Install python binding in deepstream-l4t-6.0.1-triton

[mmoroclash]

• Hardware Platform (Jetson / GPU). Jetson Nano
• DeepStream Version 6.0.1
• JetPack Version (valid for Jetson only) 4.6.3
• TensorRT Version 8.2.1.8
• NVIDIA GPU Driver Version (valid for GPU only). CUDA: 10.2.300
• Issue Type( questions)

I’m trying to run python-binding on the docker container deepstream-l4t-6.0.1-triton,
but it didn’t installed in the right way, here the version that i’m tring to install DeepStream_Python_Apps_Bindings_v1.1.1

but the code can’t create plugins nvstreammux, nvinferserver, nvosd ?

[yingliu]

nvstreammux is a key plugin for DeepStream, it comes with deepstream docker or as part of native installation. Can you check if this plugin exists or not after starting the docker container? Let’s be sure the basic function if fine before installing extra Python packages.

[mmoroclash]

not exist, the image is deepstream-l4t-6.0.1-triton, and i didn’t touch anything inside.

[yingliu]

@mmoroclash

the image is deepstream-l4t-6.0.1-triton

But your directory is deepstream-6.2, and you are using Jetson Nano, which can only support deepstream-6.0.1. So which docker image are you using?

[mmoroclash]

here is the command, that’s because I forgot and added -w with the options.

[fanzh]

you are using DS6.0.1 docker, there is no deepstream-6.2 path in docker container. and please refer to this method:Quickstart Guide — DeepStream 6.2 Release documentation

[mmoroclash]

@fanzh Thanks for the reply, but you can ignore the path, you are correct, this path was created because I forgot and add -w option when running the docker image, so it created that path, so you can just ignore it, the image still doesn’t have the python binding, and it raises many errors and the installation process is not clear at all, Deepstream is a really good framework, but it has very bad documentation.

anyway here is the script that i have used to solve the problem and install Python-binding for deepstream-6.0.1-triton

unset DISPLAY

apt update

# s - Base dependencies
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
    
# s - Initialization of submodules
cd /opt/nvidia/deepstream/deepstream/sources/
git clone https://github.com/NVIDIA-AI-IOT/deepstream_python_apps.git
cd /opt/nvidia/deepstream/deepstream/sources/deepstream_python_apps/
# siwch to tag v1.1.1 because that's compatible with deepstream 6.0.1
git checkout tags/v1.1.1
git describe --tags
git submodule update --init

# s - Installing Gst-python
apt-get install -y apt-transport-https ca-certificates -y
update-ca-certificates

cd /opt/nvidia/deepstream/deepstream/sources/deepstream_python_apps/3rdparty/gst-python/
./autogen.sh
make
make install

# pip3 install nproc

# s - Compiling the bindings
cd /opt/nvidia/deepstream/deepstream/sources/deepstream_python_apps/bindings
mkdir build
cd build
cmake ..  -DDS_VERSION=6.0.1 -DPYTHON_MAJOR_VERSION=3 -DPYTHON_MINOR_VERSION=6 -DPIP_PLATFORM=linux_aarch64 -DDS_PATH=/opt/nvidia/deepstream/deepstream
make -j4

# s - Installing the bindings
# Installing the pip wheel
# wget https://github.com/NVIDIA-AI-IOT/deepstream_python_apps/releases/download/v1.1.1/pyds-1.1.1-py3-none-linux_aarch64.whl
pip3 install ./pyds-*_aarch64.whl

# pip wheel troubleshooting
# If the wheel installation fails, upgrade the pip using the following command
python3 -m pip install --upgrade pip

rm /root/.cache/gstreamer-1.0/*
unset DISPLAY
rm /root/.cache/gstreamer-1.0/*

# s - Launching test-1 app
cd /opt/nvidia/deepstream/deepstream/sources/deepstream_python_apps/apps/deepstream-test1
python3 deepstream_test_1.py <input .h264 file> 

hopefully, that helps the other.