kernels/Dockerfile.python

# MAC Kernel — Python 3.11 with 80+ ML/DS/AI libraries pre-installed
# Build: docker build -t mac-kernel-python:latest -f Dockerfile.python .
# Usage: This image is used by the MAC kernel manager for offline notebook execution.

FROM python:3.11-slim

LABEL maintainer="MAC Team" \
      description="Python notebook kernel with 80+ pre-installed libraries for offline use"

# System dependencies for common Python packages
RUN apt-get update && apt-get install -y --no-install-recommends \
    build-essential gcc g++ gfortran \
    libffi-dev libssl-dev libxml2-dev libxslt1-dev \
    libhdf5-dev libopenblas-dev liblapack-dev \
    libjpeg-dev libpng-dev zlib1g-dev \
    graphviz libgraphviz-dev \
    git curl wget \
    && rm -rf /var/lib/apt/lists/*

# Core data science & ML stack
RUN pip install --no-cache-dir \
    numpy==1.26.4 \
    pandas==2.2.2 \
    scipy==1.14.1 \
    sympy==1.13.3 \
    statsmodels==0.14.4

# Visualization
RUN pip install --no-cache-dir \
    matplotlib==3.9.2 \
    seaborn==0.13.2 \
    plotly==5.24.1 \
    bokeh==3.5.2 \
    altair==5.4.1 \
    pillow==10.4.0

# Machine Learning
RUN pip install --no-cache-dir \
    scikit-learn==1.5.2 \
    xgboost==2.1.1 \
    lightgbm==4.5.0 \
    catboost==1.2.7

# Deep Learning (CPU versions — GPU workers get CUDA versions)
RUN pip install --no-cache-dir \
    torch==2.4.1+cpu torchvision==0.19.1+cpu torchaudio==2.4.1+cpu \
    --index-url https://download.pytorch.org/whl/cpu

RUN pip install --no-cache-dir \
    tensorflow-cpu==2.17.0 \
    keras==3.6.0

# NLP / Transformers
RUN pip install --no-cache-dir \
    transformers==4.46.2 \
    tokenizers==0.20.3 \
    datasets==3.1.0 \
    sentencepiece==0.2.0 \
    nltk==3.9.1 \
    spacy==3.8.2 \
    gensim==4.3.3

# Image / Vision
RUN pip install --no-cache-dir \
    opencv-python-headless==4.10.0.84 \
    scikit-image==0.24.0 \
    imageio==2.36.0

# Data processing & file I/O
RUN pip install --no-cache-dir \
    openpyxl==3.1.5 \
    xlsxwriter==3.2.0 \
    pyarrow==18.0.0 \
    h5py==3.12.1 \
    lxml==5.3.0 \
    beautifulsoup4==4.12.3 \
    pyyaml==6.0.2 \
    toml==0.10.2 \
    orjson==3.10.11

# HTTP / API
RUN pip install --no-cache-dir \
    requests==2.32.3 \
    httpx==0.28.1 \
    aiohttp==3.11.7 \
    fastapi==0.115.6

# Databases
RUN pip install --no-cache-dir \
    sqlalchemy==2.0.36 \
    psycopg2-binary==2.9.10 \
    redis==5.2.1

# Utilities
RUN pip install --no-cache-dir \
    tqdm==4.67.0 \
    rich==13.9.4 \
    click==8.1.7 \
    pydantic==2.10.4 \
    python-dotenv==1.0.1 \
    cryptography==43.0.3 \
    pytest==8.3.4

# Jupyter/IPython for notebook compatibility
RUN pip install --no-cache-dir \
    ipython==8.29.0 \
    ipykernel==6.29.5 \
    nbformat==5.10.4

# Additional ML/AI tools
RUN pip install --no-cache-dir \
    onnxruntime==1.20.0 \
    joblib==1.4.2 \
    optuna==4.0.0 \
    mlflow==2.18.0 \
    wandb==0.18.7

# Math/Science
RUN pip install --no-cache-dir \
    networkx==3.4.2 \
    igraph==0.11.8 \
    shapely==2.0.6

# Misc popular packages
RUN pip install --no-cache-dir \
    faker==33.0.0 \
    tabulate==0.9.0 \
    colorama==0.4.6 \
    regex==2024.9.11

# Download NLTK data (offline use)
RUN python -c "import nltk; nltk.download('punkt'); nltk.download('punkt_tab'); nltk.download('stopwords'); nltk.download('wordnet')"

WORKDIR /workspace

# Non-root user for security
RUN useradd -m -s /bin/bash coder
USER coder

CMD ["python3"]