Save *.npy features on detect.py --visualize

[Zengyf-CVer]

Add feature map to save npy files,export npy files with 32 feature maps per layer.

🛠️ PR Summary

_{Made with ❤️ by Ultralytics Actions}

🌟 Summary

Enhancements in feature visualization and data saving within YOLOv5.

📊 Key Changes

• 📁 Modified the path handling for saving feature visualization images, ensuring they are stored in the correct directory.
• 🖼️ Feature visualization now includes saving the features as .npy files for easier post-processing or analysis.

🎯 Purpose & Impact

• 🚀 The updated path handling prevents potential issues with saving files in unintended locations, improving the reliability of the feature visualization functionality.
• 🔍 Saving features as .npy files provides users with more flexibility, allowing them to more easily use the visualized data for further research or analysis.

[github-actions]

👋 Hello @Zengyf-CVer, thank you for submitting a 🚀 PR! To allow your work to be integrated as seamlessly as possible, we advise you to:

• ✅ Verify your PR is up-to-date with upstream/master. If your PR is behind upstream/master an automatic GitHub actions rebase may be attempted by including the /rebase command in a comment body, or by running the following code, replacing 'feature' with the name of your local branch:

git remote add upstream https://github.com/ultralytics/yolov5.git
git fetch upstream
git checkout feature  # <----- replace 'feature' with local branch name
git merge upstream/master
git push -u origin -f

• ✅ Verify all Continuous Integration (CI) checks are passing.
• ✅ Reduce changes to the absolute minimum required for your bug fix or feature addition. "It is not daily increase but daily decrease, hack away the unessential. The closer to the source, the less wastage there is." -Bruce Lee

[glenn-jocher]

@Zengyf-CVer if you want to add numpy outputs, it probably makes sense to export a single multi-dimensional array npy file per image rather than many npy files. i.e. instead of exporting 30 10x10 arrays you could export a single 30x10x10 array, and then no new directories are needed.

[Zengyf-CVer]

@glenn-jocher
Thank you very much for your suggestion, I will look into it again. thanks.

[Zengyf-CVer]

@glenn-jocher
I merged the 32 feature maps of each layer into one npy file and removed the folder.

[glenn-jocher]

@Zengyf-CVer hey great!

[glenn-jocher]

/rebase

[glenn-jocher]

@Zengyf-CVer I've cleaned this up so the PR only adds +1 lines of code to the repo.

[glenn-jocher]

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