Pytorch installation on Jetson AGX Orin

[dorukdilmen]

Hi,

I’m trying to install pytorch recording this links:

PyTorch for Jetson

Below are pre-built PyTorch pip wheel installers for Jetson Nano, TX1/TX2, Xavier, and Orin with JetPack 4.2 and newer. Download one of the PyTorch binaries from below for your version of JetPack, and see the installation instructions to run on your Jetson. These pip wheels are built for ARM aarch64 architecture, so run these commands on your Jetson (not on a host PC). You can also use the containers from jetson-containers. PyTorch pip wheels JetPack 6

PyTorch v2.3.0

JetPack 6.0 (L4T R36.2…

NVIDIA Jetson

When I do this all, I get this error:

yolo predict model=yolov8n.pt source='https://ultralytics.com/images/bus.jpg'
Traceback (most recent call last):
  File "/home/doruk/doruk/yolov8/yolov8/bin/yolo", line 5, in <module>
    from ultralytics.cfg import entrypoint
  File "/home/doruk/doruk/yolov8/ultralytics/__init__.py", line 5, in <module>
    from ultralytics.data.explorer.explorer import Explorer
  File "/home/doruk/doruk/yolov8/ultralytics/data/__init__.py", line 3, in <module>
    from .base import BaseDataset
  File "/home/doruk/doruk/yolov8/ultralytics/data/base.py", line 17, in <module>
    from ultralytics.data.utils import FORMATS_HELP_MSG, HELP_URL, IMG_FORMATS
  File "/home/doruk/doruk/yolov8/ultralytics/data/utils.py", line 19, in <module>
    from ultralytics.nn.autobackend import check_class_names
  File "/home/doruk/doruk/yolov8/ultralytics/nn/__init__.py", line 3, in <module>
    from .tasks import (
  File "/home/doruk/doruk/yolov8/ultralytics/nn/tasks.py", line 10, in <module>
    from ultralytics.nn.modules import (
  File "/home/doruk/doruk/yolov8/ultralytics/nn/modules/__init__.py", line 65, in <module>
    from .head import OBB, Classify, Detect, Pose, RTDETRDecoder, Segment, WorldDetect
  File "/home/doruk/doruk/yolov8/ultralytics/nn/modules/head.py", line 10, in <module>
    from ultralytics.utils.tal import TORCH_1_10, dist2bbox, dist2rbox, make_anchors
  File "/home/doruk/doruk/yolov8/ultralytics/utils/__init__.py", line 47, in <module>
    TORCHVISION_VERSION = importlib.metadata.version("torchvision")  # faster than importing torchvision
  File "/usr/lib/python3.8/importlib/metadata.py", line 530, in version
    return distribution(distribution_name).version
  File "/usr/lib/python3.8/importlib/metadata.py", line 503, in distribution
    return Distribution.from_name(distribution_name)
  File "/usr/lib/python3.8/importlib/metadata.py", line 177, in from_name
    raise PackageNotFoundError(name)
importlib.metadata.PackageNotFoundError: torchvision

[AastaLLL]

Hi,

importlib.metadata.PackageNotFoundError: torchvision

TorchVision and PyTorch are different libraries.
Based on the error message, TorchVision is missing in your environment.
Please follow our document above to install it.

Thanks.

[dorukdilmen]

doruk@ubuntu:~/doruk/yolov8$ source yolov8/bin/activate
(yolov8) doruk@ubuntu:~/doruk/yolov8$ python
Python 3.8.10 (default, Nov 22 2023, 10:22:35) 
[GCC 9.4.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import torchvision
>>> import torch
>>> print(torch.__version__)
2.1.0a0+41361538.nv23.06
>>> print('CUDA available: ' + str(torch.cuda.is_available()))
CUDA available: True
>>> exit()
(yolov8) doruk@ubuntu:~/doruk/yolov8$ yolo predict model=yolov8n.pt source='https://ultralytics.com/images/bus.jpg'
Traceback (most recent call last):
  File "/home/doruk/doruk/yolov8/yolov8/bin/yolo", line 5, in <module>
    from ultralytics.cfg import entrypoint
  File "/home/doruk/doruk/yolov8/ultralytics/__init__.py", line 5, in <module>
    from ultralytics.data.explorer.explorer import Explorer
  File "/home/doruk/doruk/yolov8/ultralytics/data/__init__.py", line 3, in <module>
    from .base import BaseDataset
  File "/home/doruk/doruk/yolov8/ultralytics/data/base.py", line 17, in <module>
    from ultralytics.data.utils import FORMATS_HELP_MSG, HELP_URL, IMG_FORMATS
  File "/home/doruk/doruk/yolov8/ultralytics/data/utils.py", line 19, in <module>
    from ultralytics.nn.autobackend import check_class_names
  File "/home/doruk/doruk/yolov8/ultralytics/nn/__init__.py", line 3, in <module>
    from .tasks import (
  File "/home/doruk/doruk/yolov8/ultralytics/nn/tasks.py", line 10, in <module>
    from ultralytics.nn.modules import (
  File "/home/doruk/doruk/yolov8/ultralytics/nn/modules/__init__.py", line 65, in <module>
    from .head import OBB, Classify, Detect, Pose, RTDETRDecoder, Segment, WorldDetect
  File "/home/doruk/doruk/yolov8/ultralytics/nn/modules/head.py", line 10, in <module>
    from ultralytics.utils.tal import TORCH_1_10, dist2bbox, dist2rbox, make_anchors
  File "/home/doruk/doruk/yolov8/ultralytics/utils/__init__.py", line 47, in <module>
    TORCHVISION_VERSION = importlib.metadata.version("torchvision")  # faster than importing torchvision
  File "/usr/lib/python3.8/importlib/metadata.py", line 530, in version
    return distribution(distribution_name).version
  File "/usr/lib/python3.8/importlib/metadata.py", line 503, in distribution
    return Distribution.from_name(distribution_name)
  File "/usr/lib/python3.8/importlib/metadata.py", line 177, in from_name
    raise PackageNotFoundError(name)
importlib.metadata.PackageNotFoundError: torchvision
(yolov8) doruk@ubuntu:~/doruk/yolov8$ 

