Dockerfile

# # Use the official Python image
# FROM python:3.10-slim

# # Set environment variables
# ENV PYTHONDONTWRITEBYTECODE=1
# ENV PYTHONUNBUFFERED=1
# ENV TRANSFORMERS_CACHE=/data/huggingface/cache
# ENV HF_HOME=/app/.huggingface

# # Set the working directory
# WORKDIR /app

# # Install system dependencies
# RUN apt-get update && apt-get install -y \
#     git \
#     && rm -rf /var/lib/apt/lists/*

# # Create the Hugging Face cache directory and set proper permissions
# RUN mkdir -p /app/.huggingface && chmod -R 777 /app/.huggingface

# # Copy project files
# COPY . .

# # Install Python dependencies
# RUN pip install --no-cache-dir --upgrade pip \
#     && pip install --no-cache-dir -r requirements.txt

# # Expose the port FastAPI runs on
# EXPOSE 7860

# # Start the API using uvicorn
# CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "7860"]




# # Use the official Python image
# FROM python:3.10-slim

# # Set environment variables
# ENV PYTHONDONTWRITEBYTECODE=1
# ENV PYTHONUNBUFFERED=1
# ENV HF_HOME=/app/.huggingface

# # Set the working directory
# WORKDIR /app

# # Install system dependencies
# RUN apt-get update && apt-get install -y \
#     git \
#     && rm -rf /var/lib/apt/lists/*

# # Create necessary directories and set permissions
# RUN mkdir -p /app/.huggingface /app/pegasus_model \
#     && chmod -R 777 /app/.huggingface /app/pegasus_model


# # Create necessary directories and set permissions
# RUN mkdir -p /app/.huggingface /app/pegasus_model /app/nltk_data \
#     && chmod -R 777 /app/.huggingface /app/pegasus_model /app/nltk_data

# ENV NLTK_DATA=/app/nltk_data


# # Copy project files
# COPY . .

# # Install Python dependencies
# RUN pip install --no-cache-dir --upgrade pip \
#     && pip install --no-cache-dir -r requirements.txt \
#     && pip install --no-cache-dir huggingface_hub

# # Download the Pegasus model and tokenizer
# RUN python3 -c "from huggingface_hub import snapshot_download; snapshot_download(repo_id='tuner007/pegasus_paraphrase', local_dir='/app/pegasus_model')"

# # Expose the port FastAPI runs on
# EXPOSE 7860

# # Start the API using uvicorn
# CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "7860"]


##------------------------------------------------- add nltk------------------------------------------------


# # Use the official Python image
# FROM python:3.10-slim

# # Set environment variables
# ENV PYTHONDONTWRITEBYTECODE=1
# ENV PYTHONUNBUFFERED=1
# ENV HF_HOME=/app/.huggingface
# ENV NLTK_DATA=/app/nltk_data

# # Set the working directory
# WORKDIR /app

# # Install system dependencies
# RUN apt-get update && apt-get install -y \
#     git \
#     && rm -rf /var/lib/apt/lists/*

# # Create necessary directories and set permissions
# RUN mkdir -p /app/.huggingface /app/pegasus_model /app/nltk_data \
#     && chmod -R 777 /app/.huggingface /app/pegasus_model /app/nltk_data

# # Copy project files (excluding large model weights)
# COPY . .

# # Install Python dependencies
# RUN pip install --no-cache-dir --upgrade pip \
#     && pip install --no-cache-dir -r requirements.txt \
#     && pip install --no-cache-dir huggingface_hub

# # Download the Pegasus model and tokenizer from Hugging Face
# RUN python3 -c "from huggingface_hub import snapshot_download; snapshot_download(repo_id='tuner007/pegasus_paraphrase', local_dir='/app/pegasus_model')"

# # # Download NLTK 'punkt' tokenizer only
# # RUN python3 -m nltk.downloader punkt && cp -r /root/nltk_data/tokenizers /app/nltk_data/

# # Download NLTK 'punkt' tokenizer only (to correct path)
# RUN python3 -m nltk.downloader -d /app/nltk_data punkt
# ENV NLTK_DATA=/app/nltk_data



# # Expose the port FastAPI runs on
# EXPOSE 7860

# # Start the API using uvicorn
# CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "7860"]


#######################################

# # Use the official Python image with slim variant
# FROM python:3.10-slim

# # Set environment variables
# ENV PYTHONDONTWRITEBYTECODE=1 \
#     PYTHONUNBUFFERED=1 \
#     HF_HOME=/app/.huggingface \
#     NLTK_DATA=/app/nltk_data \
#     PORT=7860 \
#     MAX_TOKENS=3500

# # Set the working directory
# WORKDIR /app

# # Install system dependencies
# RUN apt-get update && \
#     apt-get install -y --no-install-recommends git && \
#     rm -rf /var/lib/apt/lists/*

# # Create necessary directories with appropriate permissions
# RUN mkdir -p /app/.huggingface /app/pegasus_model /app/nltk_data && \
#     chmod -R 777 /app/.huggingface /app/pegasus_model /app/nltk_data

# # Install Python dependencies and NLTK data
# COPY requirements.txt .
# RUN pip install --no-cache-dir --upgrade pip && \
#     pip install --no-cache-dir -r requirements.txt && \
#     python -m nltk.downloader -d /app/nltk_data punkt punkt_tab averaged_perceptron_tagger && \
#     rm -rf /root/.cache/pip

# # Copy application code
# COPY . .

# # Download the Pegasus model
# RUN python -c "from huggingface_hub import snapshot_download; snapshot_download(repo_id='tuner007/pegasus_paraphrase', local_dir='/app/pegasus_model')"

# # Expose the port
# EXPOSE $PORT

# # Start the API
# CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "7860"]


################### work ######################

# Use the official Python image with slim variant
FROM python:3.10-slim

# Set environment variables (fixed syntax)
ENV PYTHONDONTWRITEBYTECODE=1 \
    PYTHONUNBUFFERED=1 \
    HF_HOME=/app/.huggingface \
    NLTK_DATA=/app/nltk_data \
    PORT=7860 \
    OMP_NUM_THREADS=1 \
    TOKENIZERS_PARALLELISM=false

# Set the working directory
WORKDIR /app

# Install system dependencies
RUN apt-get update && \
    apt-get install -y --no-install-recommends git && \
    rm -rf /var/lib/apt/lists/*

# Create necessary directories with appropriate permissions
RUN mkdir -p /app/.huggingface /app/pegasus_model /app/nltk_data && \
    chmod -R 777 /app/.huggingface /app/pegasus_model /app/nltk_data

# Install Python dependencies and NLTK data
COPY requirements.txt .
RUN pip install --no-cache-dir --upgrade pip && \
    pip install --no-cache-dir -r requirements.txt && \
    python -m nltk.downloader -d /app/nltk_data punkt punkt_tab && \
    rm -rf /root/.cache/pip

# Copy application code
COPY . .

# Download the Pegasus model
RUN python -c "from huggingface_hub import snapshot_download; snapshot_download(repo_id='tuner007/pegasus_paraphrase', local_dir='/app/pegasus_model')"

# Expose the port
EXPOSE $PORT

# Start the API
CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "7860"]