Quantcast
Channel: Recent Questions - Stack Overflow
Viewing all articles
Browse latest Browse all 12111

Container cannot reach the installed drivers

$
0
0

This is the error Docker is giving me:

docker: Error response from daemon: could not select device driver "" with capabilities: [[gpu]].

This is the output of my nvidia-smi:

Sun Apr  7 05:25:10 2024       +---------------------------------------------------------------------------------------+| NVIDIA-SMI 535.161.07             Driver Version: 535.161.07   CUDA Version: 12.2     ||-----------------------------------------+----------------------+----------------------+| GPU  Name                 Persistence-M | Bus-Id        Disp.A | Volatile Uncorr. ECC || Fan  Temp   Perf          Pwr:Usage/Cap |         Memory-Usage | GPU-Util  Compute M. ||                                         |                      |               MIG M. ||=========================================+======================+======================||   0  NVIDIA GeForce GTX 1650        Off | 00000000:01:00.0 Off |                  N/A || N/A   42C    P0               7W /  50W |      9MiB /  4096MiB |      0%      Default ||                                         |                      |                  N/A |+-----------------------------------------+----------------------+----------------------++---------------------------------------------------------------------------------------+| Processes:                                                                            ||  GPU   GI   CI        PID   Type   Process name                            GPU Memory ||        ID   ID                                                             Usage      ||=======================================================================================||    0   N/A  N/A    830758      G   /usr/lib/xorg/Xorg                            4MiB |+---------------------------------------------------------------------------------------+

And this is the output of nvcc --version:

nvcc: NVIDIA (R) Cuda compiler driverCopyright (c) 2005-2021 NVIDIA CorporationBuilt on Thu_Nov_18_09:45:30_PST_2021Cuda compilation tools, release 11.5, V11.5.119Build cuda_11.5.r11.5/compiler.30672275_0

For context, this is the Dockerfile:

# Use NVIDIA CUDA 12.0.0 development image based on Ubuntu 22.04 as the base imageFROM nvidia/cuda:12.4.0-devel-ubuntu22.04# Set non-interactive mode to avoid prompts during package installationENV DEBIAN_FRONTEND=noninteractive# Update the package list and install python3-pipRUN apt-get updateRUN apt-get install -y --no-install-recommends git curl wget python3 python3-pip python3-dev libatlas-base-dev nvidia-cuda-toolkit && rm -rf /var/lib/apt/lists/*# Set the working directory in the containerWORKDIR /app# Copy the requirements.txt file to the containerCOPY job.sh /app/COPY mistral-7b.py /app/COPY requirements.txt /app/RUN pip3 install -U bitsandbytesRUN pip3 install -U git+https://github.com/huggingface/transformers.gitRUN pip3 install -U git+https://github.com/huggingface/peft.gitRUN pip3 install -U git+https://github.com/huggingface/accelerate.gitRUN pip3 install trl xformers wandb datasets einops sentencepieceRUN pip3 install --upgrade --force-reinstall numpy# Command to run your applicationCMD ["python3", "mistral-7b.py"]

Where am I going wrong?


Viewing all articles
Browse latest Browse all 12111

Trending Articles



<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>