yolov8 框架自带模型体验功能

news2025/7/15 10:22:54

简介

YOLOv8 是 ultralytics 公司在 2023 年 1月 10 号开源的 YOLOv5 的下一个重大更新版本，目前支持图像分类、物体检测和实例分割任务。

YOLOv8 是一个 SOTA 模型，它建立在以前 YOLO 版本的成功基础上，并引入了新的功能和改进，以进一步提升性能和灵活性。在图像检测识别领域yolov8和yolov5是使用较多的两款框架，兼顾精度和速度。

本文讲解yolov8自带的一个实时目标检测页面的使用。页面如下，可以在页面上体验yolov8上所有的模型，包括目标检测、分类、分割、姿态、定向框。

安装

创建虚拟环境

conda create --name yolov8 python=3.10 -y

conda activate yolov8

安装pytorch

conda install pytorch2.2.0 torchvision0.17.0 torchaudio==2.2.0 pytorch-cuda=12.1 -c pytorch -c nvidia

由于numpy版本会导致报错，需要降低numpy版本

conda install numpy==1.26.2

下载工程

git clone --branch v8.2.103 https://github.com/ultralytics/ultralytics.git

使用 --branch v8.2.103 选择yolov8版本

安装工程

pip install -e . -i https://pypi.tuna.tsinghua.edu.cn/simple

测试推理

yolo predict model=yolov8n.pt source=‘ultralytics/assets/zidane.jpg’ device=0

yolo predict model=yolov8n.pt source='ultralytics/assets/zidane.jpg' device=0
Ultralytics YOLOv8.2.103 🚀 Python-3.10.16 torch-2.2.0 CUDA:0 (NVIDIA GeForce RTX 2080 Ti, 11004MiB)
YOLOv8n summary (fused): 168 layers, 3,151,904 parameters, 0 gradients, 8.7 GFLOPs

image 1/1 /nfs/user_home/lijinkui/projects/ultralytics/ultralytics/assets/zidane.jpg: 384x640 2 persons, 1 tie, 86.8ms
Speed: 7.0ms preprocess, 86.8ms inference, 209.4ms postprocess per image at shape (1, 3, 384, 640)
Results saved to runs/detect/predict
💡 Learn more at https://docs.ultralytics.com/modes/predict

推理结果在：Results saved to runs/detect/predict

到此为止就说明yolov8安装没有问题了。

训练

使用官方数据测试训练：

yolo detect train data=coco8.yaml model=yolov8n.pt epochs=10 imgsz=640

页面展示

yolov8自带一个推理的前端页面，下面说明开启的步骤。

首先使用 yolo help 这个命令查看yolo的帮助信息

(yolov8) lijinkui@node07:~/$ yolo help

    Arguments received: ['yolo', 'help']. Ultralytics 'yolo' commands use the following syntax:

        yolo TASK MODE ARGS

        Where   TASK (optional) is one of {'segment', 'detect', 'classify', 'pose', 'obb'}
                MODE (required) is one of {'predict', 'val', 'train', 'export', 'track', 'benchmark'}
                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=yolov8n.pt epochs=10 lr0=0.01

    2. Predict a YouTube video using a pretrained segmentation model at image size 320:
        yolo predict model=yolov8n-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=yolov8n.pt data=coco8.yaml batch=1 imgsz=640

    4. Export a YOLOv8n classification model to ONNX format at image size 224 by 128 (no TASK required)
        yolo export model=yolov8n-cls.pt format=onnx imgsz=224,128

    5. Explore your datasets using semantic search and SQL with a simple GUI powered by Ultralytics Explorer API
        yolo explorer data=data.yaml model=yolov8n.pt
    
    6. Streamlit real-time webcam inference GUI
        yolo streamlit-predict
        
    7. Run special commands:
        yolo help
        yolo checks
        yolo version
        yolo settings
        yolo copy-cfg
        yolo cfg

    Docs: https://docs.ultralytics.com
    Community: https://community.ultralytics.com
    GitHub: https://github.com/ultralytics/ultralytics