Optimize NVIDIA Jetson Docs

[lakshanthad]

@glenn-jocher

Now all model exports are working with NVIDIA Jetson Orin.

• Added benchmarks for all the model exports
• Added onnxruntime-gpu install
• Make minor changes to format

🛠️ PR Summary

_{Made with ❤️ by Ultralytics Actions}

🌟 Summary

Updates and enhancements for NVIDIA Jetson setup and Parking Management documentation.

📊 Key Changes

• Updated NVIDIA Jetson documentation to include clearer setup instructions for Docker and non-Docker environments.
• Added details on installing Ultralytics package with optional dependencies on NVIDIA Jetson.
• Provided steps for PyTorch and Torchvision installation compatible with Jetson's ARM64 architecture.
• Added notes on manually installing onnxruntime-gpu due to a lack of aarch64 binaries on PyPI for Jetson.
• Introduced a note to downgrade numpy to version 1.23.5 after installing onnxruntime-gpu.
• Detailed instructions for using TensorRT on NVIDIA Jetson for improved performance.
• Included a comprehensive NVIDIA Jetson Orin YOLOv8 benchmarks comparison chart.
• Reorganized Parking Management documentation for clarity by fixing section references.

🎯 Purpose & Impact

• Enhances setup guides: The updated documentation makes setting up Ultralytics on NVIDIA Jetson devices more straightforward, aiding both new and existing users in getting started quickly.
• Improves performance: Instructions for TensorRT usage can significantly boost inference performance on Jetson devices, making it valuable for projects requiring high-speed object detection.
• Clarifies documentation: The swaps in the Parking Management sections ensure the documentation is more intuitive and easier to navigate.
• Broad audience utility: By simplifying the setup process and detailing performance optimizations, these changes benefit both developers and users interested in deploying AI models on embedded systems.

[codecov[bot]]

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Project coverage is 69.66%. Comparing base (40e5ac3) to head (f29cdb7).

Additional details and impacted files

```diff @@ Coverage Diff @@ ## main #11242 +/- ## ========================================== - Coverage 70.66% 69.66% -1.01% ========================================== Files 122 122 Lines 15591 15591 ========================================== - Hits 11018 10862 -156 - Misses 4573 4729 +156 ``` | [Flag](https://app.codecov.io/gh/ultralytics/ultralytics/pull/11242/flags?src=pr&el=flags&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=ultralytics) | Coverage Δ | | |---|---|---| | [Benchmarks](https://app.codecov.io/gh/ultralytics/ultralytics/pull/11242/flags?src=pr&el=flag&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=ultralytics) | `35.38% <ø> (-0.26%)` | :arrow_down: | | [GPU](https://app.codecov.io/gh/ultralytics/ultralytics/pull/11242/flags?src=pr&el=flag&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=ultralytics) | `37.36% <ø> (ø)` | | | [Tests](https://app.codecov.io/gh/ultralytics/ultralytics/pull/11242/flags?src=pr&el=flag&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=ultralytics) | `65.10% <ø> (-1.65%)` | :arrow_down: | Flags with carried forward coverage won't be shown. [Click here](https://docs.codecov.io/docs/carryforward-flags?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=ultralytics#carryforward-flags-in-the-pull-request-comment) to find out more.

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

[johnnynunez]

If you want compile onnxruntime-gpu for jetson. @glenn-jocher I've some tutorials https://hackmd.io/ZmWQz8azTdWNVoCc9Bf3QA?view and Dustin upload whl: http://jetson.webredirect.org/

git clone --recursive https://github.com/microsoft/onnxruntime
cd onnxruntime 

export PATH="/usr/local/cuda-12.4/bin:${PATH}"
export CUDACXX="/usr/local/cuda-12.4/bin/nvcc"

pip3 install -U packaging
./build.sh --config Release --update --build --parallel --build_wheel \
--use_tensorrt --cuda_home /usr/local/cuda-12.4 --cudnn_home /usr/lib/aarch64-linux-gnu \
--tensorrt_home /usr/lib/aarch64-linux-gnu

[lakshanthad]

@Burhan-Q @glenn-jocher Thank you for reviewing. I have made the changes accordingly. It should be ready to merge now.

[glenn-jocher]

@lakshanthad looks much better, thanks for the fixes!!

[glenn-jocher]

@lakshanthad @Burhan-Q PR merged! Nice work here guys!

[lakshanthad]

Thank you @glenn-jocher !