Spaces:
Runtime error
Runtime error
Update Dockerfile
Browse files- Dockerfile +5 -3
Dockerfile
CHANGED
|
@@ -3,6 +3,9 @@ FROM nvidia/cuda:${CUDA_IMAGE}
|
|
| 3 |
|
| 4 |
# We need to set the host to 0.0.0.0 to allow outside access
|
| 5 |
ENV HOST 0.0.0.0
|
|
|
|
|
|
|
|
|
|
| 6 |
|
| 7 |
RUN apt-get update && apt-get upgrade -y \
|
| 8 |
&& apt-get install -y git build-essential \
|
|
@@ -11,8 +14,6 @@ RUN apt-get update && apt-get upgrade -y \
|
|
| 11 |
libclblast-dev libopenblas-dev \
|
| 12 |
&& mkdir -p /etc/OpenCL/vendors && echo "libnvidia-opencl.so.1" > /etc/OpenCL/vendors/nvidia.icd
|
| 13 |
|
| 14 |
-
COPY . .
|
| 15 |
-
|
| 16 |
# setting build related env vars
|
| 17 |
ENV CUDA_DOCKER_ARCH=all
|
| 18 |
ENV LLAMA_CUBLAS=1
|
|
@@ -21,7 +22,8 @@ ENV LLAMA_CUBLAS=1
|
|
| 21 |
RUN python3 -m pip install --no-cache-dir --upgrade pip pytest cmake scikit-build setuptools fastapi uvicorn sse-starlette pydantic-settings starlette-context huggingface-hub==0.14.1 flask
|
| 22 |
|
| 23 |
# Install llama-cpp-python (build with cuda)
|
| 24 |
-
RUN CMAKE_ARGS="-DLLAMA_CUBLAS=on" pip install llama-cpp-python
|
|
|
|
| 25 |
|
| 26 |
# Run the server
|
| 27 |
CMD python3 -m app.py
|
|
|
|
| 3 |
|
| 4 |
# We need to set the host to 0.0.0.0 to allow outside access
|
| 5 |
ENV HOST 0.0.0.0
|
| 6 |
+
RUN useradd -m -u 1000 user
|
| 7 |
+
WORKDIR /home/user/app
|
| 8 |
+
COPY --link --chown=1000 ./ /home/user/app
|
| 9 |
|
| 10 |
RUN apt-get update && apt-get upgrade -y \
|
| 11 |
&& apt-get install -y git build-essential \
|
|
|
|
| 14 |
libclblast-dev libopenblas-dev \
|
| 15 |
&& mkdir -p /etc/OpenCL/vendors && echo "libnvidia-opencl.so.1" > /etc/OpenCL/vendors/nvidia.icd
|
| 16 |
|
|
|
|
|
|
|
| 17 |
# setting build related env vars
|
| 18 |
ENV CUDA_DOCKER_ARCH=all
|
| 19 |
ENV LLAMA_CUBLAS=1
|
|
|
|
| 22 |
RUN python3 -m pip install --no-cache-dir --upgrade pip pytest cmake scikit-build setuptools fastapi uvicorn sse-starlette pydantic-settings starlette-context huggingface-hub==0.14.1 flask
|
| 23 |
|
| 24 |
# Install llama-cpp-python (build with cuda)
|
| 25 |
+
RUN CMAKE_ARGS="-DLLAMA_CUBLAS=on" pip install --no-cache-dir llama-cpp-python
|
| 26 |
+
|
| 27 |
|
| 28 |
# Run the server
|
| 29 |
CMD python3 -m app.py
|