Dockerfile.ci

# syntax=docker/dockerfile:experimental

ARG FROM_IMAGE_NAME
FROM $FROM_IMAGE_NAME as main
ENV DEBIAN_FRONTEND=noninteractive

RUN sed -i -e 's/^APT/# APT/' -e 's/^DPkg/# DPkg/' \
      /etc/apt/apt.conf.d/docker-clean

RUN apt-get update && \
      apt-get install -y --no-install-recommends gettext && \
      apt-get clean

RUN wget https://github.com/mikefarah/yq/releases/download/v4.44.1/yq_linux_amd64 -O /usr/local/bin/yq && \
chmod a+x /usr/local/bin/yq

##### For Mamba begin #####
RUN pip uninstall -y triton && \
    pip install triton==2.1.0

# The causal-conv1d and mamba-ssm packages below are built from scratch here
# (which takes significant time) because there are no wheels available on PyPI
# for these relatively newer versions of the packages that are compatible with
# the older NGC-variant PyTorch version (e.g. version 2.2.0.dev231106) that we
# are using (in the NGC base container). Generally, if the package is not
# compatible with the PyTorch version, then it will generate a Python import
# error. The package authors tend to only release wheels for new versions of
# these pacakges which are compatible with the versions of regular PyTorch and
# NGC-variant PyTorch that are newer at the time of release. So, to use newer
# versions of these packages with relatively older versions of the NGC PyTorch
# container, we tend to have to build the packages from scratch.

RUN cd /tmp && \
    pip uninstall -y causal-conv1d && \
    git clone https://github.com/Dao-AILab/causal-conv1d.git && \
    cd causal-conv1d && \
    git checkout v1.2.2.post1 && \
    CAUSAL_CONV1D_FORCE_BUILD=TRUE pip install . && \
    cd .. && \
    rm -rf causal-conv1d

RUN cd /tmp && \
    pip uninstall -y mamba-ssm && \
    git clone https://github.com/state-spaces/mamba.git && \
    cd mamba && \
    git checkout v2.0.3 && \
    MAMBA_FORCE_BUILD=TRUE pip install . && \
    cd .. && \
    rm -rf mamba
##### For Mamba end #####

##### For JET-API start #####
RUN apt-get install -y python3-venv && \
    apt-get clean -y && \
    python -m venv /opt/jet
##### For JET-API end #####

RUN pip3 install --no-cache-dir \
      einops \
      flask-restful \
      nltk \
      pytest \
      pytest-cov \
      pytest_mock \
      pytest-random-order \
      sentencepiece \
      wrapt \
      git+https://github.com/fanshiqing/grouped_gemm@v1.1.2 \
      zarr \
      tensorstore==0.1.45 \
      wandb

COPY . /workspace/megatron-lm

COPY . /workspace/megatron-lm
RUN cp -r /workspace/megatron-lm /opt && \
    pip install /opt/megatron-lm  


##### For NVIDIANS only #####
FROM main as jet
ARG CACHEBUST=0
RUN --mount=type=secret,id=JET_INDEX_URLS \
    JET_INDEX_URLS=$(cat /run/secrets/JET_INDEX_URLS) && \
    /opt/jet/bin/pip install jet-api --upgrade $JET_INDEX_URLS 
ENV PATH="$PATH:/opt/jet/bin"
###