Hi, I am currently running a “deepstream-6.2” docker image.
I am trying to use “paddlepaddle” inside the docker image using “pip install paddlepaddle-gpu”, but then I get a “Illegal instruction (core dumped)” error whenever I try to run it.
While going through the nvidia documents, I came across a “paddlepaddle image” .
If there is a way, that I could use “paddlepaddle” inside the “deepstream-6.2” without the image, or is there any workaround?
There is a paddlepaddle docker already: PaddlePaddle | NVIDIA NGC, is it workable for your case?
The deepstream docker is mainly used for inference, there may be more dependencies to be tackled for paddlepaddle.
When I install “pip3 install paddlepaddle-gpu” inside deepstream-6.2 docker container I’m getting “Illegal instruction (core dumped)”. How we can install successfully ?
Just checked it on a server with docker nvcr.io/nvidia/deepstream:6.2-devel and confirm it works, you may refer the logs. Can “nvidia-smi” work in your docker? Not sure if the coredump is relevant with GPU runtime.
Besides, why not use two different docker, one for deepstream and one for paddlepaddle? Deepstream focuses on inference while paddlepaddle focus on model creation and training, quite differenct use case for the two dockers.
root@test:~# pip3 install paddlepaddle-gpu
Collecting paddlepaddle-gpu
Downloading paddlepaddle_gpu-2.5.1-cp38-cp38-manylinux1_x86_64.whl (541.0 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 541.0/541.0 MB 159.2 kB/s eta 0:00:00
Collecting decorator
Downloading decorator-5.1.1-py3-none-any.whl (9.1 kB)
Collecting Pillow
Downloading Pillow-10.0.0-cp38-cp38-manylinux_2_28_x86_64.whl (3.4 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 3.4/3.4 MB 404.6 kB/s eta 0:00:00
Collecting httpx
Downloading httpx-0.24.1-py3-none-any.whl (75 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 75.4/75.4 kB 775.9 kB/s eta 0:00:00
Collecting astor
Downloading astor-0.8.1-py2.py3-none-any.whl (27 kB)
Collecting opt-einsum==3.3.0
Downloading opt_einsum-3.3.0-py3-none-any.whl (65 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 65.5/65.5 kB 742.4 kB/s eta 0:00:00
Collecting paddle-bfloat==0.1.7
Downloading paddle_bfloat-0.1.7-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (385 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 385.5/385.5 kB 639.1 kB/s eta 0:00:00
Collecting numpy>=1.13
Downloading numpy-1.24.4-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (17.3 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 17.3/17.3 MB 326.6 kB/s eta 0:00:00
Collecting protobuf>=3.20.2
Downloading protobuf-4.23.4-cp37-abi3-manylinux2014_x86_64.whl (304 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 304.5/304.5 kB 280.3 kB/s eta 0:00:00
Requirement already satisfied: certifi in /usr/lib/python3/dist-packages (from httpx->paddlepaddle-gpu) (2019.11.28)
Requirement already satisfied: idna in /usr/lib/python3/dist-packages (from httpx->paddlepaddle-gpu) (2.8)
Collecting httpcore<0.18.0,>=0.15.0
Downloading httpcore-0.17.3-py3-none-any.whl (74 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 74.5/74.5 kB 382.8 kB/s eta 0:00:00
Collecting sniffio
Downloading sniffio-1.3.0-py3-none-any.whl (10 kB)
Collecting h11<0.15,>=0.13
Downloading h11-0.14.0-py3-none-any.whl (58 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 58.3/58.3 kB 383.2 kB/s eta 0:00:00
Collecting anyio<5.0,>=3.0
Downloading anyio-3.7.1-py3-none-any.whl (80 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 80.9/80.9 kB 379.4 kB/s eta 0:00:00
Collecting exceptiongroup
Downloading exceptiongroup-1.1.2-py3-none-any.whl (14 kB)
Installing collected packages: paddle-bfloat, sniffio, protobuf, Pillow, numpy, h11, exceptiongroup, decorator, astor, opt-einsum, anyio, httpcore, httpx, paddlepaddle-gpu
Successfully installed Pillow-10.0.0 anyio-3.7.1 astor-0.8.1 decorator-5.1.1 exceptiongroup-1.1.2 h11-0.14.0 httpcore-0.17.3 httpx-0.24.1 numpy-1.24.4 opt-einsum-3.3.0 paddle-bfloat-0.1.7 paddlepaddle-gpu-2.5.1 protobuf-4.23.4 sniffio-1.3.0
WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv
[notice] A new release of pip available: 22.3.1 -> 23.2.1
[notice] To update, run: python3 -m pip install --upgrade pip
root@test:~#
PaddlePaddle does inference as well. I have a requirement to use it inside the deepstream-image.
There is no update from you for a period, assuming this is not an issue anymore. Hence we are closing this topic. If need further support, please open a new one. Thanks
So did you manage to install paddlepaddle-gpu in your deepstream docker?
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.