FROM ubuntu:20.04 AS base_image

ENV DEBIAN_FRONTEND=noninteractive \
    LD_LIBRARY_PATH="${LD_LIBRARY_PATH}:/usr/local/lib"

RUN apt-get update \
 && apt-get upgrade -y \
 && apt-get autoremove -y \
 && apt-get clean \
 && rm -rf /var/lib/apt/lists/*

FROM base_image AS common

LABEL maintainer="Amazon AI"
LABEL dlc_major_version="1"

# TensorFlow major.minor version
ENV TF_VERSION=2.12

# prevent stopping by user interaction
ENV DEBIAN_FRONTEND noninteractive
ENV DEBCONF_NONINTERACTIVE_SEEN true

# Set environment variables for MKL
# For more about MKL with TensorFlow see:
# https://www.tensorflow.org/performance/performance_guide#tensorflow_with_intel%C2%AE_mkl_dnn

ENV KMP_AFFINITY=granularity=fine,compact,1,0
ENV KMP_BLOCKTIME=1
ENV KMP_SETTINGS=0

ENV PYTHONDONTWRITEBYTECODE=1
ENV PYTHONUNBUFFERED=1
ENV PYTHONIOENCODING=UTF-8
ENV LANG=C.UTF-8
ENV LC_ALL=C.UTF-8

ARG PYTHON=python3.10
ARG PYTHON_VERSION=3.10.10

ARG PIP=pip3

ARG OMPI_VERSION=4.1.5

# To be passed to ec2 and sagemaker stages
ENV PYTHON=${PYTHON}
ENV PYTHON_VERSION=${PYTHON_VERSION}
ENV PIP=${PIP}

RUN apt-get update && apt-get install -y --no-install-recommends \
    build-essential \
    openssh-client \
    openssh-server \
    ca-certificates \
    curl \
    emacs \
    git \
    libtemplate-perl \
    libssl1.1 \
    openssl \
    unzip \
    wget \
    vim \
    zlib1g-dev \
    # Install dependent library for OpenCV
    libgtk2.0-dev \
 && rm -rf /var/lib/apt/lists/*

# Install Open MPI
RUN mkdir /tmp/openmpi \
 && cd /tmp/openmpi \
 && wget --quiet https://download.open-mpi.org/release/open-mpi/v4.1/openmpi-${OMPI_VERSION}.tar.gz \
 && tar zxf openmpi-${OMPI_VERSION}.tar.gz \
 && cd openmpi-${OMPI_VERSION} \
 && ./configure --enable-orterun-prefix-by-default \
 && make -j $(nproc) all \
 && make install \
 && ldconfig \
 && rm -rf /tmp/openmpi

# Create a wrapper for OpenMPI to allow running as root by default
RUN mv /usr/local/bin/mpirun /usr/local/bin/mpirun.real \
 && echo '#!/bin/bash' > /usr/local/bin/mpirun \
 && echo 'mpirun.real --allow-run-as-root "$@"' >> /usr/local/bin/mpirun \
 && chmod a+x /usr/local/bin/mpirun

RUN echo "hwloc_base_binding_policy = none" >> /usr/local/etc/openmpi-mca-params.conf \
 && echo "rmaps_base_mapping_policy = slot" >> /usr/local/etc/openmpi-mca-params.conf

ENV LD_LIBRARY_PATH=/usr/local/openmpi/lib:$LD_LIBRARY_PATH
ENV PATH /usr/local/openmpi/bin/:$PATH

# SSH login fix. Otherwise user is kicked off after login
RUN sed 's@session\s*required\s*pam_loginuid.so@session optional pam_loginuid.so@g' -i /etc/pam.d/sshd

# Create SSH key.
RUN mkdir -p /root/.ssh/ \
 && mkdir -p /var/run/sshd \
 && ssh-keygen -q -t rsa -N '' -f /root/.ssh/id_rsa \
 && cp /root/.ssh/id_rsa.pub /root/.ssh/authorized_keys \
 && printf "Host *\n  StrictHostKeyChecking no\n" >> /root/.ssh/config

WORKDIR /

RUN apt-get update \
 && apt-get install -y --no-install-recommends \
    libbz2-dev \
    libc6-dev \
    libcurl4-openssl-dev \
    libffi-dev \
    libgdbm-dev \
    liblzma-dev \
    libncursesw5-dev \
    libreadline-gplv2-dev \
    libsqlite3-dev \
    libssl-dev \
    tk-dev \
    ffmpeg \
    libsm6 \
    libxext6 \
 && rm -rf /var/lib/apt/lists/* \
 && apt-get clean

