Dockerfile

FROM python:3.12-slim

WORKDIR /app

RUN apt-get update && \
    apt-get install -y libgomp1 openjdk-17-jre-headless wget unzip libgoogle-perftools-dev libsparsehash-dev git cmake build-essential

# download and compile fast align
RUN git clone https://github.com/clab/fast_align.git fast_align_src && \
    cd fast_align_src && \
    mkdir build && \
    cd build && \
    cmake .. && \
    make

RUN cp fast_align_src/build/fast_align .
RUN cp fast_align_src/build/atools .
RUN rm -rf fast_align_src

# download okapi tikal
RUN wget https://okapiframework.org/binaries/main/1.47.0/okapi-apps_gtk2-linux-x86_64_1.47.0.zip
RUN unzip okapi-apps_gtk2-linux-x86_64_1.47.0.zip -d okapi-apps_gtk2-linux-x86_64_1.47.0

RUN pip install -r requirements.txt
RUN python -m spacy download xx_ent_wiki_sm

RUN useradd -m appuser
USER appuser

RUN chown -R appuser:appuser /app

COPY fast_align_config ./fast_align_config
COPY src ./src
COPY gradio_app.py .
COPY requirements.txt .
run mkdir /app/tmp

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