ARG PYTHON=python3
ARG PYTHON_VERSION=3.10.8
ARG MAMBA_VERSION=22.11.1-4
ARG PYTORCH_VERSION_EC2=2.0.1
ARG PYTORCH_VERSION_SM=2.0.0

FROM ubuntu:20.04 AS base_image

ENV DEBIAN_FRONTEND=noninteractive \
    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"

ARG PYTHON
ARG PYTHON_VERSION
ARG OPEN_MPI_VERSION=4.1.5
ARG MAMBA_VERSION

# This arg required to stop docker build waiting for region configuration while installing tz data from ubuntu 20
ARG DEBIAN_FRONTEND=noninteractive

# Python won’t try to write .pyc or .pyo files on the import of source modules
# Force stdin, stdout and stderr to be totally unbuffered. Good for logging
ENV PYTHONDONTWRITEBYTECODE=1
ENV PYTHONUNBUFFERED=1
ENV PYTHONIOENCODING=UTF-8
ENV LANG=C.UTF-8
ENV LC_ALL=C.UTF-8
ENV LD_LIBRARY_PATH="/usr/local/lib:${LD_LIBRARY_PATH}"
ENV LD_LIBRARY_PATH="/opt/conda/lib:${LD_LIBRARY_PATH}"
ENV PATH=/opt/conda/bin:$PATH
ENV DGLBACKEND=pytorch
ENV DLC_CONTAINER_TYPE=training

WORKDIR /

RUN apt-get update \
 # TODO: Remove systemd upgrade once it is updated in base image
 && apt-get -y upgrade --only-upgrade systemd \
 && apt-get install -y --no-install-recommends \
    build-essential \
    ca-certificates \
    cmake \
    curl \
    emacs \
    git \
    jq \
    libcurl4-openssl-dev \
    libglib2.0-0 \
    libgl1-mesa-glx \
    libsm6 \
    libssl-dev \
    libxext6 \
    libxrender-dev \
    software-properties-common \
    unzip \
    vim \
    wget \
    zlib1g-dev \
 && rm -rf /var/lib/apt/lists/* \
 && apt-get clean

# Install Open MPI
RUN wget https://www.open-mpi.org/software/ompi/v4.1/downloads/openmpi-$OPEN_MPI_VERSION.tar.gz \
 && gunzip -c openmpi-$OPEN_MPI_VERSION.tar.gz | tar xf - \
 && cd openmpi-$OPEN_MPI_VERSION \
 && ./configure --prefix=/home/.openmpi \
 && make all install \
 && cd .. \
 && rm openmpi-$OPEN_MPI_VERSION.tar.gz \
 && rm -rf openmpi-$OPEN_MPI_VERSION


# Install OpenSSH for MPI to communicate between containers, allow OpenSSH to talk to containers without asking for confirmation
RUN apt-get update \
 && apt-get install -y --no-install-recommends openssh-client openssh-server \
 && mkdir -p /var/run/sshd \
 && cat /etc/ssh/ssh_config | grep -v StrictHostKeyChecking > /etc/ssh/ssh_config.new \
 && echo "    StrictHostKeyChecking no" >> /etc/ssh/ssh_config.new \
 && mv /etc/ssh/ssh_config.new /etc/ssh/ssh_config \
 && rm -rf /var/lib/apt/lists/* \
 && apt-get clean

# Configure OpenSSH so that nodes can communicate with each other
RUN mkdir -p /var/run/sshd \
 && sed 's@session\s*required\s*pam_loginuid.so@session optional pam_loginuid.so@g' -i /etc/pam.d/sshd

RUN rm -rf /root/.ssh/ \
 && mkdir -p /root/.ssh/ \
 && 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

# for conda ssl verification
ENV REQUESTS_CA_BUNDLE=/etc/ssl/certs/ca-certificates.crt
RUN curl -L -o ~/mambaforge.sh https://github.com/conda-forge/miniforge/releases/download/${MAMBA_VERSION}/Mambaforge-${MAMBA_VERSION}-Linux-x86_64.sh \
 && chmod +x ~/mambaforge.sh \
 && ~/mambaforge.sh -b -p /opt/conda \
 && rm ~/mambaforge.sh

# install core conda dependency packages (and pin essential packages)
RUN /opt/conda/bin/mamba install -y --override-channels \
    -c conda-forge \
    conda=23.1.0 \
    "cryptography>39.0.0" \
    cmake \
    requests \
    pyopenssl \
    python=$PYTHON_VERSION \
    cython \
    # force conda to install h5py with openmpi build, not mpich (rubik error)
    # https://anaconda.org/conda-forge/h5py/files?page=2
    "mpi4py>=3.1.4,<3.2" \
    mpi4py \
    h5py \
    mkl \
    mkl-include \
    parso \
    typing \
    # Below 2 are included in miniconda base, but not mamba so need to install
    conda-content-trust \
    charset-normalizer \
    packaging \
    opencv \
    awscli \
    boto3 \
    pybind11 \
    pyyaml \
    scipy \
    click \
    psutil \
    ipython \
    # Adding package for studio kernels
    ipykernel \
  && /opt/conda/bin/mamba install -y -c dglteam dgl \
  && /opt/conda/bin/mamba clean -afy

