mirror of
https://github.com/ggerganov/llama.cpp.git
synced 2024-10-31 23:28:51 +01:00
04ac0607e9
* python: add check-requirements.sh and GitHub workflow This script and workflow forces package versions to remain compatible across all convert*.py scripts, while allowing secondary convert scripts to import dependencies not wanted in convert.py. * Move requirements into ./requirements * Fail on "==" being used for package requirements (but can be suppressed) * Enforce "compatible release" syntax instead of == * Update workflow * Add upper version bound for transformers and protobuf * improve check-requirements.sh * small syntax change * don't remove venvs if nocleanup is passed * See if this fixes docker workflow * Move check-requirements.sh into ./scripts/ --------- Co-authored-by: Jared Van Bortel <jared@nomic.ai>
46 lines
1008 B
Docker
46 lines
1008 B
Docker
ARG UBUNTU_VERSION=22.04
|
|
|
|
# This needs to generally match the container host's environment.
|
|
ARG ROCM_VERSION=5.6
|
|
|
|
# Target the CUDA build image
|
|
ARG BASE_ROCM_DEV_CONTAINER=rocm/dev-ubuntu-${UBUNTU_VERSION}:${ROCM_VERSION}-complete
|
|
|
|
FROM ${BASE_ROCM_DEV_CONTAINER} as build
|
|
|
|
# Unless otherwise specified, we make a fat build.
|
|
# List from https://github.com/ggerganov/llama.cpp/pull/1087#issuecomment-1682807878
|
|
# This is mostly tied to rocBLAS supported archs.
|
|
ARG ROCM_DOCKER_ARCH=\
|
|
gfx803 \
|
|
gfx900 \
|
|
gfx906 \
|
|
gfx908 \
|
|
gfx90a \
|
|
gfx1010 \
|
|
gfx1030 \
|
|
gfx1100 \
|
|
gfx1101 \
|
|
gfx1102
|
|
|
|
COPY requirements.txt requirements.txt
|
|
COPY requirements requirements
|
|
|
|
RUN pip install --upgrade pip setuptools wheel \
|
|
&& pip install -r requirements.txt
|
|
|
|
WORKDIR /app
|
|
|
|
COPY . .
|
|
|
|
# Set nvcc architecture
|
|
ENV GPU_TARGETS=${ROCM_DOCKER_ARCH}
|
|
# Enable ROCm
|
|
ENV LLAMA_HIPBLAS=1
|
|
ENV CC=/opt/rocm/llvm/bin/clang
|
|
ENV CXX=/opt/rocm/llvm/bin/clang++
|
|
|
|
RUN make
|
|
|
|
ENTRYPOINT [ "/app/main" ]
|