RUN wget https://www.python.org/ftp/python/$PYTHON_VERSION/Python-$PYTHON_VERSION.tgz \
    && tar -xvf Python-$PYTHON_VERSION.tgz \
    && cd Python-$PYTHON_VERSION \
    && ./configure \
    && make -j $(nproc) \
    && make install \
    && rm -rf ../Python-$PYTHON_VERSION*

RUN ${PIP} --no-cache-dir install --upgrade \
    pip \
    setuptools

# Some TF tools expect a "python" binary
RUN ln -s $(which ${PYTHON}) /usr/local/bin/python \
    && ln -s $(which ${PIP}) /usr/bin/pip

RUN apt-get update && apt-get -y install protobuf-compiler

# install PyYAML==5.1.2 to avoid conflict with latest awscli
RUN ${PIP} install --no-cache-dir -U \
    pybind11 \
    cmake \
    scipy \
    Pillow \
    python-dateutil \
    "pyYAML>=5.4,<5.5" \
    requests \
    "awscli<2" \
    mpi4py \
    # Let's install TensorFlow separately in the end to avoid
    # the library version to be overwritten
 && ${PIP} install --no-cache-dir -U \
    h5py \
    absl-py \
    opencv-python \
    werkzeug \
    psutil \
    urllib3 \
    "protobuf<4"

ADD https://raw.githubusercontent.com/aws/deep-learning-containers/master/src/deep_learning_container.py /usr/local/bin/deep_learning_container.py

RUN chmod +x /usr/local/bin/deep_learning_container.py

RUN curl https://aws-dlc-licenses.s3.amazonaws.com/tensorflow-${TF_VERSION}/license.txt -o /license.txt

########################################################
#  _____ ____ ____    ___
# | ____/ ___|___ \  |_ _|_ __ ___   __ _  __ _  ___
# |  _|| |     __) |  | || '_ ` _ \ / _` |/ _` |/ _ \
# | |__| |___ / __/   | || | | | | | (_| | (_| |  __/
# |_____\____|_____| |___|_| |_| |_|\__,_|\__, |\___|
#                                         |___/
#  ____           _
# |  _ \ ___  ___(_)_ __   ___
# | |_) / _ \/ __| | '_ \ / _ \
# |  _ <  __/ (__| | |_) |  __/
# |_| \_\___|\___|_| .__/ \___|
#                  |_|
########################################################

FROM common AS ec2
ARG TF_URL=https://framework-binaries.s3.us-west-2.amazonaws.com/tensorflow/r2.12_aws/cpu/2023-03-23-00-31/tensorflow_cpu-2.12.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl

RUN ${PIP} install --no-cache-dir -U \
    ${TF_URL} \
    tensorflow-io \
    tensorflow-datasets

# Install Horovod with TensorFlow
RUN HOROVOD_WITH_MPI=1 \
   && HOROVOD_WITH_TENSORFLOW=1 HOROVOD_WITHOUT_MXNET=1 HOROVOD_WITHOUT_PYTORCH=1 \
   && ${PIP} install --no-cache-dir horovod==0.27.0

RUN HOME_DIR=/root \
   && curl -o ${HOME_DIR}/oss_compliance.zip https://aws-dlinfra-utilities.s3.amazonaws.com/oss_compliance.zip \
   && unzip ${HOME_DIR}/oss_compliance.zip -d ${HOME_DIR}/ \
   && cp ${HOME_DIR}/oss_compliance/test/testOSSCompliance /usr/local/bin/testOSSCompliance \
   && chmod +x /usr/local/bin/testOSSCompliance \
   && chmod +x ${HOME_DIR}/oss_compliance/generate_oss_compliance.sh \
   && ${HOME_DIR}/oss_compliance/generate_oss_compliance.sh ${HOME_DIR} ${PYTHON} \
   && rm -rf ${HOME_DIR}/oss_compliance*

# remove tmp files
RUN rm -rf /tmp/*

CMD ["/bin/bash"]

#################################################################
#  ____                   __  __       _
# / ___|  __ _  __ _  ___|  \/  | __ _| | _____ _ __
# \___ \ / _` |/ _` |/ _ \ |\/| |/ _` | |/ / _ \ '__|
#  ___) | (_| | (_| |  __/ |  | | (_| |   <  __/ |
# |____/ \__,_|\__, |\___|_|  |_|\__,_|_|\_\___|_|
#              |___/
#  ___                              ____           _
# |_ _|_ __ ___   __ _  __ _  ___  |  _ \ ___  ___(_)_ __   ___
#  | || '_ ` _ \ / _` |/ _` |/ _ \ | |_) / _ \/ __| | '_ \ / _ \
#  | || | | | | | (_| | (_| |  __/ |  _ <  __/ (__| | |_) |  __/
# |___|_| |_| |_|\__,_|\__, |\___| |_| \_\___|\___|_| .__/ \___|
#                      |___/                        |_|
#################################################################

