`ultralytics 8.3.72` Fix NVIDIA Jetson DLA core support for DLA inference

Laughing-q commented 5 days ago

🛠️ PR Summary

🌟 Summary

This PR updates the CI workflow, adds detailed NVIDIA Jetson DLA hardware information, and improves TensorRT export functionality for better device compatibility. 🚀

📊 Key Changes

CI Workflow Adjustment: Removed ubuntu-24.04-arm from test matrix due to temporary issues.
Jetson Docs Update: Enhanced documentation with a detailed table of devices and their DLA core specifications, including frequency and core count.
TensorRT Export Improvements:
- Enabled explicit DLA core selection during export and inference (dla:0/dla:1).
- Added metadata handling for DLA-specific inference settings.
Version Update: Incremented version from 8.3.71 to 8.3.72.

🎯 Purpose & Impact

Improved Compatibility: Adjustments to the CI prevent workflow disruptions and ensure reliable testing on supported platforms.
Enhanced Usability: The new Jetson DLA table clarifies hardware capabilities, helping users optimize performance.
Simplified Inference: Export and inference on Jetson DLA cores are now more user-friendly and reliable.
Version Tracking: The version update reflects these changes for clearer package management.

🌟 These updates collectively improve the developer and user experience, particularly for NVIDIA Jetson users leveraging DLA hardware.

sentry-io[bot] commented 5 days ago

🔍 Existing Issues For Review

Your pull request is modifying functions with the following pre-existing issues:

📄 File: ultralytics/engine/exporter.py

Function	Unhandled Issue
`__call__`	AssertionError: ERROR ❌️ argument 'int8' is not supported for format='onnx' ultralytics.engin... `Event Count:` 1
`__call__`	RuntimeError: [enforce fail at alloc_cpu.cpp:117] err == 0. DefaultCPUAllocator: can't allocate memory: you tri... ... `Event Count:` 1
`__call__`	SyntaxError: /content/drive/MyDrive/weedai_yolo/runs/detect/train/args.yaml 'train:' key missing ❌. ... `Event Count:` 1

_{Did you find this useful? React with a 👍 or 👎}

UltralyticsAssistant commented 5 days ago

👋 Hello @Laughing-q, thank you for submitting an ultralytics/ultralytics 🚀 PR! To ensure a seamless integration of your work, please review the following checklist:

✅ Define a Purpose: Clearly explain the purpose of your fix or feature in your PR description, and link to any relevant issues. Ensure your commit messages are clear, concise, and adhere to the project's conventions.
✅ Synchronize with Source: Confirm your PR is synchronized with the ultralytics/ultralytics main branch. If it's behind, update it by clicking the 'Update branch' button or by running git pull and git merge main locally.
✅ Ensure CI Checks Pass: Verify all Ultralytics Continuous Integration (CI) checks are passing. If any checks fail, please address the issues.
✅ Update Documentation: Update the relevant documentation for any new or modified features.
✅ Add Tests: If applicable, include or update tests to cover your changes, and confirm that all tests are passing.
✅ Sign the CLA: Please ensure you have signed our Contributor License Agreement if this is your first Ultralytics PR by writing "I have read the CLA Document and I sign the CLA" in a new message.
✅ Minimize Changes: Limit your changes to the minimum necessary 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

🔍 Additionally: It looks like your PR focuses on improving support for NVIDIA Deep Learning Accelerator (DLA) devices, which is an exciting enhancement! 🎉 For clarity and completeness, if there are any edge cases or additional behavior that your code addresses (e.g., specific scenarios involving DLA devices), you may want to include them in your testing or documentation updates.

For more guidance, please refer to our Contributing Guide. An Ultralytics engineer will review your PR soon and provide further feedback if necessary. Don’t hesitate to leave a comment if you have any questions. Thank you for contributing to Ultralytics! ⚡

codecov[bot] commented 5 days ago

Codecov Report

Attention: Patch coverage is 33.33333% with 4 lines in your changes missing coverage. Please review.

Project coverage is 73.09%. Comparing base (84a8b06) to head (e25d9b5). Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
ultralytics/nn/autobackend.py	0.00%	3 Missing :warning:
ultralytics/engine/exporter.py	50.00%	1 Missing :warning:

Additional details and impacted files

```diff @@ Coverage Diff @@ ## main #19078 +/- ## ========================================== + Coverage 73.08% 73.09% +0.01% ========================================== Files 129 129 Lines 17485 17486 +1 ========================================== + Hits 12779 12782 +3 + Misses 4706 4704 -2 ``` | [Flag](https://app.codecov.io/gh/ultralytics/ultralytics/pull/19078/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/19078/flags?src=pr&el=flag&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=ultralytics) | `34.10% <33.33%> (+0.01%)` | :arrow_up: | | [GPU](https://app.codecov.io/gh/ultralytics/ultralytics/pull/19078/flags?src=pr&el=flag&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=ultralytics) | `37.94% <16.66%> (-0.01%)` | :arrow_down: | | [Tests](https://app.codecov.io/gh/ultralytics/ultralytics/pull/19078/flags?src=pr&el=flag&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=ultralytics) | `66.91% <33.33%> (+<0.01%)` | :arrow_up: | 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.

lakshanthad commented 5 days ago

@Laughing-q thanks for the support here! I have tested on Jetson and it all works now! @glenn-jocher this should be good to go.

Laughing-q commented 5 days ago

@glenn-jocher a fix of this PR. In this PR it'll automatically save the dla in our metadata when exporting a engine format model within dla specified and directly use it in AutoBackend class so model gets to know the correct dla device. No need to specify device while inference.

glenn-jocher commented 5 days ago

@Laughing-q looks good! @lakshanthad can you review? Do we need any Docs updates after this?

UltralyticsAssistant commented 4 days ago

🥳 Amazing work, team! Huge thanks to @Laughing-q for leading this fantastic PR and to @glenn-jocher and @lakshanthad for the collaborative efforts behind the scenes. Merging this is a big step forward for improved compatibility and usability, especially for our NVIDIA Jetson community. 🚀

As Henry Ford once said, "Coming together is a beginning, staying together is progress, and working together is success." This PR embodies that spirit, combining teamwork and innovation to unlock new potentials for DLA-specific workflows.

Your hard work and dedication continue to inspire. Thank you for your meaningful contributions! ❤️

ultralytics / ultralytics