Improved Usage example docstrings

[glenn-jocher]

🛠️ PR Summary

_{Made with ❤️ by Ultralytics Actions}

🌟 Summary

Enhancements made to enable YOLOv5 classification and detection on more media types and export to various formats.

📊 Key Changes

• Classification and Detection: Expanded the ability of YOLOv5's classification and detection to handle various sources like images and videos, and formats including PyTorch, ONNX, TensorFlow, CoreML, etc.
• Training and Validation: Updated usage instructions for training and validating classification and detection models, highlighting single-GPU and multi-GPU distributed data parallel (DDP) training.
• Export Functionality: Clarified export script usage to support different model formats for subsequent inference.
• Documentation Clarity: Improved instructions and comments in scripts to facilitate user understanding of models, datasets, and general usage.

🎯 Purpose & Impact

• Versatility: Users can now more easily use YOLOv5 for classification/detection across different formats and sources, making the tool more flexible. 🔄
• Usability: Updated scripts and clearer descriptions improve accessibility for new users and reduce potential confusion. 💡
• Performance: Multi-GPU DDP training support enhances performance, allowing for faster and more efficient training on larger datasets. ⚙️
• Model Deployment: Expanded export options make it easier for users to deploy YOLOv5 models across a wider range of platforms and devices. 🚀