---
license: mit
datasets:
- ds4sd/DocLayNet
language:
- en
metrics:
- accuracy
pipeline_tag: object-detection
---

## How to Get Started with the Model

### Install `ultralytics YOLO` package

``` shell
$ pip install ultralytics
```

### Perform Inference as per [kurkurzz](https://github.com/kurkurzz/custom-yolov8-auto-annotation-cvat-blueprint/blob/master/main.py)

``` python
from ultralytics import YOLO
from json import dumps

checkpoint_path = "path/to/model/weight.pt" # e.g weights/best.pt in this directory
model = YOLO(checkpoint_path)

image_path = "path/to/image"
infered = model(image_path)
results = infered[0]

boxes = result.boxes.data[:,:4]
confs = result.boxes.conf
clss = result.boxes.cls
class_name = result.names

#detected = results[0].boxes.xywh # or xywhn, xyxy pr xyxyn

detections = []
threshold = 0.3 # 0 < threshold <= 1

for box, conf, cls in zip(boxes, confs, clss):
    label = class_name[int(cls)]
    if conf >= threshold:
        # must be in this format
        detections.append({
            'confidence': str(float(conf)),
            'label': label,
            'points': box.tolist(),
            'type': 'rectangle',
        })

detected_objects = dumps(detections)
print(detected_objects)
                        
```