Machine Learning Docker Environment

This Docker environment is set up for machine learning development with Python 3.11, including support for TensorFlow, PyTorch, and Jupyter Lab.

Features

• Python 3.11 base environment
• Jupyter Lab interface
• Support for Streamlit applications
• Pre-installed ML libraries:
  • TensorFlow
  • PyTorch (CPU version)
  • Hugging Face Transformers
  • scikit-learn
  • and more

Quick Start

Build the Docker Image

docker build -t llm-jupyter-v11 .

Run with Docker Run Command

Below replace C:/path/to/your/notebooks with your actual notebooks directory path:

docker run -d --name llm-jupyter-v11-streamlit \
  -v C:/path/to/your/notebooks/:/home/git \
  -p 8888:8888 -p 8501:8501 \
  llm-jupyter-v11:latest \
  jupyter lab --port=8888 --ip=0.0.0.0 --allow-root

Running Streamlit Applications

To run a Streamlit app within the container:

docker exec -it llm-jupyter-v11-streamlit bash -c "cd /home/git && streamlit run globebotter.py --server.address 0.0.0.0"

Accessing the Applications

• Jupyter Lab: http://localhost:8888
• Streamlit: http://localhost:8501

Volume Mounts

• Your notebooks directory is mounted to /home/git inside the container
• Ensure you replace C:/path/to/your/notebooks with your actual path

Notes

• The environment uses Python 3.11 with a slim base image
• No authentication is required for Jupyter Lab (suitable for local development only)
• All Python dependencies are managed through pip and requirements.txt
• The container exposes ports:
  • 8888 (Jupyter Lab)
  • 8501 (Streamlit)
  • 5678 (Debug port)