NVIDIA Jetson Xavier NX developer kit | deepstream python示例程序与绑定文件（applications and bindings）

1. 准备gst-python

桌面打开terminal，运行。

安装python3的基础包，避免安装gst-python时出现找不到头文件的错误。

sudo apt-get install python3-dev libpython3-dev

 安装gst-python

$ sudo apt-get install python-gi-dev
$ export GST_LIBS="-lgstreamer-1.0 -lgobject-2.0 -lglib-2.0"
# 如果是x86_64系统，把aarch64替换掉
$ export GST_CFLAGS="-pthread -I/usr/include/gstreamer-1.0 -I/usr/include/glib-2.0 -I/usr/lib/aarch64-linux-gnu/glib-2.0/include"
$ git clone https://github.com/GStreamer/gst-python.git
$ cd gst-python
$ git checkout 1a8f48a  #根据版本有所不同
$ ./autogen.sh PYTHON=python3
$ ./configure PYTHON=python3
$ make
$ sudo make install

2. 下载deepstream的python文件与安装python bindings

下载

cd /opt/nvidia/deepstream/deepstream-6.0/sources
sudo su
git clone https://github.com/NVIDIA-AI-IOT/deepstream_python_apps

安装bindings

2.1 安装基础依赖

#直接复制这两行代码，粘贴运行
apt install -y git python-dev python3 python3-pip python3.6-dev python3.8-dev cmake g++ build-essential \
    libglib2.0-dev libglib2.0-dev-bin python-gi-dev libtool m4 autoconf automake libgirepository1.0-dev libcairo2-dev

2.2 初始化子模型

bash
cd /opt/nvidia/deepstream/deepstream/sources/deepstream_python_apps/
git submodule update --init

 2.3 安装gst-python

# 更新证书
bash
sudo apt-get install -y apt-transport-https ca-certificates -y
sudo update-ca-certificates
# 安装
bash
cd 3rdparty/gst-python/
./autogen.sh
make
sudo make install

3. 安装与运行sample applicaitions

在sources文件夹下创立了deepstream_python_apps文件夹，进入其中。运行每个模型请查看模型文件夹中的readme文件。参考下篇文章

但模型运行仍需下载其他的前提条件，在此一并介绍安装步骤。

3.1 安装python包：opencv与numpy

$ sudo apt update
$ sudo apt install python3-numpy python3-opencv -y

3.2 安装 GstRtspServer and instrospection typelib 

$ sudo apt update
$ sudo apt-get install libgstrtspserver-1.0-0 gstreamer1.0-rtsp

3.3 安装 the introspection typelib package

$ sudo apt-get install libgirepository1.0-dev
$ sudo apt-get install gobject-introspection gir1.2-gst-rtsp-server-1.0

3.4 安装pyds包

# 在deepstream_python_apps的bindings中运行terminal
mkdir build
cd build
cmake ..  -DDS_VERSION=6.0 \
    -DPYTHON_MAJOR_VERSION=3 -DPYTHON_MINOR_VERSION=6 \
    -DPIP_PLATFORM=linux_aarch64 -DDS_PATH=/opt/nvidia/deepstream/deepstream
make

#注意修改pyds版本号和系统类型，可在bindings/build文件夹中查看这个whl文件
sudo pip3 install ./pyds-1.1.3-py3-none-linux_aarch64.whl

参考：Nvidia Deepstream极致细节：0. Deepstream 6.0以及Deepstream Python 1.1.0的安装_破浪会有时的博客-CSDN博客

3.5 下载peoplenet model等模型

参考/opt/nvidia/deepstream/deepstream/samples/configs/tao_pretrained_models的readme

#下载配置文件
$ cd /opt/nvidia/deepstream/deepstream/sources/apps/sample_apps
$ sudo git clone https://github.com/NVIDIA-AI-IOT/deepstream_reference_apps.git
$ cd /opt/nvidia/deepstream/deepstream/
$ sudo cp \
  sources/apps/sample_apps/deepstream_reference_apps/deepstream_app_tao_configs/* \
  samples/configs/tao_pretrained_models/

#下载模型（有时网络连接与下载速度较慢）
$ sudo apt install -y wget zip
$ cd /opt/nvidia/deepstream/deepstream/samples/configs/tao_pretrained_models/
$ sudo ./download_models.sh

#完成，运行可参考/opt/nvidia/deepstream/deepstream-6.0/samples/configs/tao_pretrained_models/README

运行时可能会有报错，如果README阅读不细致会发生一些版本配置的问题，如deepstream6.0系统，但配置文件中默认是deepstream6.1，需要进入配置文件中修改。

3.6 安装triton接口

# 前往triton接口安装文件的所在地址
cd /opt/nvidia/deepstream/deepstream-6.0/samples

# 运行triton接口支持的安装sh文件
sudo ./triton_backend_setup.sh

# 准备分类视频样例，或者不运行该部分命令，把自己的分类视频复制到samples/streams/地址下
sudo apt-get update && sudo apt-get install ffmpeg
./prepare_classification_test_video.sh

#准备tensorRT，tensorflow，ONNX models，将下载到samples/triton-model-repo
./prepare_ds_triton_model_repo.sh

#完成，运行可参考/opt/nvidia/deepstream/deepstream-6.0/samples/configs/deepstream-app-triton/README

3.7 message broker相关

# Azure Iot:
    $ sudo apt-get install -y libcurl4-openssl-dev libssl-dev uuid-dev libglib2.0 libglib2.0-dev

# Kafka:
    $ sudo apt-get install libglib2.0 libglib2.0-dev
    $ sudo apt-get install libjansson4  libjansson-dev
    $ sudo apt-get install librdkafka1=0.11.3-1build1

# AMQP (rabbitmq):
    $ git clone -b v0.8.0  --recursive https://github.com/alanxz/rabbitmq-c.git
    $ cd rabbitmq-c
    $ mkdir build && cd build
    $ cmake ..
    $ cmake --build .
    $ sudo cp librabbitmq/librabbitmq.so.4 /opt/nvidia/deepstream/deepstream-<version>/lib/
    $ sudo ldconfig