Dockerfile

# ── Build stage ───────────────────────────────────────────────────────────────
FROM python:3.11-slim

WORKDIR /app

# System libraries required by OpenCV + PaddlePaddle + transformers
RUN apt-get update && apt-get install -y --no-install-recommends \
        libglib2.0-0 \
        libgl1 \
        libgomp1 \
        libgcc-s1 \
        wget \
    && rm -rf /var/lib/apt/lists/*

# ── Install CPU-only PyTorch first (avoids pulling 2 GB CUDA wheels) ──────────
RUN pip install --no-cache-dir \
        torch==2.3.0 \
        torchvision==0.18.0 \
        --index-url https://download.pytorch.org/whl/cpu

# ── Install remaining Python dependencies ─────────────────────────────────────
COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt

# ── Copy source code and models into the container ────────────────────────────
COPY . .

# Tell PaddleOCR 3.x (paddlex backend) where to find offline model weights
ENV PADDLE_PDX_MODEL_HOME=/app/models/paddleocr

# HuggingFace Spaces must expose port 7860
EXPOSE 7860

CMD ["python", "app.py"]