FROM common AS sagemaker

LABEL maintainer="Amazon AI"
LABEL dlc_major_version="1"

ARG TF_URL=https://framework-binaries.s3.us-west-2.amazonaws.com/tensorflow/r2.12_aws/cpu/2023-03-23-00-31/tensorflow_cpu-2.12.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl

# sagemaker-specific environment variable
ENV SAGEMAKER_TRAINING_MODULE sagemaker_tensorflow_container.training:main

# dependencies for opencv
# these dependencies are not needed for gpu image
RUN apt-get update && apt-get install -y --no-install-recommends \
    libgtk2.0-dev \
    && apt-get install -y -qq libkrb5-dev \
    && apt-get install -y -qq libsasl2-dev libsasl2-modules \
    && apt-get install -y -qq krb5-user \
&& rm -rf /var/lib/apt/lists/*


# https://github.com/tensorflow/models/issues/9267
# tf-models does not respect existing installations of TF and always installs open source TF
RUN ${PIP} install --no-cache-dir -U \
    tf-models-official==2.12.0 \
    tensorflow-text==2.12.0

RUN ${PIP} uninstall -y tensorflow tensorflow-gpu \
    ; ${PIP} install --no-cache-dir -U \
    ${TF_URL} \
    tensorflow-io \
    tensorflow-datasets

# Install Horovod with TensorFlow
RUN HOROVOD_WITH_MPI=1 \
   && HOROVOD_WITH_TENSORFLOW=1 HOROVOD_WITHOUT_MXNET=1 HOROVOD_WITHOUT_PYTORCH=1 \
   && ${PIP} install --no-cache-dir horovod==0.27.0

RUN $PYTHON -m pip install --no-cache-dir -U \
    numba \
    bokeh \
    imageio \
    opencv-python \
    plotly \
    seaborn \
    shap

RUN $PYTHON -m pip install --no-cache-dir -U \
    "sagemaker<3" \
    sagemaker-experiments==0.* \
    sagemaker-tensorflow-training \
    sagemaker-training \
    "sagemaker-studio-analytics-extension<1" \
    "sparkmagic<1" \
    "sagemaker-studio-sparkmagic-lib<1" \
    smclarify

# install smdebug directly the specific branch
ARG SMDEBUG_TAG=1.0.33
RUN git clone https://github.com/awslabs/sagemaker-debugger.git \
    && cd sagemaker-debugger \
    && git checkout tags/$SMDEBUG_TAG -b $SMDEBUG_TAG \
    && $PYTHON setup.py install && cd .. && rm -rf sagemaker-debugger

# install sagemaker-tensorflow from source
ARG SAGEMAKER_TENSORFLOW_TAG=v1.18.0
RUN git clone -b tf-2 https://github.com/aws/sagemaker-tensorflow-extensions.git \
    && cd sagemaker-tensorflow-extensions \
    && git checkout tags/$SAGEMAKER_TENSORFLOW_TAG -b $SAGEMAKER_TENSORFLOW_TAG \
    && $PYTHON -m pip install . \
    && cd .. && rm -rf sagemaker-tensorflow-extensions

# Remove python kernel installed by sparkmagic
RUN /usr/local/bin/jupyter-kernelspec remove -f python3

# remove tmp files
RUN rm -rf /tmp/*

RUN HOME_DIR=/root \
   && curl -o ${HOME_DIR}/oss_compliance.zip https://aws-dlinfra-utilities.s3.amazonaws.com/oss_compliance.zip \
   && unzip ${HOME_DIR}/oss_compliance.zip -d ${HOME_DIR}/ \
   && cp ${HOME_DIR}/oss_compliance/test/testOSSCompliance /usr/local/bin/testOSSCompliance \
   && chmod +x /usr/local/bin/testOSSCompliance \
   && chmod +x ${HOME_DIR}/oss_compliance/generate_oss_compliance.sh \
   && ${HOME_DIR}/oss_compliance/generate_oss_compliance.sh ${HOME_DIR} ${PYTHON} \
   && rm -rf ${HOME_DIR}/oss_compliance*

CMD ["/bin/bash"]