bbrito/singularity_images:1.7

$ singularity pull shub://bbrito/singularity_images:1.7

Singularity Recipe

Bootstrap: docker
From: ubuntu:16.04

%labels
Maintainer Bruno Brito
Version v18.05.06

%help
This container runs Tensorflow-GPU.

%environment
    # nvidia driver libs specific cuda version libs are mounted by --bind command at run
    # required for GPU enabled container
    SHELL=/bin/bash
    CPATH="/cuda/include:$CPATH"
    PATH="/cuda/bin:/nvidia:$PATH"
    LD_LIBRARY_PATH="/cuda/lib64:/nvidia:$LD_LIBRARY_PATH"
    CUDA_HOME="/cuda"
    export PATH LD_LIBRARY_PATH CPATH CUDA_HOME

%post
    # default mount points
    mkdir -p /scratch/global /scratch/local /rcc/stor1/refdata /rcc/stor1/projects /rcc/stor1/depts
    touch /usr/bin/nvidia-smi

    # Install necessary packages
    apt-get update && apt-get install -y --no-install-recommends \
        build-essential \
        gcc-multilib \
        libatlas-base-dev \
        libboost-all-dev \
        libhdf5-serial-dev \
        libprotobuf-dev \
        protobuf-compiler \
        libopenblas-dev \
        liblapack-dev \
        gfortran \
        libcurl4-openssl-dev \
        python-pip \
        python3-pip \
        pkg-config \
        python-dev \
        python3-dev \
        python-setuptools \
        python3-setuptools
    apt-get clean

    # Update pip
    pip install --no-cache-dir --upgrade pip==9.0.3
    pip3 install --no-cache-dir --upgrade pip==9.0.3

    # Install TensorFlow-GPU
    export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-1.7.0-cp27-none-linux_x86_64.whl
    pip install --no-cache-dir --ignore-installed --upgrade $TF_BINARY_URL

    # Install python packages
    pip install --upgrade keras tflearn numpy nibabel h5py scikit-learn pandas scipy matplotlib ipykernel jupyter

    # Install TensorFlow-GPU
    export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-1.7.0-cp35-cp35m-linux_x86_64.whl
    pip3 install --no-cache-dir --ignore-installed --upgrade $TF_BINARY_URL

    # Install python packages
    pip3 install --upgrade keras tflearn numpy nibabel h5py scikit-learn pandas scipy matplotlib ipykernel jupyter

Collection


View on Datalad

Metrics

key value
id /containers/bbrito-singularity_images-1.7
collection name bbrito/singularity_images
branch master
tag 1.7
commit 29f8b186c173f234c471df2afb29bc98add96c21
version (container hash) bba0308ef60fee0653bcfa9e41e9142a
build date 2018-05-02T13:59:14.797Z
size (MB) 4633
size (bytes) 1351495711
SIF Download URL (please use pull with shub://)
Datalad URL View on Datalad
Singularity Recipe Singularity Recipe on Datalad
We cannot guarantee that all containers will still exist on GitHub.