TFDetect dynamic anchor count assignment fix

[nrupatunga]

1. Currently the TFDetect layer is hardcoded to 3 anchors. change it to take care of a dynamic number of anchors in the model

🛠️ PR Summary

_{Made with ❤️ by Ultralytics Actions}

🌟 Summary

Enhancement to output tensor reshaping in TensorFlow model wrapper.

📊 Key Changes

• Modified the reshaping of the output tensor from 3 * ny * nx to self.na * ny * nx.

🎯 Purpose & Impact

• Purpose: To fix the logic for reshaping the output tensor to correctly reflect the anchor boxes count (self.na) used in the object detection model.
• Impact: This change ensures accurate output dimensions, which is critical for proper functioning of object detection. It might affect downstream tasks such as object localization and classification performance. Users working with TensorFlow models will have more reliable output, potentially improving the model's overall predictions on image data.

[github-actions]

👋 Hello @nrupatunga, 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]

@nrupatunga good catch! If we want to use the number of anchors though, we should use the Detect na attribute defined here:

yolov5/models/tf.py

Line 207 in e80a09b

self.na = len(anchors[0]) // 2 # number of anchors

[nrupatunga]

@glenn-jocher Oops! missed it altogether. That's a clean change to do. Fixed the changes.

[glenn-jocher]

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