# Ultralytics YOLO 🚀, AGPL-3.0 license
# Crack-seg dataset by Ultralytics
# Documentation: https://docs.ultralytics.com/datasets/segment/crack-seg/
# Example usage: yolo train data=crack-seg.yaml
# parent
# ├── ultralytics
# └── datasets
#     └── crack-seg  ← downloads here (91.2 MB)
# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
path: D:/ultralytics-seg/data/crack-seg # dataset root dir
train: train/images # train images (relative to 'path') 3717 images
val: valid/images # val images (relative to 'path') 112 images
test: test/images # test images (relative to 'path') 200 images
# Classes
names:
  0: crack

import warnings
warnings.filterwarnings('ignore')
from ultralytics import YOLO
if __name__ == '__main__':
    model = YOLO('ultralytics/cfg/models/11/yolo11-seg.yaml')
    #model.load('yolov8n.pt') # loading pretrain weights
    model.train(data='data/crack-seg.yaml',
                cache=False,
                imgsz=640,
                epochs=200,
                batch=16,
                close_mosaic=10,
                device='0',
                optimizer='SGD', # using SGD
                project='runs/train',
                name='exp',
                )

YOLO11-seg summary (fused): 265 layers, 2,834,763 parameters, 0 gradients, 10.2 GFLOPs
                 Class     Images  Instances      Box(P          R      mAP50  mAP50-95)     Mask(P          R      mAP50  mAP50-95): 100%|██████████| 7/7 [00:07<00:00,  1.06s/it]
                   all        200        249       0.83      0.784      0.816      0.632      0.746      0.707      0.673      0.228