Sign inSign up

drstoop/data_science

By drstoop

•Updated about 8 years ago

nvidia-docker run -it -p 8888:8888 -p 6006:6006 drstoop/data_science:<tag>

Image
0

175

drstoop/data_science repository overview

⁠Versions

Ubuntu 16.04, Python 3.5.2

⁠Tags:
  • tf_torch_gpu_jupyter: tensorflow-gpu 1.10.1, torch 0.4.1, cuda 9.0, cudnn 7.1
  • tf-1.4.0_torch_gpu_jupyter: tensorflow-gpu 1.4.0, torch 0.4.1, cuda 8.0, cudnn 6 (for old GPUs with cuda capability 3.0, e.g. GeForce 750M)
⁠Build files

Following files were used for building image:

  • Dockerfile
  • jupyter_notebook_config.partial.py
  • requirements.txt (here: empty file, but required for additional pip3-package-installations)

⁠Dockerfile (tf_torch_gpu_jupyter)

NOTE: Multiple RUN commands were required to ensure sequential installation prerequisits.

# docker build -t drstoop/data_science:tf_torch_gpu_jupyter .

FROM nvcr.io/nvidia/cuda:9.0-cudnn7.1-devel-ubuntu16.04

LABEL maintainer="Maximilian Berr <[email protected]>"

# append notebook and ipython config
ENV JUP_CONFIG_PARTIAL /tmp/jupyter_notebook_config.partial.py
ENV REQUIREMENTS /tmp/requirements.txt

COPY jupyter_notebook_config.partial.py ${JUP_CONFIG_PARTIAL}
COPY requirements.txt ${REQUIREMENTS}

RUN apt-get update && apt-get install -y --no-install-recommends \
        apt-utils \
        build-essential \
        curl \
        git \
        golang \
        libcurl3-dev \
        libfreetype6-dev \
        libpng12-dev \
        libzmq3-dev \
        openjdk-8-jdk \
        openjdk-8-jre-headless \
        pkg-config \
        python-dev \
        python-pip \
        python3-pip \
        python3-setuptools \
        rsync \
        software-properties-common \
        unzip \
        wget \
        zip \
        zlib1g-dev \
    && \
    apt-get clean && apt-get update \
    && \
    rm -rf /var/lib/apt/lists/*

RUN pip --no-cache-dir install --upgrade \
        pip setuptools \
    && \
    pip3 --no-cache-dir install --upgrade \
         pip setuptools

RUN python3 -m pip install tensorflow-gpu \
   && \
   pip3 --no-cache-dir install \
         ipykernel \
         jupyter \
         matplotlib \
         numpy \
         scipy \
         sklearn \
         pandas \
         xarray \
         wheel \
         torch torchvision \
    && \
    pip3 --no-cache-dir install \
         --requirement ${REQUIREMENTS} \
    && \
    python3 -m ipykernel.kernelspec \
    && \
    jupyter notebook --generate-config --allow-root \
    && \
    ipython profile create \
    && \
    cat ${JUP_CONFIG_PARTIAL} >> /root/.jupyter/jupyter_notebook_config.py \
    && \
    rm ${JUP_CONFIG_PARTIAL} ${REQUIREMENTS}

# tensorboard
EXPOSE 6006
# IPython
EXPOSE 8888
⁠jupyter_notebook_config.partial.py
import os
# from IPython.lib import passwd

# c = c  # pylint:disable=undefined-variable

c.NotebookApp.ip = '0.0.0.0'
c.NotebookApp.open_browser = False
c.NotebookApp.port = int(os.getenv('PORT', 8888))
⁠requirements.txt
# add additional pipy-packages here


Tag summary

Content type

Image

Digest

Size

2.3 GB

Last updated

about 8 years ago

docker pull drstoop/data_science:tf-1.4.0_torch_gpu_jupyter