9th

Dockerfile

@@ -1,17 +1,9 @@
-# Use an NVIDIA CUDA base image
-ARG CUDA_IMAGE="12.3.1-base-ubuntu22.04"
+ARG CUDA_IMAGE="12.1.1-devel-ubuntu22.04"
 FROM nvidia/cuda:${CUDA_IMAGE}
 
+# We need to set the host to 0.0.0.0 to allow outside access
 ENV HOST 0.0.0.0
 
-# Set the working directory in the container to /app
-#WORKDIR /app
-
-RUN mkdir -p /app/cache && chmod -R 777 /app/cache
-
-ENV HF_HOME=/app/cache
-
-# Install Python and pip
 RUN apt-get update && apt-get upgrade -y \
     && apt-get install -y git build-essential \
     python3 python3-pip gcc wget \
@@ -19,28 +11,23 @@ RUN apt-get update && apt-get upgrade -y \
     libclblast-dev libopenblas-dev \
     && mkdir -p /etc/OpenCL/vendors && echo "libnvidia-opencl.so.1" > /etc/OpenCL/vendors/nvidia.icd
 
+COPY . .
+
+# setting build related env vars
 ENV CUDA_DOCKER_ARCH=all
 ENV LLAMA_CUBLAS=1
 
-# Copy the current directory contents into the container at /app
-COPY . /app
+# Install depencencies
+RUN python3 -m pip install --upgrade pip pytest cmake \
+    scikit-build setuptools fastapi uvicorn sse-starlette \
+    pydantic-settings starlette-context gradio huggingface_hub hf_transfer
 
-# Install required packages from requirements.txt
-COPY ./requirements.txt /app/requirements.txt
-RUN pip3 install --no-cache-dir -r /app/requirements.txt
-
-# Expose the ports for FastAPI and Streamlit
-EXPOSE 8000
-EXPOSE 8501
-
-# Copy and give execute permissions to the start script
-COPY start_server.sh /app/start_server.sh
-RUN chmod +x /app/start_server.sh
+# Install llama-cpp-python (build with cuda)
+RUN CMAKE_ARGS="-DLLAMA_CUBLAS=on" pip install llama-cpp-python
 
 RUN useradd -m -u 1000 user
 # Switch to the "user" user
 USER user
-WORKDIR /home/user/app
 # Set home to the user's home directory
 ENV HOME=/home/user \
 	PATH=/home/user/.local/bin:$PATH \
@@ -52,9 +39,9 @@ ENV HOME=/home/user \
 	GRADIO_THEME=huggingface \
 	SYSTEM=spaces
 
+WORKDIR $HOME/app
+
 # Copy the current directory contents into the container at $HOME/app setting the owner to the user
-COPY --chown=user . /home/user/app
+COPY --chown=user . $HOME/app
 
-# Run the start script
-#CMD ["/app/start_server.sh"]
-CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "8000"]
+CMD ["python3", "app.py"]