pip install ultralytics
会在python安装目录下的Scripts目录下安装yolo和ultralytics两个工具
yolo help
Arguments received: ['yolo', 'help']. Ultralytics 'yolo' commands use the following syntax:
yolo TASK MODE ARGS
Where TASK (optional) is one of frozenset({'pose', 'obb', 'detect', 'classify', 'segment'})
MODE (required) is one of frozenset({'benchmark', 'predict', 'export', 'val', 'track', 'train'})
ARGS (optional) are any number of custom 'arg=value' pairs like 'imgsz=320' that override defaults.
See all ARGS at https://docs.ultralytics.com/usage/cfg or with 'yolo cfg'
1. Train a detection model for 10 epochs with an initial learning_rate of 0.01
yolo train data=coco8.yaml model=yolo11n.pt epochs=10 lr0=0.01
2. Predict a YouTube video using a pretrained segmentation model at image size 320:
yolo predict model=yolo11n-seg.pt source='https://youtu.be/LNwODJXcvt4' imgsz=320
3. Val a pretrained detection model at batch-size 1 and image size 640:
yolo val model=yolo11n.pt data=coco8.yaml batch=1 imgsz=640
4. Export a YOLO11n classification model to ONNX format at image size 224 by 128 (no TASK required)
yolo export model=yolo11n-cls.pt format=onnx imgsz=224,128
5. Ultralytics solutions usage
yolo solutions count or in ['heatmap', 'queue', 'speed', 'workout', 'analytics', 'trackzone', 'inference'] source="path/to/video/file.mp4"
6. Run special commands:
yolo help
yolo checks
yolo version
yolo settings
yolo copy-cfg
yolo cfg
yolo solutions help
Docs: https://docs.ultralytics.com
Solutions: https://docs.ultralytics.com/solutions/
Community: https://community.ultralytics.com
GitHub: https://github.com/ultralytics/ultralytics
预测
yolo predict model=yolo11n.pt source='
https://ultralytics.com/images/bus.jpg'
在当前目录下下载yolo11n.pt模型文件，并将预测结果放到当前目录的 runs\detect\predict 目录下
代码实现：
from ultralytics import YOLO
if __name__ == "__main__":
# Load a model
model = YOLO("yolo11s.pt") # https://github.com/ultralytics/assets/releases/tag/v8.3.0
# Perform object detection on an image
img = "data/images/1.png"
results = model(img)
results[0].show()
# Export the model to ONNX format
# path = model.export(format="onnx") # return path to exported model
v8.3.0 现有模型如下：
FastSAM-s.pt
FastSAM-x.pt
mobile_sam.pt
rtdetr-l.pt
rtdetr-x.pt
sam2.1_b.pt
sam2.1_l.pt
sam2.1_s.pt
sam2.1_t.pt
sam2_b.pt
sam2_l.pt
sam2_s.pt
sam2_t.pt
sam_b.pt
sam_l.pt
yolo11l-cls.pt
yolo11l-obb.pt
yolo11l-pose.pt
yolo11l-seg.pt
yolo11l.pt
yolo11m-cls.pt
yolo11m-obb.pt
yolo11m-pose.pt
yolo11m-seg.pt
yolo11m.pt
yolo11n-cls.pt
yolo11n-obb.pt
yolo11n-pose.pt
yolo11n-seg.pt
yolo11n.pt
yolo11s-cls.pt
yolo11s-obb.pt
yolo11s-pose.pt
yolo11s-seg.pt
yolo11s.pt
yolo11x-cls.pt
yolo11x-obb.pt
yolo11x-pose.pt
yolo11x-seg.pt
yolo11x.pt
yolo12l.pt
yolo12m.pt
yolo12n.pt
yolo12s.pt
yolo12x.pt
yolov10b.pt
yolov10l.pt
yolov10m.pt
yolov10n.pt
yolov10s.pt
yolov10x.pt
yolov3-sppu.pt
yolov3-tinyu.pt
yolov3u.pt
yolov5l6u.pt
yolov5lu.pt
yolov5m6u.pt
yolov5mu.pt
yolov5n6u.pt
yolov5nu.pt
yolov5s6u.pt
yolov5su.pt
yolov5x6u.pt
yolov5xu.pt
yolov8l-cls.pt
yolov8l-human.pt
yolov8l-obb.pt
yolov8l-oiv7.pt
yolov8l-pose.pt
yolov8l-seg.pt
yolov8l-world-cc3m.pt
yolov8l-world.pt
yolov8l-worldv2-cc3m.pt
yolov8l-worldv2.pt
yolov8l.pt
yolov8m-cls.pt
yolov8m-human.pt
yolov8m-obb.pt
yolov8m-oiv7.pt
yolov8m-pose.pt
yolov8m-seg.pt
yolov8m-world.pt
yolov8m-worldv2.pt
yolov8m.pt
yolov8n-cls.pt
yolov8n-human.pt
yolov8n-obb.pt
yolov8n-oiv7.pt
yolov8n-pose.pt
yolov8n-seg.pt
yolov8n.pt
yolov8s-cls.pt
yolov8s-human.pt
yolov8s-obb.pt
yolov8s-oiv7.pt
yolov8s-pose.pt
yolov8s-seg.pt
yolov8s-world.pt
yolov8s-worldv2.pt
yolov8s.pt
yolov8x-cls.pt
yolov8x-human.pt
yolov8x-obb.pt
yolov8x-oiv7.pt
yolov8x-pose-p6.pt
yolov8x-pose.pt
yolov8x-seg.pt
yolov8x-world.pt
yolov8x-worldv2.pt
yolov8x.pt
yolov8x6-500.pt
yolov8x6-oiv7.pt
yolov8x6.pt
yolov9c-seg.pt
yolov9c.pt
yolov9e-seg.pt
yolov9e.pt
yolov9m.pt
yolov9s.pt
yolov9t.pt
yolo_nas_l.pt
yolo_nas_m.pt
yolo_nas_s.pt
