YOLO Object Detection

pip install ultralytics

from ultralytics import YOLO

model = YOLO("yolov8n.pt")           # n = nano, পরের s/m/l/x
results = model("dhaka_street.jpg")  # বা video, webcam (0), URL

for r in results:
    r.save("out.jpg")                # bbox আঁকা ছবি
    for box, cls, conf in zip(r.boxes.xyxy, r.boxes.cls, r.boxes.conf):
        print(model.names[int(cls)], float(conf), box.tolist())

from ultralytics import YOLO
import cv2

model = YOLO("yolov8n.pt")
cap = cv2.VideoCapture(0)
while True:
    ok, frame = cap.read()
    if not ok: break
    res = model(frame, verbose=False, imgsz=640, conf=0.4)[0]
    cv2.imshow("yolo", res.plot())
    if cv2.waitKey(1) == 27: break

data/
├── images/
│   ├── train/  001.jpg ...
│   └── val/
├── labels/
│   ├── train/  001.txt   (class cx cy w h — normalized 0-1)
│   └── val/
└── data.yaml

path: ./data
train: images/train
val:   images/val
names:
  0: person
  1: rickshaw
  2: bus
  3: car

from ultralytics import YOLO

model = YOLO("yolov8n.pt")           # pre-trained start
model.train(
    data="data/data.yaml",
    epochs=50, imgsz=640, batch=16,
    device=0,                        # GPU
    name="dhaka-traffic",
)
metrics = model.val()                # mAP50, mAP50-95

from ultralytics import YOLO

model = YOLO("yolov8n.pt")
for r in model.track(source="traffic.mp4", persist=True,
                     tracker="bytetrack.yaml", conf=0.4):
    for box, tid, cls in zip(r.boxes.xyxy, r.boxes.id, r.boxes.cls):
        print(int(tid), model.names[int(cls)])