Can't install onnxruntime on Jetpack 5.1.2

Robotics & Edge Computing Jetson & Embedded Systems Jetson Orin Nano
1

I downloaded the latest version of onnxruntime from Jetson Zoo but I’m getting this when installing:

$ wget https://nvidia.box.com/shared/static/iizg3ggrtdkqawkmebbfixo7sce6j365.whl -O onnxruntime_gpu-1.16.0-cp38-cp38m-linux_aarch64.whl
$ pip install onnxruntime_gpu-1.16.0-cp38-cp38m-linux_aarch64.whl 
ERROR: onnxruntime_gpu-1.16.0-cp38-cp38m-linux_aarch64.whl is not a supported wheel on this platform.

I tried both 1.15.1 and 1.16.0. Here’s the Jetson information:

$ sudo apt-cache show nvidia-jetpack
Package: nvidia-jetpack
Version: 5.1.2-b104
Architecture: arm64
Maintainer: NVIDIA Corporation
Installed-Size: 194
Depends: nvidia-jetpack-runtime (= 5.1.2-b104), nvidia-jetpack-dev (= 5.1.2-b104)
Homepage: http://developer.nvidia.com/jetson
Priority: standard
Section: metapackages
Filename: pool/main/n/nvidia-jetpack/nvidia-jetpack_5.1.2-b104_arm64.deb
Size: 29304
SHA256: fda2eed24747319ccd9fee9a8548c0e5dd52812363877ebe90e223b5a6e7e827
SHA1: 78c7d9e02490f96f8fbd5a091c8bef280b03ae84
MD5sum: 6be522b5542ab2af5dcf62837b34a5f0
Description: NVIDIA Jetpack Meta Package
Description-md5: ad1462289bdbc54909ae109d1d32c0a8

Are the Jetpack 5.1.1 Python wheels also compatible with Jetpack 5.1.2?

3

Hi,

Could you give pip3 a try?
Thanks

4 
$ pip3 install wheels/onnxruntime_gpu-1.16.0-cp310-cp310-linux_aarch64.whl 
ERROR: onnxruntime_gpu-1.16.0-cp310-cp310-linux_aarch64.whl is not a supported wheel on this platform.

Weirdly enough, it seems to work in Python 3.10:

(py38-venv) ... $ pip install wheels/onnxruntime_gpu-1.16.0-cp38-cp38m-linux_aarch64.whl 
ERROR: onnxruntime_gpu-1.16.0-cp38-cp38m-linux_aarch64.whl is not a supported wheel on this platform.

(py310-venv) ... $ pip install wheels/onnxruntime_gpu-1.16.0-cp310-cp310-linux_aarch64.whl 
Processing ./wheels/onnxruntime_gpu-1.16.0-cp310-cp310-linux_aarch64.whl
Requirement already satisfied: protobuf in ./py310-venv/lib/python3.10/site-packages (from onnxruntime-gpu==1.16.0) (4.25.1)
Requirement already satisfied: flatbuffers in ./py310-venv/lib/python3.10/site-packages (from onnxruntime-gpu==1.16.0) (23.5.26)
Requirement already satisfied: packaging in ./py310-venv/lib/python3.10/site-packages (from onnxruntime-gpu==1.16.0) (23.2)
Requirement already satisfied: coloredlogs in ./py310-venv/lib/python3.10/site-packages (from onnxruntime-gpu==1.16.0) (15.0.1)
Requirement already satisfied: sympy in ./py310-venv/lib/python3.10/site-packages (from onnxruntime-gpu==1.16.0) (1.12)
Requirement already satisfied: numpy>=1.25.2 in ./py310-venv/lib/python3.10/site-packages (from onnxruntime-gpu==1.16.0) (1.26.3)
Requirement already satisfied: humanfriendly>=9.1 in ./py310-venv/lib/python3.10/site-packages (from coloredlogs->onnxruntime-gpu==1.16.0) (10.0)
Requirement already satisfied: mpmath>=0.19 in ./py310-venv/lib/python3.10/site-packages (from sympy->onnxruntime-gpu==1.16.0) (1.3.0)
Installing collected packages: onnxruntime-gpu
Successfully installed onnxruntime-gpu-1.16.0

