Dockerfile.gpu

# Dockerfile.gpu — GPU build for Hugging Face Pro GPU Space (or any CUDA host).
# Base: CUDA 12.4 runtime + cuDNN 9 (what onnxruntime-gpu 1.19 needs).
# To use on HF: switch the Space to GPU hardware and set this as the Dockerfile.

# ── Stage 1: build the React frontend ────────────────────────────────────────
FROM node:20-slim AS frontend-builder
WORKDIR /app
COPY frontend/package*.json frontend/
RUN cd frontend && npm ci --silent
COPY frontend/ frontend/
RUN mkdir -p static/react && cd frontend && npm run build


# ── Stage 2: CUDA 12 + cuDNN 9 Python runtime ────────────────────────────────
FROM nvidia/cuda:12.4.1-cudnn-runtime-ubuntu22.04

ENV DEBIAN_FRONTEND=noninteractive
RUN apt-get update && apt-get install -y --no-install-recommends \
        python3.10 python3-pip python3.10-dev \
        build-essential cmake \
        libgl1 libglib2.0-0 libsm6 libxrender1 libxext6 \
        wget git \
    && ln -sf /usr/bin/python3.10 /usr/bin/python \
    && rm -rf /var/lib/apt/lists/*

WORKDIR /app

# Step 1: torch + torchvision CUDA 12.1 wheels (separate index)
RUN pip install --no-cache-dir \
    --extra-index-url https://download.pytorch.org/whl/cu121 \
    torch==2.3.0+cu121 torchvision==0.18.0+cu121

# Step 2: the rest (PyPI), including onnxruntime-gpu
COPY requirements-gpu.txt .
RUN pip install --no-cache-dir -r requirements-gpu.txt

# React build from stage 1
COPY --from=frontend-builder /app/static/react ./static/react

# App source
COPY . .
RUN mkdir -p models uploads/temp outputs/results

ENV PORT=7860
EXPOSE 7860

# Fix CRLF (Windows -> Linux) and make executable
RUN sed -i 's/\r$//' startup.sh && chmod +x startup.sh

CMD ["./startup.sh"]