| # 1. Base Image: Use a slim Python image | |
| FROM python:3.10-slim | |
| # 2. Set Environment Variables | |
| # General Python/PIP settings | |
| ENV PYTHONUNBUFFERED=1 | |
| ENV PIP_NO_CACHE_DIR=off | |
| ENV PIP_DISABLE_PIP_VERSION_CHECK=on | |
| # Hugging Face cache settings | |
| ENV HF_HOME="/app/huggingface_cache" | |
| ENV TRANSFORMERS_CACHE="/app/huggingface_cache/transformers" | |
| # Optional: For Hugging Face token (if needed for private models, Gemma is public) | |
| # ENV HUGGING_FACE_HUB_TOKEN="your_hf_token_here" # Pass at runtime or via secrets | |
| # 3. Set Working Directory | |
| WORKDIR /app | |
| # 4. Copy requirements file and install dependencies | |
| # This is done before copying the rest of the app to leverage Docker layer caching. | |
| COPY requirements.txt . | |
| RUN apt-get update && apt-get install -y --no-install-recommends \ | |
| build-essential \ | |
| && rm -rf /var/lib/apt/lists/* | |
| RUN pip install --no-cache-dir -r requirements.txt | |
| # 5. Copy the rest of the application code | |
| COPY app.py . | |
| # 6. Create the cache directory and set permissions | |
| RUN mkdir -p $HF_HOME && chmod -R 777 $HF_HOME | |
| # Note: chmod 777 is permissive; for production, consider a more specific user/group. | |
| # 7. Expose the port the app runs on | |
| EXPOSE 8000 | |
| # 8. Command to run the application | |
| CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "8000"] |