Download Dockerfile from Devisri515/Agentic_RAG_Knowledge_Search: direct link, hf CLI and curl.
- Browser
- Download file 799 Bytes
-
https://huggingface.co/spaces/Devisri515/Agentic_RAG_Knowledge_Search/resolve/main/Dockerfile
- Command line
-
hf download hf://spaces/Devisri515/Agentic_RAG_Knowledge_Search/Dockerfile
-
curl -L -o Dockerfile https://huggingface.co/spaces/Devisri515/Agentic_RAG_Knowledge_Search/resolve/main/Dockerfile
799 Bytes
| # Docker image for HuggingFace Spaces (SDK: docker, app_port: 7860) | |
| FROM python:3.10-slim | |
| # HF Spaces run containers as a non-root user; create one with a writable home | |
| RUN useradd -m -u 1000 user | |
| USER user | |
| ENV HOME=/home/user \ | |
| PATH=/home/user/.local/bin:$PATH \ | |
| HF_HOME=/home/user/.cache/huggingface \ | |
| PYTHONUNBUFFERED=1 \ | |
| TOKENIZERS_PARALLELISM=false | |
| WORKDIR /app | |
| # Install Python dependencies | |
| COPY --chown=user:user requirements.txt . | |
| RUN pip install --no-cache-dir --user -r requirements.txt | |
| # Copy the application code | |
| COPY --chown=user:user . . | |
| # Bake the embedding + NLI models into the image (fast, offline cold starts) | |
| RUN python -m src.prefetch_models | |
| # Gradio UI (public) runs on 7860; FastAPI backend runs internally on 8000 | |
| EXPOSE 7860 | |
| CMD ["bash", "start.sh"] | |