0 x00: overture

 

 

Some readers recently asked how they could install a deep learning environment on Nvidia’s Jetson platform, which they’ve been working on recently. See some online tutorials to write messy, so I plan to record my configuration process, for everyone to do a reference.

 

0x01: Environment and Hardware

 

The author uses the nvidia JetsonNX hardware platform and the system environment is JetPack 4.4.1. For other platforms and versions, follow this step.

Python version 3.6.

Install the following deep learning environment:

  • TensorFlow = = 2.3.0
  • PyTorch = = 1.8.0 comes with
  • MxNet = = 1.6.0

 

0x02: Installation steps

 

Some resources needed for a deep learning development environment can be found here: elinux.org/Jetson_Zoo

 

First install the PIP tool:

sudo apt-get install python3-pip
Copy the code

Convenient acceleration, can be configured PIP

pip3 config set global.index-url https://pypi.tuna.tsinghua.edu.cn/simple
Copy the code

 

0x021: TensorFlow Installation

Here it is: elinux.org/Jetson_Zoo

Find the TF version corresponding to Jetson (JetPack 4.4.1) system

Here the author installed 2.3.0: you can click on the corresponding version link to download the spare.

Then execute:

# install prerequisites
sudo apt-get install libhdf5-serial-dev hdf5-tools libhdf5-dev zlib1g-dev zip libjpeg8-dev liblapack-dev libblas-dev gfortran

# install and upgrade pip3Sudo pip3 install -u PIP testResources setuptools=49.6.0# install the following python packagesSudo pip3 install -u numpy==1.16.1 Future ==0.18.2 mock== 3.0.5H5PY ==2.10.0 keras_preprocessing==1.1.1 Keras_applications ==1.0.8 Gast ==0.2.2 futures Protobuf Pybind11# to install TensorFlow 1.15 for JetPack 4.4:Sudo pip3 install TensorFlow 1.15# or install latest version of TensorFlow (2.3) for JetPack 4.4Sudo pip3 install TensorFlow 2.3Copy the code

 

0x022: PyTorch Installation

Here it is: elinux.org/Jetson_Zoo

Find the corresponding PyTorch version for Jetson (JetPack 4.4.1) system

Here the author installed 1.8.0: you can click the corresponding version link to download the spare.

 

Then execute:

# install OpenBLAS and OpenMPI
sudo apt-get install libopenblas-base libopenmpi-dev

# Python 3.6 (download pip wheel from above)Pip3 Install Cython pip3 install numpyCopy the code

0x023: MXNet Installation

Here it is: elinux.org/Jetson_Zoo

Find the corresponding MXNet version for the Jetson (JetPack 4.4.1) system

Here the author installed 1.6.0: you can click the corresponding version link to download the spare.

 

Then execute:

# Python 2.7Sudo apt-get install -y git build-essential libatlas-base-dev libopencv-dev graphviz sudo install# Python 3.6Sudo apt-get install -y git build-essential libatlas-base-dev libopencv-dev graphviz sudo installCopy the code

Other tools: Keras and ONNX can be found at elinux.org/Jetson_Zoo

 

If you encounter: “Illegal instruction (cpre dumped)” error, can I read this post address: xiaosongshine.blog.csdn.net/article/det…

 

0x03: Test function

 

Check out my previous post on “Taking You to AI” to see if the GPU version of the Deep learning framework is properly installed. Methods :TensorFlow, PyTorch, MXNet, PaddlePaddle

 

0 x04: reference

1.0 ：elinux.org/Jetson_Zoo