[notice] A new release of pip is available: 23.0.1 -> 23.3.2
[notice] To update, run: pip install --upgrade pip
(py310-venv) ... $ python3
Python 3.10.13 (main, Aug 25 2023, 13:20:03) [GCC 9.4.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import onnxruntime
>>>
5

Hi @aurelien.montmejat, that’s because you are running Python 3.10 in that virtualenv. My guess is that outside of virtualenv, your pip3 is for Python 3.8 (you can check this with pip3 --version). So try installing:

$ wget https://nvidia.box.com/shared/static/iizg3ggrtdkqawkmebbfixo7sce6j365.whl -O onnxruntime_gpu-1.16.0-cp38-cp38m-linux_aarch64.whl
$ pip3 install onnxruntime_gpu-1.16.0-cp38-cp38m-linux_aarch64.whl
6

hi,

I also tried to install onnxruntime_gpu-1.16.0-cp38-cp38m-linux_aarch64.whl on jetson orin nano and got the same error.

… $ pip3 install onnxruntime_gpu-1.16.0-cp38-cp38m-linux_aarch64.whl
Defaulting to user installation because normal site-packages is not writeable
ERROR: onnxruntime_gpu-1.16.0-cp38-cp38m-linux_aarch64.whl is not a supported wheel on this platform.

pip3 version is as follows:

… $ pip3 --version
pip 23.3.2 from /usr/local/lib/python3.8/dist-packages/pip (python 3.8)

Jetpack version is

… $ sudo apt-cache show nvidia-jetpack
Package: nvidia-jetpack
Version: 5.1.2-b104
Architecture: arm64
Maintainer: NVIDIA Corporation
Installed-Size: 194
Depends: nvidia-jetpack-runtime (= 5.1.2-b104), nvidia-jetpack-dev (= 5.1.2-b104)
Homepage: Jetson - Embedded AI Computing Platform | NVIDIA Developer
Priority: standard
Section: metapackages
Filename: pool/main/n/nvidia-jetpack/nvidia-jetpack_5.1.2-b104_arm64.deb
Size: 29304
SHA256: fda2eed24747319ccd9fee9a8548c0e5dd52812363877ebe90e223b5a6e7e827
SHA1: 78c7d9e02490f96f8fbd5a091c8bef280b03ae84
MD5sum: 6be522b5542ab2af5dcf62837b34a5f0
Description: NVIDIA Jetpack Meta Package
Description-md5: ad1462289bdbc54909ae109d1d32c0a8

7

Hi

We can install the package without issue.
Could you double-check the package? Here is the checksum for your reference:

a50941a48fbd216da67be82a5314ee5f  onnxruntime_gpu-1.16.0-cp38-cp38-linux_aarch64.whl

Thanks.

8

Hi, sorry I had a typo in my comment. I was actually trying to install the correct ONNX wheel file:

... $ pip3 --version
pip 20.0.2 from /usr/lib/python3/dist-packages/pip (python 3.8)
... $ which python3
/usr/bin/python3
... $ python3 --version
Python 3.8.10
... $ pip3 install onnxruntime_gpu-1.16.0-cp38-cp38m-linux_aarch64.whl 
ERROR: onnxruntime_gpu-1.16.0-cp38-cp38m-linux_aarch64.whl is not a supported wheel on this platform.
9

Okay so I found the issue!

Make sure that you rename the file onnxruntime_gpu-1.16.0-cp38-cp38-linux_aarch64.whl and not onnxruntime_gpu-1.16.0-cp38-cp38m-linux_aarch64.whl.

Notice the extra m. I guess it could be useful to modify the tutorial to rectify this.

@rskim

1 Like
10

Aha, thanks @aurelien.montmejat - it appears that with Python 3.8 and newer, the m is dropped from the filename. I’ve updated the tutorial to make this clearer 👍

12

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