NVIDIA Jetson Orin NX 16G安裝tensorflow無法使用gpu

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

我的型號是NVIDIA Jetson Orin NX 16G
python版本是3.8.10
輸入sudo jetson_release指令，可得以下
Software part of jetson-stats 4.3.1 - (c) 2024, Raffaello Bonghi
Model: NVIDIA Orin NX Developer Kit - Jetpack 5.1.3 [L4T 35.5.0]
NV Power Mode[2]: 15W
Serial Number: [XXX Show with: jetson_release -s XXX]
Hardware:

  • P-Number: p3767-0000
  • Module: NVIDIA Jetson Orin NX (16GB ram)
    Platform:
  • Distribution: Ubuntu 20.04 focal
  • Release: 5.10.192-tegra
    jtop:
  • Version: 4.3.1
  • Service: Active
    Libraries:
  • CUDA: 11.8.89
  • cuDNN: 8.6.0.166
  • TensorRT: 8.5.2.2
  • VPI: 2.4.8
  • Vulkan: 1.3.204
  • OpenCV: 4.5.4 - with CUDA: NO
  • 我是用pip3 install --upgrade --extra-index-url Index of /compute/redist/jp/v511 tensorflow來安裝tensorflow的，安裝過程中沒看到跳出錯誤或者警告，但是在我運行以下程式時
  • import tensorflow as tf
  • from tensorflow.python.client import device_lib
  • print(device_lib.list_local_devices())
  • print("Num GPUs Available: ", len(tf.config.list_physical_devices(‘GPU’)))
  • 結果是[name: “/device:CPU:0”
    device_type: “CPU”
    memory_limit: 268435456
    locality {
    }
    incarnation: 6290729694689125584
    xla_global_id: -1
    ]
    Num GPUs Available: 0

我是有什麼沒注意到導致我沒辦法再用tensorflow時使用到gpu?
備註: 我是用另一台linux筆電來刷機這台NVIDIA Jetson Orin NX(用sdk)

4

Hi,

Do you manually upgrade the CUDA into 11.8?
Our package is built with the default JetPack software so it need the CUDA 11.4.

Thanks.

5

對 我手動升級cuda，11.4–>11.8
因為我之前在cuda11.4的情況下，安裝tensorflow，結果一樣是gpu無法被使用
上網找到以下這個網址 CUDA for Tegra ，裡面說要升級到11.8

6

我這裡修正一下， "CUDA for Tegra"的網址內應該沒有說gpu沒被使用到是因為cuda沒升級的問題，但是我查到說我的cuda11.4跟cudnn8.6版本不兼容，說要升級到11.8才會兼容
我想之前重刷機，結果都是無法使用gpu(tensorflow)，所以我打算升級到11.8試試看

7

如果cuda要11.4版本的話，我也確實安裝cuda==11.4.315了，結果還是沒辦法使用gpu，我該給你什麼資訊，又或者我該怎麼做才可以在NVIDIA Jetson Orin NX上使用tensorflow的gpu

8

Hi,

Could you share the command that you using for installing TensorFlow?
Thanks.

9

你好
我是用以下指令安裝 : pip3 install --upgrade --extra-index-url Index of /compute/redist/jp/v511 tensorflow

也用過 pip3 install --upgrade --extra-index-url Index of /compute/redist/jp/v513 tensorflow

補充說明 : 我安裝tensorflow之前，會輸入以下
sudo apt-get update
sudo apt-get install python3-pip
sudo -H pip install jetson-stats
sudo apt-get upgrade
sudo apt-get dist-upgrade

10

Hi,

The package works well in our side.
Please give it a try.

Setup with JetPack 5.1.4 and install TensorFlow with:

$ sudo pip3 install -U testresources setuptools==65.5.0
$ sudo pip3 install -U numpy==1.23.1 future==0.18.2 mock==3.0.5 keras_preprocessing==1.1.2 keras_applications==1.0.8 gast==0.4.0 protobuf pybind11 cython pkgconfig packaging h5py==3.7.0
$ sudo pip3 install --extra-index-url https://developer.download.nvidia.com/compute/redist/jp/v512 tensorflow==2.12.0+nv23.06

Test:

$ python3
Python 3.8.10 (default, Nov  7 2024, 13:10:47) 
[GCC 9.4.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> from tensorflow.python.client import device_lib
>>> print(device_lib.list_local_devices())

[name: "/device:CPU:0"
device_type: "CPU"
memory_limit: 268435456
locality {
}
incarnation: 5049998604901546960
xla_global_id: -1
, name: "/device:GPU:0"
device_type: "GPU"
memory_limit: 25924761856
locality {
  bus_id: 1
  links {
  }
}
incarnation: 7857006595459309742
physical_device_desc: "device: 0, name: Xavier, pci bus id: 0000:00:00.0, compute capability: 7.2"
xla_global_id: 416903419
]
>>>

Thanks.

11

您好

你的方法讓我可以使用到gpu了

感謝

12

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