RUN curl -o /license.txt https://aws-dlc-licenses.s3.amazonaws.com/pytorch-2.0/license.txt

# The ENV variables declared below are changed in the previous section
# Grouping these ENV variables in the first section causes
# ompi_info to fail. This is only observed in CPU containers
ENV PATH="/home/.openmpi/bin:$PATH"
ENV LD_LIBRARY_PATH="/home/.openmpi/lib/:$LD_LIBRARY_PATH"
RUN ompi_info --parsable --all | grep mpi_built_with_cuda_support:value

# needed by torchdata
RUN mkdir -p /etc/pki/tls/certs && cp /etc/ssl/certs/ca-certificates.crt /etc/pki/tls/certs/ca-bundle.crt

# Removing the cache as it is needed for security verification
RUN rm -rf /root/.cache | true

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

FROM common AS ec2

ARG PYTHON

# PyTorch Binaries
ARG PT_EC2_TRAINING_URL
ARG PT_TORCHVISION_URL
ARG PT_TORCHAUDIO_URL
ARG PT_TORCHDATA_URL

# Install ec2 AWS-PyTorch
RUN /opt/conda/bin/mamba install -y pytorch=${PYTORCH_VERSION_EC2} \
  torchvision torchaudio torchtext cpuonly \
  --override-channels \
  -c https://aws-ml-conda-ec2.s3.us-west-2.amazonaws.com \
  -c conda-forge

# Patches
RUN pip install "pillow>=9.5"
RUN /opt/conda/bin/mamba install -y -c conda-forge \
  "requests>=2.31.0"

# install fastai (depends on PyTorch)
RUN /opt/conda/bin/mamba install -y -c conda-forge \
  # needed by fastai.distributed
  accelerate \
  && /opt/conda/bin/mamba install -y -c fastai fastai \
  # fastai conda strictly depends on dataclasses
  && pip uninstall -y dataclasses

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* \
 && rm -rf /tmp/tmp*

# Removing the cache as it is needed for security verification
RUN rm -rf /root/.cache | true

# Starts framework
CMD ["/bin/bash"]

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

FROM common AS sagemaker

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

ARG PYTHON

# The smdebug pipeline relies for following format to perform string replace and trigger DLC pipeline for validating
# the nightly builds. Therefore, while updating the smdebug version, please ensure that the format is not disturbed.
ARG SMDEBUG_VERSION=1.0.34

ENV SAGEMAKER_TRAINING_MODULE=sagemaker_pytorch_container.training:main

ARG PT_SM_TRAINING_URL
ARG PT_TORCHVISION_URL
ARG PT_TORCHAUDIO_URL
ARG PT_TORCHDATA_URL

# Install SM AWS-PyTorch (force remove will increase the docker image size drastically)
RUN /opt/conda/bin/mamba install -y pytorch=${PYTORCH_VERSION_SM} \
  torchvision torchaudio torchtext cpuonly \
  --override-channels \
  -c https://aws-ml-conda-pre-prod-ec2.s3.us-west-2.amazonaws.com \
  -c conda-forge

# install fastai (depends on PyTorch)
RUN /opt/conda/bin/mamba install -y -c conda-forge \
  # needed by fastai.distributed
  accelerate \
  && /opt/conda/bin/mamba install -y -c fastai fastai \
  # fastai conda strictly depends on dataclasses
  && pip uninstall -y dataclasses

WORKDIR /

# Copy workaround script for incorrect hostname
COPY changehostname.c /
COPY start_with_right_hostname.sh /usr/local/bin/start_with_right_hostname.sh

RUN chmod +x /usr/local/bin/start_with_right_hostname.sh

RUN pip install --no-cache-dir --upgrade pip --trusted-host pypi.org --trusted-host files.pythonhosted.org \
 && pip install --no-cache-dir -U \
    # disable smdebug pip install until available stable smdebug releases
    # smdebug==${SMDEBUG_VERSION} \
    smclarify \
    "sagemaker>=2,<3" \
    "sagemaker-experiments<1" \
    sagemaker-pytorch-training

# Install smdebug from source
RUN cd /tmp \
  && git clone https://github.com/awslabs/sagemaker-debugger --depth 1 --single-branch --branch ${SMDEBUG_VERSION} \
  && cd sagemaker-debugger \
  && pip install . \
  && rm -rf /tmp/*

# Install extra packages
RUN /opt/conda/bin/mamba install -y -c conda-forge \
  bokeh \
  imageio \
  numba \
  pandas \
  plotly \
  scikit-learn \
  seaborn \
  shap \
 && /opt/conda/bin/mamba clean -afy

# Patches
RUN pip install "pillow>=9.5"
RUN /opt/conda/bin/mamba install -y -c conda-forge \
  "requests>=2.31.0"

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* \
 && rm -rf /tmp/tmp*

# Removing the cache as it is needed for security verification
RUN rm -rf /root/.cache | true

ENTRYPOINT ["bash", "-m", "start_with_right_hostname.sh"]
CMD ["/bin/bash"]