Auto-Documenting Jupyter Notebooks

The Jupyter notebook is a popular open source computational notebook for the python language that combines code and documentation in the form of markdown, and outputs into one file.

Documentation and explanation of thought processes is essential in data science to understand analysis, and reproducing results. However, it was found that a large number of these notebooks found in public Github repositories lacked such explanation. 

This project aims to see if it is viable to transfer the documentation found in existing notebooks to other undocumented notebooks, and if doing so will actually improve the understandability of existing notebooks. It may ultimately lead to a tool that is able to improve the understandability of Jupyter notebooks by automatically documenting them.

Transforming technologies


Computer Science

Tejas Hegde

vote for this project: TT57

Back to project list