Clean up CausalML

[Eyepatch0]

1. Understand the Core Documentation

• Read class_project/project_template/README.md
  • Is there anything unclear, incorrect, or that can be improved?

2. Study Code Examples

• Carefully examine all code in class_project/project_template
  • Make sure you understand every file and module perfectly
  • Note the structure, naming conventions, and patterns used

3. Learn the Claude Skills

• Become familiar with the Claude Skills in helpers_root/.claude/skills

  • Skills are living documentation of how we maintain and create code

• Explore the available skills using these commands:

  > md skill describe coding
  > md skill describe notebooks
  > md skill describe testing
  > md skill describe X_in_60_minutes

• Review key skills relevant to tutorials:

  • coding.format_rules: Python code conventions
  • testing.format_rules: Unit test patterns
  • notebook.format_rules: Jupyter notebook structure
  • markdown.format_rules: Documentation style
  • E.g.,

    > md skill edit coding.format_rules

• Is there anything unclear or that could be improved in the skills?

4. Study Reference Tutorials

• Read 2-3 tutorials
  • Tutorials that are closer to the standards are:
    • Autogen
  • Internalize its structure, style, and approach
  • Note patterns, structure, and what makes them effective
  • Identify what works well and what could be improved

5. Verify Code Quality

• Review the test structure, specifically
  class_project/project_template/test/test_docker_all.py
  • Understand what tests each tutorial should include
  • Ensure your tutorial will have comprehensive test coverage