
```python
|
33 |
+
from ultralytics import YOLO
|
34 |
+
|
35 |
+
# load model
|
36 |
+
model = YOLO('camiloa2m/TabDetect-YOLOv8s')
|
37 |
+
|
38 |
+
# set model parameters
|
39 |
+
model.overrides['conf'] = 0.25 # NMS confidence threshold
|
40 |
+
model.overrides['iou'] = 0.45 # NMS IoU threshold
|
41 |
+
model.overrides['agnostic_nms'] = False # NMS class-agnostic
|
42 |
+
model.overrides['max_det'] = 1000 # maximum number of detections per image
|
43 |
+
|
44 |
+
# set image
|
45 |
+
image = 'https://github.com/ultralytics/yolov5/raw/master/data/images/zidane.jpg'
|
46 |
+
|
47 |
+
# perform inference
|
48 |
+
results = model.predict(image)
|
49 |
+
```
|
50 |
+
|
51 |
+
### Dataset
|
52 |
+
|
53 |
+
[TNCR_Dataset](https://github.com/abdoelsayed2016/TNCR_Dataset). I merged some classes: class 0 (full_lined, merged_cells), class 1 (no_lines, partial_lined, partial_lined_merged_cells).
|
54 |
+
|
55 |
+
### Model summary (fused)
|
56 |
+
|
57 |
+
| Class | Images | Instances | P | R | mAP50 | mAP50-95 |
|
58 |
+
|----------------|--------|-----------|-------|-------|-------|----------|
|
59 |
+
| all | 1313 | 1906 | 0.957 | 0.926 | 0.973 | 0.938 |
|
60 |
+
| full_lined | 1313 | 984 | 0.96 | 0.949 | 0.98 | 0.968 |
|
61 |
+
| not_full_lined | 1313 | 922 | 0.953 | 0.904 | 0.966 | 0.908 |
|
62 |
+
|