目标检测

• YOLO 系列
  • YOLO 简介
    • YOLO 是什么
    • YOLO 历史
      • YOLOv5
      • YOLOv4
      • YOLOv3
      • YOLOv2
      • YOLOv1
    • 模型描述
    • 训练
  • YOLO 教程
    • 自定义数据 📌
      • 训练
        • 创建 dataset.yaml
        • 创建标签
        • 组织目录
        • 选择一个模型
        • 训练
      • 可视化
        • 权重和偏差日志
        • 本地日志
      • 环境
      • 状态
    • Tips for Best Training Results
      • Dataset
      • Model Selection
      • Training Settings
      • Further Reading
    • Weights & Biases Logging 🆕
      • About Weights & Biases
      • Before You Start
      • First-Time Setup
      • Viewing Runs
      • Reports
      • Environments
      • Status
    • Supervisely Ecosystem 🆕
    • Table of Contents
    • 🔥 About Supervisely
    • Prerequisites
    • 🎉 YOLO v5 Apps Collection
    • For Developers
    • Contact & Questions & Suggestions
    • Multi-GPU Training
      • Before You Start
      • Training
        • Single GPU
        • Multi-GPU DataParallel Mode (⚠️ not recommended)
        • Multi-GPU DistributedDataParallel Mode (✅ recommended)
        • Notes
      • Results
      • FAQ
      • Environments
      • Status
      • Credits
    • PyTorch Hub
      • 以前的做法
      • 使用 PyTorch Hub 加载 YOLOv5
        • 简单例子
        • 实例细节
        • 推理设置
        • 输入通道
        • 类的数量
        • 强制重新加载
        • 训练
        • Base64 结果
        • JSON 结果
      • 自定义模型
    • TorchScript, ONNX, CoreML Export
      • Before You Start
      • Export a Trained YOLOv5 Model
      • TensorRT Deployment
      • Environments
      • Status
    • Test-Time Augmentation (TTA)
      • Before You Start
      • Test Normally
      • Test with TTA
      • Inference with TTA
        • PyTorch Hub TTA
      • Environments
      • Status
    • Model Ensembling
      • Before You Start
      • Test Normally
      • Ensemble Test
      • Ensemble Inference
      • Environments
      • Status
    • Model Pruning/Sparsity
      • Before You Start
      • Test YOLOv5x on COCO (default)
      • Test YOLOv5x on COCO (0.30 sparsity)
      • Environments
      • Status
    • Hyperparameter Evolution
      • Before You Start
      • 1. Initialize Hyperparameters
      • 2. Define Fitness
      • 3. Evolve
      • 4. Visualize
      • Environments
      • Status
    • Transfer Learning with Frozen Layers
      • Before You Start
      • Freeze Backbone
      • Freeze All Layers
      • Results
        • Accuracy Comparison
        • GPU Utilization Comparison
      • Environments
      • Status