请问怎么转onnx到rv1106，类似例子中给的1-3-85-80-80的输出

[pengyang1225]

官方给的例子，转出来是1-255-80-80，但是rv1106不支持ncwh的输出，我讲1-255-80-80 reshape到1-3-85-80-80，但是结果不对

🛠️ PR Summary

_{Made with ❤️ by Ultralytics Actions}

📊 Key Changes

• Optimization guidance for exporting YOLOv5 models to the RKNN format, with a focus on the RKNN toolkit for Rockchip NPU acceleration support.
• Introduction of several environment variable checks and condition-based initializations to enable or disable RKNN-specific model tweaks during the model export process.
• Modifications in export.py to support RKNN-targeted exports by inserting RKNN-specific layers and adjustments to the YOLOv5 focus/SPPF blocks and final layer outputs.
• Addition of a new file, README_rkopt.md, providing instructions and details on the optimizations for RKNN.
• Adjustments in models/common.py and other related files to the Focus, SPP, and SPPF blocks including changes to underlying operations and activation functions for RKNN compatibility.
• Environment variable RKNN_model_hack is introduced to activate RKNN-specific changes.

🎯 Purpose & Impact

• The changes enable YOLOv5's neural network models to be exported and optimized for deployment on Rockchip NPU hardware, which is valuable for developers aiming to deploy AI models on embedded devices or IoT applications powered by Rockchip NPUs.
• The optimizations are designed to maintain the quality of inference results while improving performance on targeted hardware.
• The documentation provided in the new README_rkopt.md helps users understand how to properly export and optimize their models using the provided scripts and underlying code changes.

🌟 Summary

This PR adds support and optimization for exporting YOLOv5 models to RKNN, aimed at boosting performance on Rockchip NPU accelerated devices. 🚀📲

[github-actions]

👋 Hello @pengyang1225, thank you for submitting a YOLOv5 🚀 PR! To allow your work to be integrated as seamlessly as possible, we advise you to:

• ✅ Verify your PR is up-to-date with ultralytics/yolov5 master branch. If your PR is behind you can update your code by clicking the 'Update branch' button or by running git pull and git merge master locally.
• ✅ Verify all YOLOv5 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

[github-actions]

👋 Hello there! We wanted to let you know that we've decided to close this pull request due to inactivity. We appreciate the effort you put into contributing to our project, but unfortunately, not all contributions are suitable or aligned with our product roadmap.

We hope you understand our decision, and please don't let it discourage you from contributing to open source projects in the future. We value all of our community members and their contributions, and we encourage you to keep exploring new projects and ways to get involved.

For additional resources and information, please see the links below:

• Docs: https://docs.ultralytics.com
• HUB: https://hub.ultralytics.com
• Community: https://community.ultralytics.com

Thank you for your contributions to YOLO 🚀 and Vision AI ⭐

[glenn-jocher]

@pengyang1225 你好，感谢你的问题。对于RV1106，你可以尝试将ONNX模型输出reshape为1-3-85-80-80。另外，我建议你查看RV1106的文档，以确保你正确理解了模型需要的输入格式。希望这对你有所帮助。

[github-actions]

👋 Hello there! We wanted to let you know that we've decided to close this pull request due to inactivity. We appreciate the effort you put into contributing to our project, but unfortunately, not all contributions are suitable or aligned with our product roadmap.

We hope you understand our decision, and please don't let it discourage you from contributing to open source projects in the future. We value all of our community members and their contributions, and we encourage you to keep exploring new projects and ways to get involved.

For additional resources and information, please see the links below:

• Docs: https://docs.ultralytics.com
• HUB: https://hub.ultralytics.com
• Community: https://community.ultralytics.com

Thank you for your contributions to YOLO 🚀 and Vision AI ⭐