Dockerfile

# Use CUDA base for GPU support
FROM nvidia/cuda:13.0.1-runtime-ubuntu22.04

# Set timezone non-interactively
ENV DEBIAN_FRONTEND=noninteractive
ENV TZ=UTC

# Install Python and basic dependencies
RUN apt-get update && apt-get install -y \
    software-properties-common \
    && add-apt-repository ppa:deadsnakes/ppa \
    && apt-get update && apt-get install -y \
    python3.11 \
    python3.11-dev \
    python3.11-venv \
    python3.11-distutils \
    git \
    libsndfile1 \
    ffmpeg \
    curl \
    && rm -rf /var/lib/apt/lists/* \
    && ln -sf /usr/bin/python3.11 /usr/bin/python3 \
    && ln -sf /usr/bin/python3.11 /usr/bin/python \
    && curl -sS https://bootstrap.pypa.io/get-pip.py | python3.11

WORKDIR /app

# Copy and install Python dependencies
COPY pyproject.toml poetry.lock* ./
RUN python3.11 -m pip install poetry && \
    poetry config virtualenvs.create false && \
    poetry install --only=main

# Copy application code
COPY models/ ./models/
COPY config/ ./config/
COPY app/ ./app/
COPY scripts/ ./scripts/
COPY src/ ./src/

# Create cache directories with proper permissions
RUN mkdir -p /app/.cache/huggingface /app/.cache/torch /tmp/numba_cache && \
    chmod -R 777 /app/.cache /tmp/numba_cache

# Set environment
ENV PYTHONPATH="/app"
ENV HF_HOME="/app/.cache/huggingface"
ENV TRANSFORMERS_CACHE="/app/.cache/huggingface"
ENV TORCH_HOME="/app/.cache/torch"
ENV NUMBA_CACHE_DIR="/tmp/numba_cache"
ENV NUMBA_DISABLE_JIT=0
ENV MUSICLIME_NUM_SAMPLES=1000
ENV MUSICLIME_NUM_FEATURES=10
ENV PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True

# Hugging Face Spaces specific, expose port 7860
EXPOSE 7860

# Run on port 7860 for HF Spaces
CMD ["uvicorn", "app.server:app", "--host", "0.0.0.0", "--port", "7860", "--timeout-keep-alive", "600"]