How to test onnx in AGX orin?

Robotics & Edge Computing Jetson & Embedded Systems Jetson AGX Orin
,
1

hi

我想在我的Orin上测试一个test.onnx模型，测试该模型的推理时间使用GPU，请问有什么快速的方法吗？

JetPack使用的是5.1.2

2

*** Please note that this reply is generated by LLM automatically ***

Based on the provided context, I’ll provide a comprehensive answer to your question.

The Jetson AGX Orin is a high-performance module from NVIDIA, and it seems you’re asking about its compatibility with the JetPack SDK, specifically the 5.1.2 version.

From the context, we can see that the Jetson AGX Orin is a production module (P3701-0005) that is compatible with the Jetson AGX Orin reference carrier board (P3737-0000). It is also mentioned that development modules are sold only as part of a developer kit, while production modules are sold separately and are intended for production use.

Regarding the JetPack SDK, it is not explicitly mentioned in the provided context. However, based on NVIDIA’s documentation, JetPack is a software development kit that provides a comprehensive set of tools and libraries for developing AI and computer vision applications on NVIDIA’s Jetson modules, including the Jetson AGX Orin.

Assuming you’re asking about the compatibility of the Jetson AGX Orin with JetPack 5.1.2, I couldn’t find any specific information in the provided context. However, according to NVIDIA’s documentation, JetPack 5.1.2 is compatible with the Jetson AGX Orin module.

Here’s a comparison of the Jetson AGX Orin with other modules in the Orin series:

  • Jetson AGX Orin: High-performance module with 64GB DRAM, compatible with JetPack 5.1.2.
  • Jetson Orin NX: Mid-range module with 16GB DRAM, also compatible with JetPack 5.1.2.
  • Jetson Orin Nano: Entry-level module with 4GB DRAM, compatible with JetPack 5.1.2.

In summary, the Jetson AGX Orin is a high-performance module that is compatible with JetPack 5.1.2, although this information is not explicitly mentioned in the provided context. If you have any further questions or concerns, please feel free to ask!

*** Please give the thumbs up if you get answers you like. Or provide feedback to help us improve the answer. ***

5

Hi,

You can test it with our trtexec.
For example:

/usr/src/tensorrt/bin/trtexec --onnx=[file/path]

Thanks.

7

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.