Mirrors
/
hf_text-generation-inference
mirror of https://github.com/huggingface/text-generation-inference.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Use cuda devel instead

This commit is contained in:
OlivierDehaene 2023-10-11 10:40:06 +02:00
parent 57795685d1
commit 8a7771a33c
1 changed files with 12 additions and 19 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

31
Dockerfile
View File

@ -37,7 +37,7 @@ RUN cargo build --release
# Python builder # Python builder
# Adapted from: https://github.com/pytorch/pytorch/blob/master/Dockerfile # Adapted from: https://github.com/pytorch/pytorch/blob/master/Dockerfile
FROM debian:bullseye-slim as pytorch-install FROM nvidia/cuda:12.1.0-devel-ubuntu20.04 as pytorch-install
ARG PYTORCH_VERSION=2.1.0 ARG PYTORCH_VERSION=2.1.0
ARG PYTHON_VERSION=3.9 ARG PYTHON_VERSION=3.9
@ -49,9 +49,12 @@ ARG INSTALL_CHANNEL=pytorch
# Automatically set by buildx # Automatically set by buildx
ARG TARGETPLATFORM ARG TARGETPLATFORM
ENV PATH /opt/conda/bin:$PATH # Conda env
ENV PATH=/opt/conda/bin:$PATH \
CONDA_PREFIX=/opt/conda
RUN apt-get update && DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends \ RUN apt-get update && DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends \
ninja-build \
build-essential \ build-essential \
ca-certificates \ ca-certificates \
ccache \ ccache \
@ -79,18 +82,8 @@ RUN case ${TARGETPLATFORM} in \
esac && \ esac && \
/opt/conda/bin/conda clean -ya /opt/conda/bin/conda clean -ya
# CUDA kernels builder image
FROM pytorch-install as kernel-builder
RUN apt-get update && DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends \
ninja-build \
&& rm -rf /var/lib/apt/lists/*
RUN /opt/conda/bin/conda install -c "nvidia/label/cuda-12.1.0" cuda==12.1 && \
/opt/conda/bin/conda clean -ya
# Build Flash Attention CUDA kernels # Build Flash Attention CUDA kernels
FROM kernel-builder as flash-att-builder FROM pytorch-install as flash-att-builder
WORKDIR /usr/src WORKDIR /usr/src
@ -100,7 +93,7 @@ COPY server/Makefile-flash-att Makefile
RUN make build-flash-attention RUN make build-flash-attention
# Build Flash Attention v2 CUDA kernels # Build Flash Attention v2 CUDA kernels
FROM kernel-builder as flash-att-v2-builder FROM pytorch-install as flash-att-v2-builder
WORKDIR /usr/src WORKDIR /usr/src
@ -110,35 +103,35 @@ COPY server/Makefile-flash-att-v2 Makefile
RUN make build-flash-attention-v2 RUN make build-flash-attention-v2
# Build Transformers exllama kernels # Build Transformers exllama kernels
FROM kernel-builder as exllama-kernels-builder FROM pytorch-install as exllama-kernels-builder
WORKDIR /usr/src WORKDIR /usr/src
COPY server/exllama_kernels/ . COPY server/exllama_kernels/ .
# Build specific version of transformers # Build specific version of transformers
RUN TORCH_CUDA_ARCH_LIST="8.0;8.6+PTX" python setup.py build RUN TORCH_CUDA_ARCH_LIST="8.0;8.6+PTX" python setup.py build
# Build Transformers awq kernels # Build Transformers awq kernels
FROM kernel-builder as awq-kernels-builder FROM pytorch-install as awq-kernels-builder
WORKDIR /usr/src WORKDIR /usr/src
COPY server/Makefile-awq Makefile COPY server/Makefile-awq Makefile
# Build specific version of transformers # Build specific version of transformers
RUN TORCH_CUDA_ARCH_LIST="8.0;8.6+PTX" make build-awq RUN TORCH_CUDA_ARCH_LIST="8.0;8.6+PTX" make build-awq
# Build eetq kernels # Build eetq kernels
FROM kernel-builder as eetq-kernels-builder FROM pytorch-install as eetq-kernels-builder
WORKDIR /usr/src WORKDIR /usr/src
COPY server/Makefile-eetq Makefile COPY server/Makefile-eetq Makefile
# Build specific version of transformers # Build specific version of transformers
RUN TORCH_CUDA_ARCH_LIST="8.0;8.6+PTX" make build-eetq RUN TORCH_CUDA_ARCH_LIST="8.0;8.6+PTX" make build-eetq
# Build Transformers CUDA kernels # Build Transformers CUDA kernels
FROM kernel-builder as custom-kernels-builder FROM pytorch-install as custom-kernels-builder
WORKDIR /usr/src WORKDIR /usr/src
COPY server/custom_kernels/ . COPY server/custom_kernels/ .
# Build specific version of transformers # Build specific version of transformers
RUN python setup.py build RUN python setup.py build
# Build vllm CUDA kernels # Build vllm CUDA kernels
FROM kernel-builder as vllm-builder FROM pytorch-install as vllm-builder
WORKDIR /usr/src WORKDIR /usr/src