Objective

Refactor the Jupyter Notebook from the Titanic Kaggle competition into a well-structured repository with proper software engineering practices.

Tasks

1. Getting started:
   • One team member forks the repository (https://github.com/cambridge-FDS/ps1_ex4_nb_to_repo).
   • The other team members clone the repository from the team member’s GitHub page.
2. Code Modularisation:
   • Rewrite code as functions from the Jupyter Notebook.
   • Create appropriate Python modules to house these functions.
   • Write clear docstrings for all functions.
3. Repository Structure:
   • Organise the project into a clear directory structure.
   • Create an environment.yml file for dependency management.
4. Notebook Refactoring:
   • Maintain the Jupyter Notebook to present your analysis.
   • But replace code cells with functions you import from your own modules to abstract the implementation details away.
   • Implement solutions for cells marked with "TODO": often that means calling a function you will have written before.
5. Version Control and GitHub:
   • Use Git for version control.
   • Create separate branches for different features/modules.
   • Create PRs for your changes and conduct thorough code reviews.

Workflow

1. Fork and clone the repository.
2. Create a new branch for each module or major feature.
3. Implement the modularisation and any TODO items.
4. Create a pull request for review.
5. Address feedback and iterate as necessary.
6. Merge approved changes into the main branch.

Remember, the goal is to transform the initial Jupyter Notebook into a well-structured, maintainable, and collaborative Python project while preserving the analysis workflow.