huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-09-11 04:14:52 +00:00
Code Issues Packages Projects Releases Wiki Activity

Non system uv.

Browse Source
This commit is contained in:
Nicolas Patry 2025-02-06 23:47:12 +01:00
parent 4ba53eeec3
commit 3a533bcd69
No known key found for this signature in database
GPG Key ID: 3942D952A9BD2F09
1 changed files with 12 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

25
Dockerfile_intel
View File

@ -53,7 +53,7 @@ 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
@ -102,7 +102,6 @@ WORKDIR /usr/src
COPY proto proto COPY proto proto
COPY server server COPY server server
COPY server/Makefile server/Makefile COPY server/Makefile server/Makefile
ENV UV_SYSTEM_PYTHON=1
RUN cd server && \ RUN cd server && \
pip install -U pip uv && \ pip install -U pip uv && \
uv sync --frozen --extra gen --extra accelerate --extra compressed-tensors --extra quantize --extra peft --extra outlines --no-install-project && \ uv sync --frozen --extra gen --extra accelerate --extra compressed-tensors --extra quantize --extra peft --extra outlines --no-install-project && \
@ -115,12 +114,12 @@ RUN cd server && \
pwd && \ pwd && \
text-generation-server --help text-generation-server --help
RUN . ./server/.venv/bin/activate && pip install https://intel-optimized-pytorch.s3.cn-north-1.amazonaws.com.cn/ipex_dev/xpu/torch-2.5.0a0%2Bgite84e33f-cp311-cp311-linux_x86_64.whl --no-cache-dir RUN . ./server/.venv/bin/activate && uv pip install https://intel-optimized-pytorch.s3.cn-north-1.amazonaws.com.cn/ipex_dev/xpu/torch-2.5.0a0%2Bgite84e33f-cp311-cp311-linux_x86_64.whl --no-cache-dir
RUN . ./server/.venv/bin/activate && pip install https://intel-optimized-pytorch.s3.cn-north-1.amazonaws.com.cn/ipex_dev/xpu/torchaudio-2.5.0a0%2B56bc006-cp311-cp311-linux_x86_64.whl --no-cache-dir RUN . ./server/.venv/bin/activate && uv pip install https://intel-optimized-pytorch.s3.cn-north-1.amazonaws.com.cn/ipex_dev/xpu/torchaudio-2.5.0a0%2B56bc006-cp311-cp311-linux_x86_64.whl --no-cache-dir
RUN . ./server/.venv/bin/activate && pip install https://intel-optimized-pytorch.s3.cn-north-1.amazonaws.com.cn/ipex_dev/xpu/torchvision-0.20.0a0%2B8e8a208-cp311-cp311-linux_x86_64.whl --no-cache-dir # RUN . ./server/.venv/bin/activate && uv pip install https://intel-optimized-pytorch.s3.cn-north-1.amazonaws.com.cn/ipex_dev/xpu/torchvision-0.20.0a0%2B8e8a208-cp311-cp311-linux_x86_64.whl --no-cache-dir
RUN . ./server/.venv/bin/activate && pip install https://intel-optimized-pytorch.s3.cn-north-1.amazonaws.com.cn/ipex_dev/xpu/oneccl_bind_pt-2.5.0%2Bxpu-cp311-cp311-linux_x86_64.whl --no-cache-dir # RUN . ./server/.venv/bin/activate && uv pip install https://intel-optimized-pytorch.s3.cn-north-1.amazonaws.com.cn/ipex_dev/xpu/oneccl_bind_pt-2.5.0%2Bxpu-cp311-cp311-linux_x86_64.whl --no-cache-dir
RUN . ./server/.venv/bin/activate pip install triton-xpu==3.0.0b2 --no-cache-dir RUN . ./server/.venv/bin/activate && uv pip install triton-xpu==3.0.0b2 --no-cache-dir
ENV LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/opt/intel/oneapi/pti/0.9/lib:/opt/conda/lib ENV LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/opt/intel/oneapi/pti/0.9/lib:/opt/conda/lib
ENV CCL_ZE_IPC_EXCHANGE=sockets ENV CCL_ZE_IPC_EXCHANGE=sockets
@ -194,11 +193,11 @@ RUN case ${TARGETPLATFORM} in \
RUN conda install -c conda-forge gperftools mkl RUN conda install -c conda-forge gperftools mkl
RUN 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 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
RUN pip install https://download.pytorch.org/whl/nightly/cpu/torchaudio-2.4.0.dev20240815%2Bcpu-cp311-cp311-linux_x86_64.whl RUN . ./server/.venv/bin/activate && uv pip install https://download.pytorch.org/whl/nightly/cpu/torchaudio-2.4.0.dev20240815%2Bcpu-cp311-cp311-linux_x86_64.whl
RUN pip install triton==3.1.0 py-libnuma RUN . ./server/.venv/bin/activate && uv pip install triton==3.1.0 py-libnuma
WORKDIR /usr/src WORKDIR /usr/src
@ -206,9 +205,9 @@ RUN git clone https://github.com/intel/intel-extension-for-pytorch && cd intel-e
RUN git clone https://github.com/intel/torch-ccl.git && cd torch-ccl && git checkout v2.4.0+cpu+rc0 RUN git clone https://github.com/intel/torch-ccl.git && cd torch-ccl && git checkout v2.4.0+cpu+rc0
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 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 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 && 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