PaddlePaddle for Jetson Nano - version 1.5.2 now available

Below are pre-built PaddlePaddle pip wheel installers for Python 3.6 on Jetson Nano with JetPack 4.2.2.

install

pip install https://github.com/yxnal/PaddlePaddle_Jetson/blob/master/paddlepaddle_gpu-1.5.2-cp36-cp36m-linux_aarch64.whl?raw=true

NOTE: the Paddle-python requirement:
requests>=2.20.0
numpy>=1.12
protobuf>=3.1.0
matplotlib<=2.2.4 ; python_version<“3.6”
scipy>=0.19.0, <=1.2.1 ; python_version<“3.5”
nltk>=3.2.2, <=3.4 ; python_version<“3.5”
matplotlib ; python_version>=“3.6”
scipy ; python_version>=“3.5”
nltk ; python_version>=“3.5”
rarfile
Pillow
graphviz
six
funcsigs
pyyaml
decorator
prettytable

Verification
To verify that PaddlePaddle has been installed correctly on your system, launch an interactive Python interpreter from terminal,
and run the following:

>>> import paddle.fluid
>>> paddle.fluid.install_check.run_check()
Your Paddle Fluid works well on SINGLE GPU or CPU.
Your Paddle Fluid works well on MUTIPLE GPU or CPU.
Your Paddle Fluid is installed successfully! Let's start deep Learning with Paddle Fluid now

Great! thanks very much!

Thanks for sharing! I’ve added it to the Nano wiki here: https://elinux.org/Jetson_Nano#Deep_Learning

If you have the build procedure you followed, would you mind posting it?