jetson orin nano 部署FastDeploy

残影飞雪2025-10-18 8:49

可能无法编译成功，或编译成功但是无法使用！

据了解飞浆不支持到jetsonr orin nano所使用的jetpack 5.x 6.x版本

先安装PaddlePaddle基础

https://blog.csdn.net/canyingfeixue/article/details/153033153?spm=1001.2014.3001.5501

https://blog.csdn.net/qq_41251963/article/details/130265105

官方编译

https://github.com/PaddlePaddle/FastDeploy/blob/release/2.2/docs/get_started/installation/nvidia_gpu.md

一、下载编译

复制代码

git clone https://github.com/PaddlePaddle/FastDeploy.git
cd FastDeploy/

export BUILD_ON_JETSON=ON
export ENABLE_VISION=ON
 
python setup.py build
python setup.py bdist_wheel

或者

bash build.sh 1 python false [80,90]

import paddle
from paddle.jit.marker import unified
# Verify GPU availability
paddle.utils.run_check()
# Verify FastDeploy custom operators compilation
from fastdeploy.model_executor.ops.gpu import beam_search_softmax

二、报错处理

安装triton

复制代码

git clone https://github.com/triton-lang/triton.git
cd triton

pip install -r python/requirements.txt # build-time dependencies
pip install -e .