NvCaffeParser.h is missing inside nvcr.io/nvidia/deepstream-l4t:6.3-samples

[rishikesan]

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson )
**• DeepStream Version - 6.3
**• JetPack Version 5.1.2-b104
**• TensorRT Version 8.5.2.2
• NVIDIA GPU Driver Version (valid for GPU only)
**• Docker container used : nvcr.io/nvidia/deepstream-l4t:6.3-samples
• Issue Type( questions, new requirements, bugs)
Error
• How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing)

Compile the sample app objectDetector_Yolo which is inside the docker container nvcr.io/nvidia/deepstream-l4t:6.3-samples inside the path /opt/nvidia/deepstream/deepstream-6.3/sources/objectDetector_Yolo/
Compile command is sudo -E make -C nvdsinfer_custom_impl_Yolo

The Error was

root@ubuntu:/realtime/objectDetector_Yolo# sudo -E make -C nvdsinfer_custom_impl_Yolo
make: Entering directory '/realtime/objectDetector_Yolo/nvdsinfer_custom_impl_Yolo'
g++ -c -o nvdsinfer_yolo_engine.o -Wall -std=c++11 -shared -fPIC -Wno-error=deprecated-declarations -I../includes -I/usr/local/cuda-11.4/include nvdsinfer_yolo_engine.cpp
In file included from nvdsinfer_yolo_engine.cpp:23:
../includes/nvdsinfer_custom_impl.h:127:10: fatal error: NvCaffeParser.h: No such file or directory
  127 | #include "NvCaffeParser.h"
      |          ^~~~~~~~~~~~~~~~~
compilation terminated.
make: *** [Makefile:61: nvdsinfer_yolo_engine.o] Error 1
make: Leaving directory '/realtime/objectDetector_Yolo/nvdsinfer_custom_impl_Yolo'

When i check the file it wasn’t there inside the container

But it is there outside the container in /usr/includes

Please advice on how we can solve this

[fanzh]

please install the “Prerequisites” and start the docker according to this doc.

[rishikesan]

Below is the docker file i am running , hope that has all the prerequisites

ARG BASE_IMAGE=nvcr.io/nvidia/deepstream-l4t:6.3-samples
FROM ${BASE_IMAGE}

RUN sh /opt/nvidia/deepstream/deepstream/user_additional_install.sh

RUN mkdir /realtime

WORKDIR /realtime

RUN apt-get update && \
    apt-get install -y libsqlite3-dev nmap libatlas-base-dev \
                       ffmpeg python3-pip  openssh-server psmisc cmake sudo

RUN wget https://github.com/google/glog/archive/refs/tags/v0.6.0.zip && \
      unzip v0.6.0.zip && \
      cd glog-0.6.0/ && \
      cmake -S . -B build -G "Unix Makefiles" && \
      cmake --build build && \
      cmake --build build --target install && \
      cd .. && \
      rm -r glog-0.6.0 && \
      rm v0.6.0.zip && \
      echo 'export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/lib/' >> ~/.bashrc

RUN apt-get install -y libgstreamer-plugins-base1.0-dev libgstreamer1.0-dev \
   libgstrtspserver-1.0-dev libx11-dev librabbitmq-dev libglib2.0 libglib2.0-dev \
   rabbitmq-server redis-tools

RUN apt-get install -y gstreamer1.0-tools gstreamer1.0-nice gstreamer1.0-plugins-bad gstreamer1.0-plugins-ugly gstreamer1.0-plugins-good libgstreamer1.0-dev git libgstreamer-plugins-bad1.0-dev libsoup2.4-dev libjson-glib-dev

RUN apt-get install -y arp-scan && \
    apt-get clean && \
    rm -rf /var/lib/apt/lists/*

RUN chmod u+s /usr/sbin/arp-scan

RUN apt-get update && apt-get install -y iproute2

RUN apt install libboost-all-dev -y && \
    git clone https://github.com/alanxz/SimpleAmqpClient.git && \
    cd SimpleAmqpClient && \
    mkdir build && \
    cd build && \
    cmake .. && \
    make && \
    make install

COPY requirements_edge.txt requirements_edge.txt

RUN pip3 install -r requirements_edge.txt

RUN pip3 install jetson-stats==4.2.3

Can you please advice whether this needs any changes

[fanzh]

if you build docker image successfully, please refer to the following command-line to start the docker.

docker run --gpus all -it --rm --net=host --privileged -v /tmp/.X11-unix:/tmp/.X11-unix -e DISPLAY=$DISPLAY -w /opt/nvidia/deepstream/deepstream-6.4 nvcr.io/nvidia/deepstream:6.4-gc-triton-devel

[rishikesan]

@fanzh

/tmp/.X11-unix

Is this command works for Jetson device

[fanzh]

yes, Jeston also has this directory. it is used for display.

[rishikesan]

Even though i run the docker as below , inside the docker container i couldn’t find the file NvCaffeParser.h and when i run inside the docker container sudo -E make -C nvdsinfer_custom_impl_Yolo i get the same error as below

sudo docker run -it --rm --runtime nvidia --net=host --privileged -v /tmp/.X11-unix:/tmp/.X11-unix -v <path to the home>/realtime:/realtime -v /run/systemd/system:/run/systemd/system -v /var/run/dbus/system_bus_socket:/var/run/dbus/system_bus_socket -v /etc:/etc -v /var/lib:/var/lib registry.gitlab.com/aoye/ai-ninjas/ai_deployment_pipeline/realtime:yolo_sample bash

Error

root@ubuntu:/realtime/objectDetector_Yolo# sudo -E make -C nvdsinfer_custom_impl_Yolo
make: Entering directory '/realtime/objectDetector_Yolo/nvdsinfer_custom_impl_Yolo'
g++ -c -o nvdsinfer_yolo_engine.o -Wall -std=c++11 -shared -fPIC -Wno-error=deprecated-declarations -I../includes -I/usr/local/cuda-11.4/include nvdsinfer_yolo_engine.cpp
In file included from nvdsinfer_yolo_engine.cpp:23:
../includes/nvdsinfer_custom_impl.h:127:10: fatal error: NvCaffeParser.h: No such file or directory
  127 | #include "NvCaffeParser.h"
      |          ^~~~~~~~~~~~~~~~~
compilation terminated.
make: *** [Makefile:61: nvdsinfer_yolo_engine.o] Error 1
make: Leaving directory '/realtime/objectDetector_Yolo/nvdsinfer_custom_impl_Yolo'
root@ubuntu:/realtime/objectDetector_Yolo# exit
exit

[fanzh]

please refer to DS_docker_containers. 6.3-samples is for deployment. please use nvcr.io/nvidia/deepstream:6.3-gc-triton-devel instead.

[rishikesan]

But the container is recommended to use for dGPU not for Jetson i am working with Jetson

In the same material the container “docker pull nvcr.io/nvidia/deepstream-l4t:6.3-samples” is recommended for Jetson i think , this is the same container i am using

[fanzh]

There is no update from you for a period, assuming this is not an issue any more. Hence we are closing this topic. If need further support, please open a new one. Thanks.
one correction. it should be nvcr.io/nvidia/deepstream:6.3-triton-multiarch on Jetson. please refer to DS_docker_containers