Updated Segmentation and Classification usage

[glenn-jocher]

Signed-off-by: Glenn Jocher glenn.jocher@ultralytics.com

🛠️ PR Summary

_{Made with ❤️ by Ultralytics Actions}

🌟 Summary

Enhanced export script with dynamic path and script handling for different model types.

📊 Key Changes

• Automatic detection of the model type (detection, segmentation, classification) during export.
• Dynamic generation of the output directory and prediction script based on the model type.
• Tailored output messages for each model type, providing specific instructions for detection, validation, and visualization.

🎯 Purpose & Impact

• Purpose: Streamline the export process for various types of models (detection, segmentation, classification) within the YOLOv5 framework.
• Impact: Users can now export models with greater ease, receiving model-specific instructions that reduce the chance of errors and improve the overall user experience. It simplifies the workflow for developers working with different kinds of models, promoting a more structured and effortless deployment.