[AastaLLL]

Hi,

Could you add --system-site-packages when creating the virtual environment?
Thanks.

[antonio_mejiasvello]

In the web Installing PyTorch for Jetson Platform - NVIDIA Docs " Installing PyTorch for Jetson Platform". There is a missing “;” in the command, after ‘1.16.1’. This causes to torch not being installed. If you could change this it will be good for all the newbies in Jetson.

python3 -m pip install --upgrade pip; python3 -m pip install numpy==’1.26.1’ (missing ; ) python3 -m pip install --no-cache $TORCH_INSTALL

[AastaLLL]

Hi,

Thanks for the feedback.
We will update this with our internal team.

[ismael.rivera]

Hi, did you solve the problem? I am having the same problem. I followed same steps in ultralytics docs.

(env3.8) jetson-orin@jetsonorin-desktop:~/Desktop/gpu$ python3
Python 3.8.10 (default, Nov 22 2023, 10:22:35)
[GCC 9.4.0] on linux
Type “help”, “copyright”, “credits” or “license” for more information.

import ultralytics

Traceback (most recent call last):
File “”, line 1, in
File “/home/jetson-orin/Desktop/gpu_cocacola/env3.8/lib/python3.8/site-packages/ultralytics/init.py”, line 10, in
from ultralytics.data.explorer.explorer import Explorer
File “/home/jetson-orin/Desktop/gpu_cocacola/env3.8/lib/python3.8/site-packages/ultralytics/data/init.py”, line 3, in
from .base import BaseDataset
File “/home/jetson-orin/Desktop/gpu_cocacola/env3.8/lib/python3.8/site-packages/ultralytics/data/base.py”, line 17, in
from ultralytics.data.utils import FORMATS_HELP_MSG, HELP_URL, IMG_FORMATS
File “/home/jetson-orin/Desktop/gpu_cocacola/env3.8/lib/python3.8/site-packages/ultralytics/data/utils.py”, line 19, in
from ultralytics.nn.autobackend import check_class_names
File “/home/jetson-orin/Desktop/gpu_cocacola/env3.8/lib/python3.8/site-packages/ultralytics/nn/init.py”, line 3, in
from .tasks import (
File “/home/jetson-orin/Desktop/gpu_cocacola/env3.8/lib/python3.8/site-packages/ultralytics/nn/tasks.py”, line 10, in
from ultralytics.nn.modules import (
File “/home/jetson-orin/Desktop/gpu_cocacola/env3.8/lib/python3.8/site-packages/ultralytics/nn/modules/init.py”, line 20, in
from .block import (
File “/home/jetson-orin/Desktop/gpu_cocacola/env3.8/lib/python3.8/site-packages/ultralytics/nn/modules/block.py”, line 8, in
from ultralytics.utils.torch_utils import fuse_conv_and_bn
File “/home/jetson-orin/Desktop/gpu_cocacola/env3.8/lib/python3.8/site-packages/ultralytics/utils/init.py”, line 47, in
TORCHVISION_VERSION = importlib.metadata.version(“torchvision”) # faster than importing torchvision
File “/usr/lib/python3.8/importlib/metadata.py”, line 530, in version
return distribution(distribution_name).version
File “/usr/lib/python3.8/importlib/metadata.py”, line 503, in distribution
return Distribution.from_name(distribution_name)
File “/usr/lib/python3.8/importlib/metadata.py”, line 177, in from_name
raise PackageNotFoundError(name)
importlib.metadata.PackageNotFoundError: torchvision

[AastaLLL]

Hi,

You will need to install TorchVision to run the app.
Please find the building instructions below:

PyTorch for Jetson

Below are pre-built PyTorch pip wheel installers for Jetson Nano, TX1/TX2, Xavier, and Orin with JetPack 4.2 and newer. Download one of the PyTorch binaries from below for your version of JetPack, and see the installation instructions to run on your Jetson. These pip wheels are built for ARM aarch64 architecture, so run these commands on your Jetson (not on a host PC). You can also use the containers from jetson-containers. PyTorch pip wheels JetPack 6

PyTorch v2.3.0

JetPack 6.0 (L4T R36.2…

Instructions

• Installation
  > torchvision

Thanks.