huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-09-11 12:24:53 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fixing the docker environment hopefully.

Browse Source
This commit is contained in:
Nicolas Patry 2025-02-06 23:56:12 +01:00
parent 3a533bcd69
commit 04462263dc
No known key found for this signature in database
GPG Key ID: 4242CEF24CB6DBF9
1 changed files with 17 additions and 18 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

35
Dockerfile_intel
View File

@ -169,7 +169,7 @@ ARG MAMBA_VERSION=23.1.0-1
ARG PYTHON_VERSION='3.11.10' ARG PYTHON_VERSION='3.11.10'
# Automatically set by buildx # Automatically set by buildx
ARG TARGETPLATFORM ARG TARGETPLATFORM
ENV PATH /opt/conda/bin:$PATH ENV PATH="/opt/conda/bin:$PATH"
# TGI seem to require libssl.so.1.1 instead of libssl.so.3 so we can't use ubuntu 22.04. Ubuntu 20.04 has python==3.8, and TGI requires python>=3.9, hence the need for miniconda. # TGI seem to require libssl.so.1.1 instead of libssl.so.3 so we can't use ubuntu 22.04. Ubuntu 20.04 has python==3.8, and TGI requires python>=3.9, hence the need for miniconda.
# Install mamba # Install mamba
@ -192,6 +192,21 @@ RUN case ${TARGETPLATFORM} in \
RUN conda install -c conda-forge gperftools mkl RUN conda install -c conda-forge gperftools mkl
# Install server
COPY proto proto
COPY server/pyproject.toml server/pyproject.toml
COPY server/Makefile server/Makefile
RUN cd server && \
pip install -U pip uv && \
uv sync --frozen --extra gen --extra accelerate --extra compressed-tensors --extra quantize --extra peft --extra outlines --no-install-project && \
. ./.venv/bin/activate && \
make gen-server-raw
RUN cd server && \
uv sync --frozen --extra gen --extra accelerate --extra compressed-tensors --extra quantize --extra peft --extra outlines && \
. ./.venv/bin/activate && \
pwd && \
text-generation-server --help
RUN . ./server/.venv/bin/activate && uv pip install https://download.pytorch.org/whl/nightly/cpu/torch-2.5.0.dev20240815%2Bcpu-cp311-cp311-linux_x86_64.whl RUN . ./server/.venv/bin/activate && uv pip install https://download.pytorch.org/whl/nightly/cpu/torch-2.5.0.dev20240815%2Bcpu-cp311-cp311-linux_x86_64.whl
RUN . ./server/.venv/bin/activate && uv pip install https://download.pytorch.org/whl/nightly/cpu/torchvision-0.20.0.dev20240815%2Bcpu-cp311-cp311-linux_x86_64.whl RUN . ./server/.venv/bin/activate && uv pip install https://download.pytorch.org/whl/nightly/cpu/torchvision-0.20.0.dev20240815%2Bcpu-cp311-cp311-linux_x86_64.whl
@ -207,7 +222,7 @@ RUN git clone https://github.com/intel/torch-ccl.git && cd torch-ccl && git chec
RUN sed -i 's/VERSION_MINOR 6/VERSION_MINOR 5/' intel-extension-for-pytorch/version.txt RUN sed -i 's/VERSION_MINOR 6/VERSION_MINOR 5/' intel-extension-for-pytorch/version.txt
RUN . ./server/.venv/bin/activate && cd intel-extension-for-pytorch && git submodule sync && git submodule update --init --recursive && python setup.py install RUN . ./server/.venv/bin/activate && cd intel-extension-for-pytorch && git submodule sync && git submodule update --init --recursive && python setup.py install
RUN . ./server/.venv/bin/activate && cd torch-ccl && git submodule sync && git submodule update --init --recursive && pip install . RUN . ./server/.venv/bin/activate && cd torch-ccl && git submodule sync && git submodule update --init --recursive && uv pip install .
ENV LD_PRELOAD=/opt/conda/lib/libtcmalloc.so ENV LD_PRELOAD=/opt/conda/lib/libtcmalloc.so
ENV CCL_ROOT=/opt/conda/lib/python3.11/site-packages/oneccl_bindings_for_pytorch ENV CCL_ROOT=/opt/conda/lib/python3.11/site-packages/oneccl_bindings_for_pytorch
@ -216,22 +231,6 @@ ENV FI_PROVIDER_PATH=/opt/conda/lib/python3.11/site-packages/oneccl_bindings_for
ENV LD_LIBRARY_PATH=/opt/conda/lib/python3.11/site-packages/oneccl_bindings_for_pytorch/opt/mpi/libfabric/lib:/opt/conda/lib/python3.11/site-packages/oneccl_bindings_for_pytorch/lib ENV LD_LIBRARY_PATH=/opt/conda/lib/python3.11/site-packages/oneccl_bindings_for_pytorch/opt/mpi/libfabric/lib:/opt/conda/lib/python3.11/site-packages/oneccl_bindings_for_pytorch/lib
ENV LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/opt/conda/lib/" ENV LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/opt/conda/lib/"
# Install server
COPY proto proto
COPY server server
COPY server/Makefile server/Makefile
ENV UV_SYSTEM_PYTHON=1
RUN cd server && \
pip install -U pip uv && \
uv sync --frozen --extra gen --extra accelerate --extra compressed-tensors --extra quantize --extra peft --extra outlines --no-install-project && \
. ./.venv/bin/activate && \
make gen-server-raw
RUN cd server && \
uv sync --frozen --extra gen --extra accelerate --extra compressed-tensors --extra quantize --extra peft --extra outlines && \
. ./.venv/bin/activate && \
pwd && \
text-generation-server --help
# Install benchmarker # Install benchmarker
COPY --from=builder /usr/src/target/release-opt/text-generation-benchmark /usr/local/bin/text-generation-benchmark COPY --from=builder /usr/src/target/release-opt/text-generation-benchmark /usr/local/bin/text-generation-benchmark