Shuffle all 4(or 9) images in mosaic augmentation

[kimnamu]

Thank you for sharing nice open-source codes 👍

This PR is. for shuffling the order of all 4(or 9) images in mosaic augmentation

Currently, the order of images in mosaic augmentation is not completely random.
The remaining images except the first are randomly arranged. Apply shuffle all to increase the diversity of data composition.

🛠️ PR Summary

_{Made with ❤️ by Ultralytics Actions}

🌟 Summary

Enhanced image indexing for data augmentation in YOLOv5 🖼️✨

📊 Key Changes

• Added random.shuffle(indices) in the load_mosaic and load_mosaic9 methods.

🎯 Purpose & Impact

• Purpose: To introduce more randomness in selecting images for mosaic data augmentation, which stitches together multiple images to create a new training sample.
• Impact: This change likely leads to better model generalization due to increased variability in training data, potentially improving the performance of YOLOv5 object detection models. 🚀

[glenn-jocher]

@kimnamu great idea! This will add a bit more randomness to the augmentation.

[glenn-jocher]

@kimnamu PR is merged. Thank you for your contributions to YOLOv5 🚀 and Vision AI ⭐